Epri Wind Systems

2010 TECHNICAL REPORT
Renewable Energy Technology Guide: 2010

.
10581090
10581090
Renewable Energy Technology
Guide: 2010
1019760
Final Report, December 2010
EPRI Project Manager

C. McGowin
Other EPRI Staff Contributing to this Report

T. Coleman
P. Jacobson
C. Libby
D. O’Connor
T. Peterson
A. Tuohy
ELECTRIC POWER RESEARCH INSTITUTE

3420 Hillview Avenue, Palo Alto, California 94304-1338 ▪ PO Box 10412, Palo Alto, California 94303-0813 ▪ USA
800.313.3774 ▪ 650.855.2121 ▪ askepri@epri.com ▪ www.epri.com
10581090
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE
ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR
SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL
PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S
CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER

(INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR
SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.
THE FOLLOWING ORGANIZATION PREPARED THIS REPORT:
Electric Power Research Institute (EPRI)
NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or
e-mail askepri@epri.com.
Electric Power Research Institute, EPRI, and TOGETHER…SHAPING THE FUTURE OF ELECTRICITY
are registered service marks of the Electric Power Research Institute, Inc.
Copyright © 2010 Electric Power Research Institute, Inc. All rights reserved.
10581090
PRODUCT DESCRIPTION
First published in 2000 as the Renewable Energy Technical Assessment Guide—TAG-RE, the
annual Electric Power Research Institute (EPRI) Renewable Energy Technology Guide provides
a consistent basis for evaluating the economic feasibility of renewable generation technologies,
including wind, solar photovoltaic (PV), solar thermal, biomass, geothermal, and emerging ocean
energy conversion technologies.
Results and Findings

Based on rated capacity, the leading worldwide sources of renewable energy at the end of 2009
were wind (160 GW), biomass (67 GW), solar PV (21.5 GW), geothermal (10.7 GW), and solar
thermal (1.1 GW). Counting hydroelectric generation as a renewable resource, renewable energy
contributed 8.2% of U.S. primary energy consumption and 9.9% of electricity generation during
2009, most of which was contributed by biomass and hydro.
Challenges and Objectives

Because so many conflicting and overly optimistic claims are made about the performance and
economic potential of all energy sources—conventional fossil, nuclear, hydro, alternative fuel,
and renewables—it is important to conduct careful, objective assessments of the status,
performance, cost, environmental impacts, and other aspects of the technologies. The primary
objective of this report is to provide timely and unbiased assessments of the status, performance,
and cost of the renewable generation technologies.
Applications, Value, and Use

The renewable generation technology data reported in the EPRI Renewable Energy Technology
Guide are of value to system planners and others who make decisions about whether renewable
technology belongs in the generation portfolio and who develop a long-term strategy and plan for
a sustainable generation portfolio.
EPRI Perspective
An energy producer can use renewable energy to strengthen ties to the community, attract new
customers, and diversify its portfolio. EPRI believes that companies that have not yet
incorporated renewables into their generation mix should consider these technologies to
determine which ones might benefit their generation strategy. The information summarized in
this report encompasses more than three decades of EPRI research on renewable technologies.
The Renewable Energy Technology Guide will continue to be updated on an annual basis to
ensure that EPRI members have access to the most current information on the technical and
economic status of renewable technologies.
iii
10581090
Approach
Information and data were collected and analyzed from both EPRI and outside sources on the
status, performance, cost, installed capacity, and markets for the renewable energy generation
technologies from a variety of sources, including EPRI studies, U.S. government and other
publicly available reports, and additional sources. Then the information in each section of the
2009 Renewable Energy Technology Guide (EPRI report 1021379) was updated.
Keywords
Biomass
Geothermal
Renewable energy
Solar photovoltaic
Solar thermal
Wind energy
10581090
ABSTRACT
The Renewable Energy Technology Guide: 2010 provides a basic reference to inform the
technical and economic assessment of renewable energy generation technologies. Design, cost,
and performance information contained in this report will enable EPRI members to perform
preliminary capital investment evaluations in a consistent and informed manner. Renewable
technologies covered include: wind, biomass, solar photovoltaics, geothermal, and solar
thermal—technologies that are commercially available or on the threshold of commercialization.
Additional sections address the emerging ocean tidal and wave energy technologies, river
instream energy technologies, and the potential to reduce greenhouse gas emissions by using
renewable energy.
v
10581090
10581090
CONTENTS
1 INTRODUCTION ....................................................................................................................1-1
1.1 Background .....................................................................................................................1-1
1.2 EPRI’s Role .....................................................................................................................1-5
1.3 Objective .........................................................................................................................1-5
1.4 Definitions and Units .......................................................................................................1-6
1.5 Scope ..............................................................................................................................1-6
1.6 References ......................................................................................................................1-8
2 ECONOMIC METHODOLOGY AND ASSUMPTIONS ..........................................................2-1

2.1 Regulated Utility Power Projects .....................................................................................2-1
2.1.1 Fixed Charge Rates.................................................................................................2-1
2.1.1.1 Annual Fixed Charge Rates.............................................................................2-1
2.1.1.2 Nominal Levelized Annual Charges and Nominal Levelized Fixed
Charge Rates...............................................................................................................2-3
2.1.1.3 Real Levelized Annual Charges and Real Levelized Fixed Charge Rates ......2-3
2.1.2 Example Annual Fixed Charge Rates for a 200-MW Wind Power Plant .................2-5
2.1.3 Other Capital Expenses—Capital Additions ............................................................2-5
2.1.4 Calculating Costs per Kilowatt-Hour........................................................................2-5
2.1.4.1 Levelized Costs per Kilowatt-Hour...................................................................2-5
2.2 Non-Utility Generator Power Projects .............................................................................2-8
2.2.1 Types of Non-Utility Generators ..............................................................................2-9
2.2.1.1 PURPA Cogenerators and Small Power Producers ......................................2-10
2.2.1.2 Exempt Wholesale Generators ......................................................................2-11
2.2.1.3 Merchant Power Plants..................................................................................2-11
2.2.2 Development of an Economic Pro Forma for a Merchant Plant ............................2-12
2.2.2.1 Differences Between Regulated Utility and Non-Utility Power Projects.........2-12
2.2.2.1.1 Pricing ....................................................................................................2-12
2.2.2.1.2 Income Taxes and Depreciation ............................................................2-12
vii
10581090
2.2.2.1.3 Capital Expenditures ..............................................................................2-13
2.2.2.1.4 Debt Financing .......................................................................................2-13
2.2.2.1.5 Economic Methodology..........................................................................2-13
2.2.3 Conceptualizing the Analysis.................................................................................2-13
2.2.3.1 Total Capital Requirement .............................................................................2-14
2.2.3.1.1 Construction Cost...................................................................................2-14
2.2.3.1.2 Financing Cost .......................................................................................2-16
2.2.3.1.3 Development Cost..................................................................................2-16
2.2.3.1.4 Other Costs ............................................................................................2-16
2.2.3.2 EPRI Capital Cost Definitions ........................................................................2-17
2.2.3.3 Income Statement..........................................................................................2-17
2.2.3.3.1 Revenues ...............................................................................................2-19
2.2.3.3.2 Operating and Maintenance Expenses ..................................................2-21
2.2.3.4 Income Taxes ................................................................................................2-21
2.2.3.5 Cash Flow Statement ....................................................................................2-22
2.2.3.6 Economic Measures ......................................................................................2-24
2.2.3.6.1 Busbar Cost ...........................................................................................2-24
2.2.3.6.2 Net Present Value (NPV) .......................................................................2-25
2.2.3.6.3 Internal Rate of Return (IRR) .................................................................2-26
2.2.3.6.4 Payback Period ......................................................................................2-26
2.2.3.6.5 Debt Coverage Ratios............................................................................2-27
2.2.3.7 Sensitivity Analysis ........................................................................................2-27
2.2.3.8 Project Risks and Financing ..........................................................................2-29
2.2.4 Limitations of Examples.........................................................................................2-31
2.3 Guidelines for Economic Evaluation of Renewable Energy Projects ............................2-31
2.3.1 Design/Cost Estimate ............................................................................................2-31
2.3.2 Accuracy Ranges ..................................................................................................2-32
2.3.3 Definitions of Economic Terms..............................................................................2-32
2.4 References ....................................................................................................................2-35
3 WIND POWER........................................................................................................................3-1
3.1 Introduction .....................................................................................................................3-1
3.2 Installed Wind Capacity...................................................................................................3-2
3.3 Wind Energy Principles .................................................................................................3-10
3.4 Developments in Wind Turbine Technology..................................................................3-12
10581090
3.4.1 Generators & Power Electronics ...........................................................................3-12
3.4.1.1 Squirrel Cage Induction Generator ................................................................3-13
3.4.1.2 Variable-Speed Turbine .................................................................................3-13
3.4.1.3 Power Electronics ..........................................................................................3-14
3.4.1.4 Squirrel Cage Induction Generator with Full Power Conversion....................3-14
3.4.1.5 Doubly-Fed Induction Generator....................................................................3-14
3.4.1.6 Synchronous Generator.................................................................................3-14
3.4.1.7 Direct-Drive Low-Speed Wound Rotor Generators........................................3-15
3.4.1.8 Direct-Drive Low-Speed Permanent Magnet Generators ..............................3-15
3.4.1.9 Superconducting Low-Speed Generators......................................................3-16
3.4.2 Blades and Rotor...................................................................................................3-17
3.4.2.1 Passive Aerodynamic Control........................................................................3-20
3.4.2.2 Active Aerodynamic Control...........................................................................3-20
3.4.2.3 “Stealth” Rotor Blade .....................................................................................3-21
3.4.2.4 “Smart” Blades ...............................................................................................3-21
3.4.2.5 Flatback Airfoils..............................................................................................3-21
3.4.2.6 Blade Manufacturing Processes ....................................................................3-22
3.4.2.7 Blade Transport and Shipping .......................................................................3-22
3.4.2.8 Blade Test Facilities.......................................................................................3-22
3.4.3 Yaw Systems.........................................................................................................3-22
3.4.4 Sensors .................................................................................................................3-23
3.4.4.1 Optical Strain Gages......................................................................................3-24
3.4.4.2 Wind Measurements—LIDAR........................................................................3-24
3.4.4.3 Condition Monitoring ......................................................................................3-25
3.4.5 Controls .................................................................................................................3-25
3.4.6 SCADA Data Collection & Transmittal...................................................................3-27
3.4.7 Drive Train .............................................................................................................3-27
3.4.7.1 Drive Trains with Gearboxes..........................................................................3-27
3.4.7.2 Hydrodynamic Fluid Coupling ........................................................................3-29
3.4.7.3 Direct Drive Train without Gearbox................................................................3-30
3.4.8 Foundation.............................................................................................................3-30
3.4.9 Tower.....................................................................................................................3-31
3.4.9.1 Conventional Steel Towers ............................................................................3-31
3.4.9.2 Concrete and Hybrid Towers .........................................................................3-31
ix
10581090
3.4.9.3 Integrated Towers and Foundations ..............................................................3-32
3.4.9.4 Tall Towers ....................................................................................................3-33
3.4.10 Offshore Foundations ..........................................................................................3-33
3.4.10.1 Bottom-Fixed Offshore Foundation Designs................................................3-33
3.4.10.2 Floating Foundation Concepts .....................................................................3-35
3.5 Trends in the Turbine Supply Market ............................................................................3-37
3.6 Trends in Wind Turbine and Plant Sizes .......................................................................3-39
3.7 Onshore Capital and O&M Cost Trends........................................................................3-41
3.8 Developments in Offshore Wind Technology ................................................................3-44
3.8.1 Trends in the Offshore Turbine Supply Market......................................................3-45
3.8.2 Offshore Transmission Technology Status............................................................3-47
3.8.3 Offshore Capital and O&M Cost Trends................................................................3-47
3.9 Technology Performance and Cost Tables ...................................................................3-48
3.9.1 Site Assumptions...................................................................................................3-48
3.9.2 Plant Performance.................................................................................................3-50
3.9.3 Total Capital Requirement.....................................................................................3-51
3.9.4 Operation and Maintenance Cost..........................................................................3-51
3.9.5 Levelized Cost of Electricity...................................................................................3-56
3.9.6 Performance and Cost Summary ..........................................................................3-57
3.10 Grid Integration............................................................................................................3-62
3.10.1 Production Variability...........................................................................................3-62
3.10.2 Ancillary Service Costs........................................................................................3-63
3.10.3 Wind Energy Forecasting and Scheduling ..........................................................3-64
3.10.4 Regulatory Update...............................................................................................3-65
3.10.5 Power Quality ......................................................................................................3-65
3.11 Project Development Process and Market ..................................................................3-65
3.11.1 Buy or Build Considerations ................................................................................3-65
3.11.2 Project Development Process .............................................................................3-66
3.11.2.1 Pre-Development Activities..........................................................................3-66
3.11.2.2 Engineering Procurement and Construction Activities (EPC) ......................3-69
3.11.2.3 Operation and Maintenance Activities .........................................................3-70
3.12 Environmental Issues ..................................................................................................3-71
3.12.1 Avian and Bat Issues...........................................................................................3-72
3.12.2 Noise ...................................................................................................................3-73
10581090
3.12.3 Visual Impact .......................................................................................................3-74
3.13 Equipment Markets and Key Participants ...................................................................3-75
3.13.1 World Markets .....................................................................................................3-76
3.13.1.1 Trend to Larger Turbines .............................................................................3-77
3.13.1.2 Offshore Market in Europe...........................................................................3-80
3.13.1.3 Market Forecast ...........................................................................................3-81
3.13.1.3.1 2009 Through 2013..............................................................................3-81
3.13.1.3.2 2013 Through 2017..............................................................................3-82
3.13.2 Key Players in the Wind Power Industry .............................................................3-83
3.14 References ..................................................................................................................3-84
3.14.1 DOE-EPRI Wind Turbine Verification Program Reports......................................3-86
3.14.2 Other Reports ......................................................................................................3-88
4 BIOMASS ENERGY ...............................................................................................................4-1

4.1 Introduction .....................................................................................................................4-2
4.1.1 Basic Issues Associated with Biomass Fuel Utilization ...........................................4-4
4.1.2.1 Commercially Available Technologies .............................................................4-5
4.1.2.2 Goal Technologies ...........................................................................................4-6
4.1.2.3 Prospectus .......................................................................................................4-7
4.1.2 Characteristics of Representative Biomass Fuels ...................................................4-7
4.2 Biomass Fuel Resources ................................................................................................4-7
4.2.1 Biofuels vs. Biomass Electricity .............................................................................4-10
4.2.1.1 Refuse Derived Fuels (RDF)..........................................................................4-11
4.2.2 Basic Properties of Solid Biomass Fuels...............................................................4-13
4.2.2.1 Proximate and Ultimate Analyses of Biomass Fuels .....................................4-13
4.2.2.2 Structural Relationships and Heteroatoms ....................................................4-15
4.2.2.3 Inorganic Constituents in Biomass.................................................................4-16
4.2.3 Performance Characteristics of Biomass Fuels.....................................................4-20
4.2.3.1 Biomass Fuel Volatility...................................................................................4-20
4.2.3.2 Fuel Nitrogen Characteristics of Biomass Fuels ............................................4-26
4.2.3.3 Ash Reactivity, Slagging, and Fouling ...........................................................4-34
4.2.4 Wastewater Treatment Gas and Landfill Gas........................................................4-37
4.2.4.1 Gas Compositions .........................................................................................4-37
4.2.4.2 Utilization .......................................................................................................4-38
4.3 Characteristics of Biomass Cofiring Technologies ........................................................4-38
xi
10581090
4.3.1 Overview of Solid Biomass Fuel Cofiring Systems................................................4-38
4.3.2 Cofiring in Cyclone Boilers ....................................................................................4-39
4.3.2.1 Cofiring System Design for Cyclone Boilers ..................................................4-41
4.3.2.2 Capital Costs of Cofiring Systems for Cyclone Boilers ..................................4-45
4.3.2.3 Cofiring Impacts in Cyclone Boilers ...............................................................4-47
4.3.3 Cofiring in Pulverized Coal (PC) Boilers ................................................................4-49
4.3.3.1 Cofiring System Designs for Pulverized Coal Boilers ....................................4-49
4.3.3.2 Capital Costs of Separate Injection PC Cofiring Systems .............................4-51
4.3.3.3 Impacts of Cofiring Biomass in PC Boilers ....................................................4-52
4.3.4 Cofiring Gaseous Biomass Fuels ..........................................................................4-54
4.3.4.1 Biomass Pre-Treatments for Cofiring with Coal.............................................4-55
4.3.4.2 The Global Biomass Pellet Market.................................................................4-58
4.3.5 Conclusions Regarding Biomass Cofiring .............................................................4-58
4.4 Stand-Alone Biomass-Fired Systems............................................................................4-59
4.4.1 Wood-Waste-Fired Boilers ....................................................................................4-59
4.4.1.1 Typical Capacities of Wood-Waste-Fired Plants............................................4-60
4.4.1.2 Wood-Waste-Fired System Design................................................................4-60
4.4.1.2.1 Fuel Handling .........................................................................................4-60
4.4.1.2.2 Boiler and Turbine-Generator Technology.............................................4-61
4.4.1.2.3 Pollution Control Systems ......................................................................4-65
4.4.1.3 Comparisons to Other Biomass-Fired Rankine-Cycle Designs .....................4-66
4.4.2 Repowering Existing Coal Units for Biomass Firing ..............................................4-67
4.4.3 Ownership Models and Institutional Issues ...........................................................4-68
4.4.4 Alternative Systems for Gaseous Biomass Fuels..................................................4-68
4.4.5 Near-Commercial Technologies ............................................................................4-69
4.4.5.1 Wood Gasification for Power and District Heating in Güssing, Austria..........4-69
4.4.5.2 Wood Gasification at Harboøre......................................................................4-71
4.4.5.3 Gasification of Wood Pellets at Skive ............................................................4-72
4.4.5.4 Cofiring Syngas at Ruien ...............................................................................4-73
4.4.5.5 Wood Gasification at Kokemäki .....................................................................4-73
4.4.5.6 Carbon Negative Effects Associated with Bio-Char from Biomass................4-74
4.4.6 Non-Commercial Technologies .............................................................................4-75
4.5 Cost and Performance Summary ..................................................................................4-76
4.5.1 100% Biomass Repowering of a Pulverized Coal Boiler .......................................4-76
10581090
4.5.1.1 Plant Design Assumptions .............................................................................4-77
4.5.1.2 Plant Performance Estimate ..........................................................................4-78
4.5.1.3 Total Performance and Cost Estimates .........................................................4-78
4.5.1.4 Levelized Cost of Electricity Estimates ..........................................................4-81
4.5.2 Biomass Cofiring with Coal in a Pulverized Coal Boiler ........................................4-82
4.5.2 Biomass Cofiring with Coal ...............................................................................4-82
4.5.2.1 Plant Design and Fuel Assumptions ..............................................................4-82
4.5.2.2 Plant Performance Estimates ........................................................................4-83
4.5.2.3 Total Capital Requirement and Operation and Maintenance Cost
Estimates ...................................................................................................................4-85
4.5.3 Biomass-Fired Bubbling Fluidized Bed Combustion Boiler Power Plants .............4-88
Estimates ...................................................................................................................4-89
4.5.4 Biomass-Fired Stoker Boiler Power Plants............................................................4-93
Estimates ...................................................................................................................4-94
4.5.3.4 Levelized Cost of Electricity Estimate ............................................................4-96
4.5.5 Landfill Gas Recovery and Power Generation ......................................................4-97
4.5.6 Biomass Technologies Grubb Curve.....................................................................4-99
4.6 Biomass to Electricity RD&D Initiatives for the Future ..................................................4-99
4.7 U.S. Technical Tax Code Modifications to promote further use of Biomass ...............4-100
4.8 Conclusions.................................................................................................................4-101
4.9 References ..................................................................................................................4-102
5 SOLAR PHOTOVOLTAICS ...................................................................................................5-1

5.1 Introduction .....................................................................................................................5-2
5.2 Cost and Economic Issues..............................................................................................5-4
5.2.1 Japan.......................................................................................................................5-7
5.2.2 Germany..................................................................................................................5-8
5.2.3 Example PV Project Costs ......................................................................................5-9
xiii
10581090
5.2.4 Technology Performance and Cost Tables .............................................................5-9
5.2.4.1 Distributed PV Performance and Cost Estimates ............................................5-9
5.2.4.2 Utility-Scale PV Plant Performance and Cost Estimates ...............................5-10
5.2.4.2.1 Design and Site Assumptions ................................................................5-10
5.2.4.2.2 Plant Performance Estimates.................................................................5-12
5.2.4.2.3 Total Capital Requirement .....................................................................5-13
5.2.4.2.4 Operation and Maintenance Cost ..........................................................5-13
5.2.4.2.5 Levelized Cost of Electricity ...................................................................5-13
5.2.4.2.6 Performance and Cost Summary...........................................................5-17
5.2.5 Performance and Cost History and Projections.....................................................5-20
5.2.6 PV Performance and Cost Estimate Model ...........................................................5-21
5.2.6.1 Resource Estimates.......................................................................................5-21
5.2.6.2 Distributed PV Cost and Performance Estimates ..........................................5-21
5.2.6.3 Hypothetical Case Studies.............................................................................5-23
5.2.6.4 Central Station PV Cost and Performance Methodology...............................5-27
5.2.6.4.1 Example Performance Calculations .......................................................5-31
5.2.6.4.2 Example Capital and O&M Cost Calculations........................................5-31
5.3 Environmental Issues ....................................................................................................5-32
5.4 Design and Deployment Issues.....................................................................................5-34
5.4.1 Utility-Scale Issues ................................................................................................5-35
5.4.2 Operating and Maintenance Labor Requirements.................................................5-35
5.4.3 Interconnection ......................................................................................................5-37
5.5 Equipment Markets and Key Participants .....................................................................5-38
5.5.1 Equipment Markets................................................................................................5-39
5.5.2 Key Participants.....................................................................................................5-41
5.5.3 Internet Resources ................................................................................................5-42
5.5.4 Applicable Codes and Standards ..........................................................................5-44
5.6 References ....................................................................................................................5-45
6 GEOTHERMAL ENERGY ......................................................................................................6-1

6.1 Introduction .....................................................................................................................6-2
6.2 Resources and Current Installed Capacity......................................................................6-3
6.2.1 Hydrothermal Geothermal Resource Assessment ..................................................6-3
6.2.1.1 Resource Temperature ....................................................................................6-3
6.2.1.2 Depth ...............................................................................................................6-6
10581090
6.2.1.3 Resource Permeability.....................................................................................6-6
6.2.2 Geo-Pressured Geothermal Resource Assessment ...............................................6-7
6.2.3 Hot Dry Rock Resource Assessment ......................................................................6-7
6.2.4 Installed Geothermal Capacity ..............................................................................6-10
6.3 Technology Description.................................................................................................6-12
6.3.1 Hydrothermal .........................................................................................................6-13
6.3.1.1 Major Common System Components and Features......................................6-14
6.3.1.2 Direct Steam ..................................................................................................6-15
6.3.1.3 Flash Steam...................................................................................................6-16
6.3.1.4 Binary Power Plants.......................................................................................6-18
6.3.1.5 Low Enthalpy/Reverse Air Conditioning Cycles.............................................6-20
6.3.2 Geo-pressured.......................................................................................................6-20
6.3.3 Hot Dry Rock (Enhanced Geothermal Systems) ...................................................6-20
6.3.3.1 Supercritical Cycles .......................................................................................6-22
6.3.4 Hot Sedimentary Aquifer (HSA).............................................................................6-22
6.3.5 Down-Hole Closed-Loop Systems.........................................................................6-23
6.3.6 Geothermal Hybrid Plants .....................................................................................6-24
6.3.6.1 Geothermal-Solar Hybrid Cycles ...................................................................6-24
6.3.6.2 Geothermal + Gas Peaker Cycles .................................................................6-25
6.3.6.3 Geothermal-Fossil Hybrid Cycles ..................................................................6-26
6.3.6.4 Geothermal-Biomass Hybrid Cycles for Moderate-Low Enthalpy
Sources......................................................................................................................6-26
6.3.7 Non-Electric (“Direct Use”) Geothermal.................................................................6-27
6.4 Technology Status.........................................................................................................6-27
6.4.1 Recent U.S. Developments ...................................................................................6-28
6.4.1.1 Alaska ............................................................................................................6-29
6.4.1.2 California........................................................................................................6-29
6.4.1.3 Hawaii ............................................................................................................6-29
6.4.1.4 Idaho ..............................................................................................................6-30
6.4.1.5 Nevada ..........................................................................................................6-30
6.4.1.6 New Mexico ...................................................................................................6-30
6.4.1.7 Utah ...............................................................................................................6-30
6.4.1.8 Wyoming ........................................................................................................6-30
6.4.1.9 Other States...................................................................................................6-30
6.4.2 Hot Dry Rock Technology Status ..........................................................................6-31
xv
10581090
6.4.3 Geothermal Technologies for Electricity Production Deployment Curve
(Grubb Curve).................................................................................................................6-32
6.4.4 Geophysical Methods in Geothermal Exploration .................................................6-33
6.4.5 Geothermal Drilling Technology ............................................................................6-35
6.5 Cost and Economic Issues............................................................................................6-36
6.5.1 Example Geothermal Project Costs ......................................................................6-38
6.5.2 Factors that Influence Geothermal Power Plant Costs..........................................6-39
6.5.3 Engineering and Economic Evaluation..................................................................6-39
6.6.1 Land Use ...............................................................................................................6-42
6.6.2 Subsidence............................................................................................................6-43
6.6.3 Emissions ..............................................................................................................6-44
6.6.4 Thermal Discharge ................................................................................................6-46
6.6.5 Induced Seismicity.................................................................................................6-46
6.7 Design, Deployment, and O&M Issues .........................................................................6-47
6.7.1 Industry Trends......................................................................................................6-47
6.7.2 Research Needs and Recommendations..............................................................6-48
6.8 Key Research Participants ............................................................................................6-49
6.9 References ....................................................................................................................6-52
7 SOLAR THERMAL.................................................................................................................7-1
7.1 Introduction .....................................................................................................................7-3
7.1.1 Solar Thermal Technologies....................................................................................7-4
7.1.2 Hybridization............................................................................................................7-5
7.2 Resources .......................................................................................................................7-7
7.3 Technology Description...................................................................................................7-9
7.3.1 Parabolic Trough .....................................................................................................7-9
7.3.2 Central Receiver....................................................................................................7-11
7.3.3 Dish/Engine ...........................................................................................................7-15
7.3.3.1 Concentrators ................................................................................................7-16
7.3.3.2 Engines ..........................................................................................................7-18
7.3.4 Compact Linear Fresnel Reflector.........................................................................7-18
7.3.5 Solar Chimney .......................................................................................................7-19
7.4 Technology Status.........................................................................................................7-21
7.4.1 Parabolic Trough ...................................................................................................7-25
10581090
7.4.3 Dish/Engine ...........................................................................................................7-31
7.5 Cost and Economic Issues............................................................................................7-33
7.5.1 Engineering and Economic Evaluation..................................................................7-34
7.5.4 Dish/Engine ...........................................................................................................7-39
7.6.1 Land Use ...............................................................................................................7-40
7.6.2 Water Use..............................................................................................................7-41
7.6.3 Molten Salt for Central Receiver............................................................................7-41
7.6.4 Oil for Heat Transfer in Trough Applications .........................................................7-41
7.6.5 Potential for Greenhouse Gas Reduction..............................................................7-41
7.7 Design, Deployment and O&M Issues ..........................................................................7-41
7.7.3 Dish/Engine ...........................................................................................................7-47
7.8 Equipment Markets and Key Participants .....................................................................7-47
7.8.1 Internet Resources ................................................................................................7-52
7.9 References ....................................................................................................................7-53
8 OCEAN TIDAL ENERGY .......................................................................................................8-1

8.1 Introduction .....................................................................................................................8-3
8.2 U.S. Tidal In-Stream Energy Highlights: Mid-2009 to Late-2010 ....................................8-5
8.2.1 Federal and State Highlights ...................................................................................8-5
8.2.1.1 Federal-Level Activities....................................................................................8-5
8.2.1.2 State of Alaska Projects...................................................................................8-6
8.2.1.3 State of Washington Projects...........................................................................8-7
8.2.1.4 State of California Projects ..............................................................................8-7
8.2.1.5 State of New York Projects ..............................................................................8-8
8.2.1.6 State of Massachusetts Projects......................................................................8-8
8.2.1.7 State of Maine Projects....................................................................................8-9
8.2.1.8 State of New Hampshire Projects ..................................................................8-10
xvii
10581090
8.2.1.9 State of New Jersey Projects.........................................................................8-11
8.2.2 U.S. Developer Highlights .....................................................................................8-11
8.2.2.1 Verdant Power ...............................................................................................8-11
8.2.2.2 Vortex Hydro Energy .....................................................................................8-12
8.2.2.3 Ocean Renewable Power Corporation ..........................................................8-12
8.3 Worldwide Tidal In-Stream Energy Highlights: Mid-2009 to Late-2010.........................8-12
8.3.1 Canada ..................................................................................................................8-12
8.3.2 United Kingdom (UK).............................................................................................8-13
8.3.2.1 Marine Current Turbines (MCT).....................................................................8-15
8.3.2.2 Pulse Tidal .....................................................................................................8-15
8.3.2.3 Swanturbines .................................................................................................8-16
8.3.2.4 Tidal Energy Ltd.............................................................................................8-17
8.3.2.5 Atlantis Resources .........................................................................................8-18
8.3.2.6 Hammerfest Strøm.........................................................................................8-18
8.3.3 Ireland....................................................................................................................8-18
8.3.3.1 OpenHydro ....................................................................................................8-18
8.3.4 Continental Europe................................................................................................8-20
8.3.5 Australia.................................................................................................................8-20
8.4 Tidal Power and Energy Resources..............................................................................8-20
8.4.1 Data Sources.........................................................................................................8-20
8.4.2 Tidal Power Flux and Average Annual Energy......................................................8-22
8.4.3 Extractable Tidal Power and Energy .....................................................................8-23
8.4.4 Tidal Power Forecasting........................................................................................8-24
8.5 Tidal In-Stream Energy Conversion (TISEC) Development ..........................................8-24
8.5.1 Harnessing Tidal Energy .......................................................................................8-25
8.5.2 TISEC System Developers....................................................................................8-27
8.5.3 Survival in Storms and Hostile Marine Environments............................................8-29
8.5.4 Effect of Tidal Power Plants on the Environment ..................................................8-29
8.5.5 Permits for Tidal Power Plants ..............................................................................8-29
8.5.6 Overview of Regulatory Status for Tidal Power Plants ..........................................8-30
8.5.6.1 FERC Pilot Project License............................................................................8-30
8.6 Design, Performance, Cost, and Economic Feasibility .................................................8-31
8.6.1 TISEC Sites ...........................................................................................................8-31
8.6.2 TISEC Devices Studied .........................................................................................8-31
10581090
8.6.3 TISEC Economic Feasibility ..................................................................................8-33
8.7 Environmental Impact Issues ........................................................................................8-34
8.8 Installed Capacity and Estimated Growth .....................................................................8-36
8.9 R&D Needs ...................................................................................................................8-37
8.10 Conclusion...................................................................................................................8-38
8.11 Internet Resources ......................................................................................................8-38
8.12 References ..................................................................................................................8-39
9 OCEAN WAVE ENERGY .......................................................................................................9-1

9.1 Introduction .....................................................................................................................9-2
9.2 U.S. Wave Energy Highlights: Mid-2009 to Late-2010....................................................9-5
9.2.1 Federal- and State-Level Highlights ........................................................................9-5
9.2.1.1 Federal-Level Developments ...........................................................................9-5
9.2.1.2 State of Hawaii Projects...................................................................................9-8
9.2.1.3 State of Oregon Projects..................................................................................9-9
9.2.1.4 State of California Projects ............................................................................9-10
9.2.1.5 State of Washington Projects.........................................................................9-12
9.2.1.6 State of Maine Projects..................................................................................9-12
9.2.2 U.S. Developer and Project Deployment Highlights ..............................................9-12
9.3 Worldwide Wave Energy Highlights: Mid-2009 to Late-2010 ........................................9-12
9.3.1 United Kingdom .....................................................................................................9-13
9.3.1.1 UK Wave Energy Developers ........................................................................9-13
9.3.1.1.1 Aquamarine Power.................................................................................9-13
9.3.1.1.2 Checkmate Sea Energy .........................................................................9-13
9.3.1.1.3 Orecon ...................................................................................................9-14
9.3.1.1.4 Pelamis WavePower ..............................................................................9-15
9.3.1.2 UK Wave Energy Test Centers......................................................................9-15
9.3.1.2.1 NaREC ...................................................................................................9-15
9.3.1.2.2 The European Marine Energy Centre (EMEC) ......................................9-15
9.3.1.2.3 The UK Wave Hub .................................................................................9-16
9.3.2 Continental Europe................................................................................................9-17
9.3.2.1 WaveRoller ....................................................................................................9-17
9.3.2.2 Ongoing Testing and Development in Denmark, Norway and Sweden.........9-17
9.3.3 Australia, New Zealand and Tasmania..................................................................9-19
9.4. Wave Power and Energy Resources ...........................................................................9-20
xix
10581090
9.4.1 Measurement Data Sources..................................................................................9-22
9.4.2 Wind-Wave Model Data Sources ..........................................................................9-23
9.4.3 Available Offshore Wave Resource Calculation Methodology ..............................9-25
9.4.4 Practically Recoverable Wave Resource Calculation Methodology ......................9-26
9.4.5 Wave Power Forecasting ......................................................................................9-26
9.5 Wave Energy Conversion (WEC) Technology Description ...........................................9-27
9.5.1 Harnessing Wave Energy......................................................................................9-27
9.5.2 WEC System Developers ......................................................................................9-29
9.5.3 Survival in Storms and Hostile Marine Environments............................................9-31
9.5.4 Effect of Wave Power Plants on the Environment.................................................9-32
9.5.4.1. Report to Congress: Potential Environmental Effects of Marine and
Hydrokinetic Energy Technologies [11] .....................................................................9-32
9.5.5 Permits for Offshore Wave Power Plants ..............................................................9-34
9.5.6 Overview of Regulatory Status for Offshore Wave Power Plants..........................9-34
9.5.7 WEC Power Plant Footprints.................................................................................9-35
9.6 Design, Performance, Cost, and Economic Feasibility Issues ......................................9-37
9.6.1 WEC Sites .............................................................................................................9-37
9.6.2 WEC Design, Performance and Cost ....................................................................9-37
9.6.4 WEC Economic Feasibility ....................................................................................9-39
9.7 Installed Capacity and Estimated Growth .....................................................................9-40
9.8 R&D Needs ...................................................................................................................9-41
9.9 Conclusions...................................................................................................................9-42
9.10 Resources and References.........................................................................................9-43
9.10.1 Internet Resources ..............................................................................................9-43
9.10.2 References ..........................................................................................................9-43
10 RIVER IN-STREAM ENERGY ............................................................................................10-1

10.1 Introduction..................................................................................................................10-2
10.1.1 Main Components of River Hydrokinetic Resource.............................................10-3
10.1.2 The Conversion of River In-Stream Hydrokinetic Energy to Electricity................10-4
10.2 U.S. In-Stream River Energy Highlights: 2009–2010 ..................................................10-6
10.2.1 Federal and State Project Highlights ...................................................................10-6
10.2.1.1 Federal-Level Projects .................................................................................10-6
10.2.1.2 State-Level Projects.....................................................................................10-6
10.2.1.3 Other Projects ..............................................................................................10-8
10581090
10.3 Canadian River In-Stream Energy Highlights: 2009–2010........................................10-11
10.4 River In-Stream Power and Energy Resources ........................................................10-13
10.5 River In-Stream Energy Conversion (RISEC) Development .....................................10-16
10.5.1 Harnessing In-Stream River Energy..................................................................10-16
10.5.2 RISEC Technology Developers.........................................................................10-18
10.5.2.1 Aero Hydro Research and Technology Associates ..................................10-19
10.5.2.2 Free Flow Power.......................................................................................10-20
10.5.2.3 Free Flow 69 .............................................................................................10-21
10.5.2.4 Hydro Green .............................................................................................10-22
10.5.2.5 Lucid Energy Technologies.......................................................................10-23
10.5.2.6 New Energy Corp......................................................................................10-24
10.5.2.7 Ocean Renewable Power Corp. ...............................................................10-25
10.5.2.8 UEK Corp..................................................................................................10-26
10.5.2.9 Verdant Power ..........................................................................................10-26
10.5.2.10 Vortex Hydro ...........................................................................................10-28
10.5.3 Survival in Storms and Hostile River Environments ..........................................10-28
10.5.4 Environmental Impacts of In-Stream River Power Plants..................................10-29
10.5.5 Permits for In-Stream River Power Plants .........................................................10-29
10.5.6 Overview of Regulatory Status for River In-Stream Energy Conversion
Projects.........................................................................................................................10-30
10.6 Design, Performance, Cost, and Economic Feasibility .............................................10-30
10.6.1 RISEC Sites.......................................................................................................10-30
10.6.2 RISEC Devices Studied.....................................................................................10-30
10.6.3 RISEC Design, Performance, and Cost ............................................................10-32
10.6.4 RISEC Economic Feasibility..............................................................................10-34
10.7 Environmental Issues ................................................................................................10-35
10.8 Installed Capacity and Estimated Growth .................................................................10-35
10.9 RISEC R&D Needs ...................................................................................................10-36
10.10 Conclusions.............................................................................................................10-38
10.11 Resources and References.....................................................................................10-38
10.11.1 Internet Resources ..........................................................................................10-38
10.11.2 References ......................................................................................................10-38
11 RENEWABLE ENERGY GRID INTEGRATION TECHNOLOGIES ...................................11-1

11.1 Renewable Generation Integration Challenges...........................................................11-2
xxi
10581090
11.2 Basics of Grid Operating Systems ..............................................................................11-4
11.2.1 Functions of System Generation .........................................................................11-4
11.2.2 Expectations Based on Traditional Generator Performance ...............................11-5
11.2.3 Typical Power System-Operating Strategy..........................................................11-7
11.2.3.1 Energy Balancing.........................................................................................11-7
11.2.3.2 Unit Commitment .........................................................................................11-9
11.2.3.3 Economic Dispatch ......................................................................................11-9
11.2.3.4 Role of Ancillary Services ............................................................................11-9
11.2.3.5 Regulation and Regulating Reserves ........................................................11-10
11.2.3.6 Reserve Generation...................................................................................11-10
11.2.3.7 Voltage Control ..........................................................................................11-11
11.2.3.8 Black Start..................................................................................................11-12
11.3 Integration Issues of Variable Generation .................................................................11-12
11.3.1 Variability ...........................................................................................................11-12
11.3.2 Uncertainty ........................................................................................................11-16
11.3.3 Limited Reactive Power Control ........................................................................11-16
11.3.4 Distributed Generation.......................................................................................11-17
11.3.5 Remote Location of Some Renewable Resources............................................11-17
11.3.6 Integration of Wind Power .................................................................................11-18
11.3.7 Integration of Solar Power .................................................................................11-19
11.4 Variable Generation Impacts and Challenges...........................................................11-22
11.4.1 Transmission Connected Generation Impacts ..................................................11-22
11.4.2 Distribution Connected Generation Impacts......................................................11-27
11.5 Integration Technologies and strategies for Transmission-Connected Renewable
Generation ........................................................................................................................11-31
11.5.1 Variable Generator Interface and Output Control ..............................................11-31
11.5.1.1 Constant-Speed and Semi-Variable Speed Induction Generator with
Direct Grid Connection.............................................................................................11-32
11.5.1.2 Variable-Speed Induction or Asynchronous Generator with Power
Electronic Interface to Grid ......................................................................................11-33
11.5.1.2.1 Doubly-Fed Induction Generator........................................................11-33
11.5.1.2.2 Full-Converter Synchronous or Induction Generator .........................11-35
11.5.1.2.3 Variable-Speed HVDC-Link to Grid....................................................11-36
11.5.2 Ride-Through During Faults and Other Grid Disturbances ...............................11-38
11.5.2.1 Adaptive Protective Relaying .....................................................................11-38
10581090
11.5.2.2 Low-Voltage Ride-Through Circuits ...........................................................11-38
11.5.3 Reactive Compensation Equipment ..................................................................11-40
11.5.3.1 Distributed Constant Power Factor ............................................................11-41
11.5.3.2 Distributed Variable Power Factor .............................................................11-42
11.5.3.3 Substation Switched Capacitor Banks .......................................................11-43
11.5.3.4 Power Electronic Custom Power Devices..................................................11-44
11.5.3.4.1 Static Shunt Compensator (DSTATCOM)..........................................11-45
11.5.3.4.2 Distribution Static VAR Compensator (DSVC) ...................................11-46
11.5.4 Wind Energy Forecasting Tools ........................................................................11-48
11.5.5 Increased Diversity through Increased Transmission and BA Cooperation ......11-52
11.5.6 Flexibility from Conventional (Non-VG) Plant ....................................................11-52
11.5.7 Energy-Storage Technology..............................................................................11-53
11.5.7.1 Energy Storage Technology Descriptions..................................................11-53
11.5.7.2 Energy Storage Applications and Benefits.................................................11-57
11.5.7.3 Energy Storage Attributes and Costs.........................................................11-61
11.5.8 Grid Planning and Operating Tools ...................................................................11-68
11.6 Integration Technologies and Strategies for Distribution-Connected Renewable
Generation ........................................................................................................................11-68
11.7 References ................................................................................................................11-75
11.7.1 Endnotes ...........................................................................................................11-75
11.7.2 General..............................................................................................................11-77
11.7.3 Renewable Generation Grid Impacts and Integration .......................................11-77
11.7.4 Battery Energy Storage .....................................................................................11-78
11.7.5 Compressed Air Energy Storage.......................................................................11-78
11.7.6 Pumped Hydro Energy Storage.........................................................................11-79
11.7.7 Flywheel Energy Storage ..................................................................................11-79
11.7.8 SMES ................................................................................................................11-80
12 GREENHOUSE GAS EMISSIONS CONTROL ..................................................................12-1

12.1 Greenhouse Gases .....................................................................................................12-1
12.2 Role of Renewable Energy Technology ......................................................................12-1
12.3 Fossil Carbon Intensity of Fuels ..................................................................................12-2
12.4 CO2 Emissions Offsets ................................................................................................12-3
12.4.1 Factors Affecting CO2 Emissions Offset ..............................................................12-3
12.4.1.1 Wind.............................................................................................................12-4
xxiii
10581090
12.4.1.2 Solar PV and Thermal..................................................................................12-4
12.4.1.3 Biomass, Geothermal, and Hydro................................................................12-4
12.4.1.4 Planting Trees and Other Fast-Growing Crops............................................12-5
12.4.2 Levelized CO2 Emissions Offsets ........................................................................12-5
12.5 Factors Affecting Generation Cost of Renewable Technologies.................................12-7
12.5.1 Wind ....................................................................................................................12-7
12.5.2 Biomass.............................................................................................................12-10
12.5.3 Solar Photovoltaics............................................................................................12-11
12.5.4 Geothermal........................................................................................................12-14
12.5.5 Solar Thermal ....................................................................................................12-15
12.6 References ................................................................................................................12-17
10581090
LIST OF FIGURES
Figure 1-1 Contributions of Renewable Energy and other Sources to Total U.S. Energy
Supply during 2009 (U.S. EIA, 2010 [1]) ............................................................................1-3
Figure 1-2 Contributions of Renewable Energy and other Sources to Net U.S. Electricity
Generation during 2009 (U.S. EIA, 2009 [2],) ....................................................................1-3
Figure 2-1 Day-Ahead Prices for the California Power Exchange ...........................................2-19
Figure 2-2 Example Price Duration Curve ...............................................................................2-20
Figure 2-3 Components of Busbar Cost for Example 200-MW Wind Generation Plant...........2-24
Figure 3-1 Historical (1991-2009) and Projected (2010-2014) Installed Wind Generation
Capacity in the U.S., Europe, and Remainder of the World (Source: National
Renewable Energy Laboratory, Wind Power Monthly, and BTM Consult ApS, March
2010) ..................................................................................................................................3-2
Figure 3-2 End-of-Year Operating Wind Capacity by Country and Region: 2008 and
2009 (Source: BTM Consult ApS, March 2010) .................................................................3-8
Figure 3-3 Installed Wind Generation Capacity in the United States by State as of July
2010 (Source: American Wind Energy Association) .........................................................3-9
Figure 3-4 Rayleigh Probability Density Function for Wind Speed ..........................................3-11
Figure 3-5 Coefficient of Performance (Cp) vs. Tip Speed/Wind Speed Ratio ........................3-11
Figure 3-6 Components of a Typical Wind Turbine (Source: Danish Wind Industry
Association, www.windpower.org) ...................................................................................3-12
Figure 3-7 Schematic of Enercon Direct Drive Turbine (Source: Enercon website,
http://www.enercon.dk/en/_home.htm. Retrieved September 10, 2010)..........................3-15
Figure 3-8 Direct-Drive 1.5-MW Generator under Testing at NREL’s National Wind
Technology Center (Source: Lee, Fingersh. NREL Photographic Information
eXchange via http://www.nrel.gov/data/pix/Jpegs/14689.jpg)..........................................3-16
Figure 3-9 Example of an Eight-Pole, Air-Cooled Superconducting Generator (Source:
Abrahamsen, A B, et. al. [73], with permission) ...............................................................3-17
Figure 3-10 Cross-Section of Wind Turbine Blade (Source: DNV-GEC) ................................3-18
Figure 3-11 Bend-Twist Coupling Achieved by Fiber Orientation (Source: Laird, Daniel.
Sandia National Laboratories 2006 Wind Turbine Blade Workshop) ...............................3-19
Figure 3-12 Vortex Generators (Counter-Rotating Array) ........................................................3-20
Figure 3-13 Gearbox Internal Schematic Showing One Planetary and Two Parallel-Shaft
Stages ..............................................................................................................................3-28
Figure 3-14 Cross-Section of Hydrodynamic Drive System for a Wind Turbine (Source:
Voith Brochure, “The WinDrive – An Innovative Drive Train Concept for Wind
Turbines,” Retrieved September 2010 via
www.voithturbo.com/windrive_publications.php3)............................................................3-29
xxv
10581090
Figure 3-15 Electrical Diagram of a Typical Direct Drive Turbine ............................................3-30
Figure 3-16 Example Hybrid Steel/Concrete and Concrete Towers ........................................3-32
Figure 3-17 Transitional Depth Foundations (30-m to 60-m Depths) (Source: “Offshore
Wind Technology” presented by Walt Musial, National Renewable Energy
Laboratory, at AWEA Offshore Wind Workshop, September 9, 2008).............................3-35
Figure 3-18 Floating Foundations (greater than 60-m Depth) (Source: “Offshore Wind
Technology” presented by Walt Musial, National Renewable Energy Laboratory, at
AWEA Offshore Wind Workshop, September 9, 2008)....................................................3-36
Figure 3-19 Manufacturing Plants for Turbine Blades and Other Components (Source:
Presentation from Tom Ashwill at 2008 Sandia Blade Workshop, Sandia National
Laboratories, Albuquerque, NM) ......................................................................................3-38
Figure 3-20 Clipper Windpower’s 2.5 MW Liberty Series Wind Turbine Nacelle, with Four
Modular Permanent-Magnet Generators (Source: Clipper Windpower Inc.)....................3-39
Figure 3-21 Breakdown of Estimated Capital Costs for Land-Based Wind Plants...................3-42
Figure 3-22 Breakdown of Estimated Capital Costs for Selected Installed Offshore UK
Wind Plants (Source: Study of the Costs of Offshore Wind Generation-A Report to
the Renewables Advisory Board and DTI, Offshore Design Engineering (ODE)
Limited, Sept 14, 2006.) ...................................................................................................3-48
th
Figure 3-23 LCOE Probabilistic Analysis Results – Wind Plant 1 (Base Case, 4
Q 2009 $) .........................................................................................................................3-57
Figure 3-24 Distribution of Average Wind Power in the United States. Source:
Renewable Resource Data Center ..................................................................................3-68
Figure 3-25 The Number of O&M Personnel at a Wind Power Plant Versus the Number
of Turbines and Project Rating.........................................................................................3-71
Figure 3-26 Wind Capacity and Forecast: 1990-2018 (Source: BTM Consult ApS, March
2010) ................................................................................................................................3-81
Figure 4-1 Renewable Electricity Projections (Billion kWh/yr) (Source: EIA, Energy
Outlook 2009).....................................................................................................................4-3
Figure 4-2 Breakdown of Biomass Feedstock ...........................................................................4-9
Figure 4-3 The Effect of Biomass Plant Scale on Feedstock Requirements and
Plantation Area Required [43] ..........................................................................................4-10
Figure 4-4 Maximum Volatile Yield of Biomass Fuels Compared to Reference Coals
(Sources: Johnson et al., 2001; Johnson et al., 2002) .....................................................4-21
Figure 4-5 Low-Temperature Devolatilization Kinetics for Fresh Sawdust (Source:
Johnson et al., 2001)........................................................................................................4-22
Figure 4-6 High-Temperature Devolatilization Kinetics for Fresh Sawdust (Source:
Johnson et al., 2001)........................................................................................................4-23
Figure 4-7 Devolatilization Kinetics for Urban Wood Waste (Source: Johnson et al.,
2002) ................................................................................................................................4-23
Figure 4-8 Devolatilization Kinetics for Fresh Switchgrass (Source: Johnson et al., 2002) .....4-24
Figure 4-9 Devolatilization Kinetics for Weathered Switchgrass (Source: Johnson et al.,
2001) ................................................................................................................................4-24
Figure 4-10 Devolatilization Kinetics for Black Thunder PRB Coal (Source: Johnson et
al., 2001) ..........................................................................................................................4-25
10581090
Figure 4-11 Devolatilization Kinetics for Pittsburgh Seam #8 Coal (Source: Johnson et
al., 2001) ..........................................................................................................................4-25
Figure 4-12 Fuel Nitrogen Concentrations for the Fuel Samples Analyzed in Detail
(Source: Johnson et al., 2001; Johnson et al., 2002).......................................................4-27
Figure 4-13 Distribution of Volatile and Char Fuel Nitrogen in Selected Biomass and
Reference Coal Samples (Source: Johnson et al., 2001; Johnson et al., 2002)..............4-28
Figure 4-14 Maximum Nitrogen Volatile Yields for Selected Biomass and Coal Samples
(Source: Johnson et al., 2001; Johnson et al., 2002).......................................................4-29
Figure 4-15 Volatile Nitrogen and Carbon Evolution from Fresh Sawdust...............................4-30
Figure 4-16 Volatile Nitrogen and Carbon Evolution from Urban Wood Waste .......................4-30
Figure 4-17 Volatile Nitrogen and Carbon Evolution from Fresh Switchgrass .........................4-31
Figure 4-18 Volatile Nitrogen and Carbon Evolution from Weathered Switchgrass.................4-31
Figure 4-19 Nitrogen and Carbon Volatilization Normalized to Total Volatiles Formed
from Sawdust ...................................................................................................................4-32
Figure 4-20 Nitrogen and Carbon Volatile Evolution from Weathered Switchgrass
Normalized to Total Volatile Evolution .............................................................................4-32
Figure 4-21 Nitrogen/Carbon Atomic Ratios for Biomass Chars Formed During DTR
Pyrolysis at Various Temperatures ..................................................................................4-33
Figure 4-22 Normalized Nitrogen/Carbon Atomic Ratios for Biomass Solid Fuel and
Char, Compared to Pittsburgh #8 Bituminous Coal and Black Thunder Powder
River Basin Subbituminous Coal......................................................................................4-34
Figure 4-23 Schematic of Biomass Cofiring with Coal in a Utility Boiler ..................................4-39
Figure 4-24 Plan View of the Willow Island Biomass Cofiring System.....................................4-43
Figure 4-25 Elevation View of the Willow Island Biomass Cofiring System .............................4-44
Figure 4-26 Receiving Sawdust at the Willow Island Generating Station ................................4-45
Figure 4-27 Influence of Cofiring in Cyclone Boilers on Furnace Exit Gas Temperature
(Source: Tillman and Payette, 2004)................................................................................4-47
Figure 4-28 Impact of Cofiring on Slag Viscosity. Fuels 2 to 9 are Various Cofiring
Scenarios Using Woody Biomass, Herbaceous Biomass, and Animal Manures with
Eastern Bituminous Coals (Source: Miller et al., 2002)....................................................4-48
Figure 4-29 Design of the Sawdust Injection System at Seward Generating Station ..............4-51
Figure 4-30 NOX Reduction at the Albright Generating Station................................................4-54
Figure 4-31 Representative Wood Fuel Handling System .......................................................4-61
Figure 4-32 Schematic of Biomass-Fired Stoker Boiler Power Plant.......................................4-62
Figure 4-33 Typical Boiler and Turbine-Generator Approach ..................................................4-63
Figure 4-34 Schematic of Biomass-Fired Atmospheric Fluidized Boiler Power Plant..............4-64
Figure 4-35 Air Pollution Control System, Cooling Tower, and Demineralizer for a Typical
Wood-Waste-Fired Plant ..................................................................................................4-66
Figure 4-36 Schematic of Biomass Gasification Application for Combined Heat and
Power ...............................................................................................................................4-70
Figure 4-37 Process Flow Diagram for the Vølund Gasifier at Harboøre ................................4-71
xxvii
10581090
Figure 4-38 Fjernvarme Process Flow Diagram for Coproduction of 6 MW Electric and
10 MW Thermal................................................................................................................4-72
Figure 4-39 Overall Configuration of the Biomass Gasifier at Ruien .......................................4-73
Figure 4-40 Schematic of Biomass Gasification and Power Generation System at
Kokomäki .........................................................................................................................4-74
Figure 4-41 Example Bubbling Bed Retrofit to a Pulverized Coal Boiler (Source: Metso).......4-77
Figure 4-42 The Grubb Curve for Biomass Technologies ......................................................4-99
Figure 5-1 Worldwide PV Industry Growth and Forecast for 2010 (Source: Navigant
Consulting) [41] ..................................................................................................................5-3
Figure 5-2 PV Field of 25-MW DeSoto Next-Generation Solar Energy Center (Source:
FPL) ...................................................................................................................................5-4
Figure 5-3 California Solar Initiative Installed Projects by Month [38]........................................5-6
Figure 5-4 Annual U.S. Installed Grid-Connected PV Capacity by Sector [39]..........................5-6
Figure 5-5 German PV installations between January 2009 and March 2010 showing
surge of projects, particularly larger than 10 kW, at year’s end in anticipation of
declining feed-in tariff. (Source: Solarenergie-Förderverein Deutschland e.V.) .................5-8
th
Figure 5-6 LCOE Probabilistic Analysis: Fixed a-Si Modules, Las Vegas, NV (4 Quarter
2009 $) .............................................................................................................................5-17
Figure 5-7 Worldwide Average PV Module Selling Price vs. Cumulative Sales ......................5-20
Figure 5-8 U.S. Shipments of PV by Market Sector (top) and End Use (bottom) in 2008.
(Source: EIA [16]).............................................................................................................5-40
Figure 5-9 Distribution of PV Applications in Countries Participating in the International
Energy Agency (IEA) Photovoltaic Power Systems Program, 1992–2008. .....................5-41
Figure 6-1 Map Illustrating the "Ring of Fire" [38] ......................................................................6-2
Figure 6-2 Distribution Curve of Geothermal Energy as a Function of Worldwide
Temperature [39]................................................................................................................6-6
Figure 6-3 U.S. Geothermal Resource Potential Map [Estimated Temperatures (°C) at 6
km Depth] (top), 2004 SMU Heat Flow Assessment (Bottom) ...........................................6-8
Figure 6-4 Estimated EGS Resources in the United States (source: USGS 2008) ...................6-9
Figure 6-5 Total Installed Capacity in 2007 (GHC Bulletin, September 2007).........................6-10
Figure 6-6 Total U.S. Installed Geothermal Capacity as of March 2009 [35]...........................6-12
Figure 6-7 Hydrothermal Power Plant at The Geysers, California ...........................................6-13
Figure 6-8 Dual-Flash (Two-Flash) Geothermal Power Plant ..................................................6-17
Figure 6-9 Binary Cycle Geothermal Power Plant (Air-Cooled Design)...................................6-18
Figure 6-10 HDR Geothermal Production Process ..................................................................6-21
Figure 6-11 Hot Sedimentary Aquifer Production Process ......................................................6-23
Figure 6-12 Geothermal-Solar Hybrid Production Process for a Geothermal Flash Steam
Plant .................................................................................................................................6-24
Figure 6-13 Geothermal-Solar Hybrid Production Process for a Hot Water/Brine non-
Flash Resource ................................................................................................................6-25
Figure 6-14 Geothermal-Gas Peaker Production Process .....................................................6-25
Figure 6-15 Geo-Biomass Hybrid Cycle ..................................................................................6-26
10581090
Figure 6-16 The Grubb Curve.................................................................................................6-33
Figure 6-17 Steam Pipelines for the Ohaaki, New Zealand Plant Run Through Land
Used for Grazing and Agroforestry ..................................................................................6-44
2
Figure 7-1 Distribution of Direct-Normal Insolation Worldwide (Top, kWh/m /yr) and U.S.
2
(Bottom, Wh/m /day) (Sources: U.S. DOE and ISET) ........................................................7-8
Figure 7-2 Solar Trough Collector Field at Kramer Junction, California. Note: The
Receiver Tube at the Reflectors’ Focal Point that Transports Heat-Transfer Fluid..........7-10
Figure 7-3 Andasol 1 and 2 Parabolic Trough Plants, Spain (Source: Solar Millennium).......7-10
Figure 7-4 Solar Two Central Receiver in Operation. The two tanks to the left of the
tower base are the storage tanks for the hot and cold molten salt solutions....................7-13
Figure 7-5 25-kW SAIC Dish/Engine System at the DOE Mesa Top Thermal Test Facility
(Source: NREL)................................................................................................................7-16
Figure 7-6 CLFR Array at Liddell Power Station, Australia (Source: SolarPACES).................7-19
Figure 7-7 Cross-Section Diagram of Solar Chimney. Sunlight heats trapped air that then
rises through the cylindrical tower. Water tubes inside the greenhouse-like structure
surrounding the chimney would help retain heat. (Source: C. Pietschiny) .......................7-20
Figure 7-8 Solar Chimney Sculpture (Source: C. Pietschiny) ..................................................7-20
Figure 7-9 Martin Next-Generation Solar Energy Center (Source: FPL) .................................7-26
Figure 7-10 Gemasolar Plant Under Construction, May 2010 (Source: SENER) ....................7-28
Figure 7-11 Gemasolar Molten Salt Storage Tanks Under Construction, June 2010
(Source: SENER) .............................................................................................................7-28
Figure 7-12 Sierra SunTower plant (Source: eSolar)...............................................................7-29
Figure 7-13 Artist’s conception of BrightSource Ivanpah Solar Power Complex. (Source:
BrightSource Energy).......................................................................................................7-30
Figure 7-14 SunCatcher™ Units Deployed at the Maricopa Solar Plant (Source: Tessera
Solar)................................................................................................................................7-32
Figure 7-15 Thirty Infinia PowerDish™ Units Demonstrated in Spain (Source: DOE
EERE) ..............................................................................................................................7-32
Figure 8-1 Earth, Moon & Sun’s Influence on Tides ..................................................................8-3
Figure 8-2 The Earth Tidal Bulge...............................................................................................8-4
Figure 8-3 Typical Hydro Kinetic Water Turbine ........................................................................8-4
Figure 8-4 Tidal Power Density..................................................................................................8-4
Figure 8-5 Vortex Induced Vibrations Oscillates Objects in Fluid Currents. ............................8-12
Figure 8-6 Open Hydro Device at the EMEC Fall of Warness Tidal Test Site .........................8-14
TM
Figure 8-7 Marine Current Turbines SeaGen ........................................................................8-15
Figure 8-8 Pulse Tidal Generator.............................................................................................8-16
Figure 8-9 Swanturbines Cygnet .............................................................................................8-17
Figure 8-10 Tidal Energy Ltd DeltaStream ..............................................................................8-17
Figure 8-11 Open Hydro Gravity Base Installation Illustration .................................................8-18
Figure 8-12 Open Hydro Gravity Base Photo ..........................................................................8-19
Figure 8-13 OpenHydro Deployment Vessel ...........................................................................8-19
xxix
10581090
Figure 8-14 United States Tidal Current Reference Stations...................................................8-21
Figure 8-15 Annual Average Tidal Current Speed Probability Distributions for Dog Island
Transect, Western Passage, Maine .................................................................................8-22
Figure 8-16 Annual Tidal Energy Resource for Six U.S. Tidal Current Sites ...........................8-23
Figure 8-17 Tidal Energy Conversion System Configurations .................................................8-25
Figure 9-1 Wind Blowing over Fetch of Water Produces Waves ...............................................9-3
Figure 9-2 Particle Motion in Different Water Depths ................................................................9-3
Figure 9-3 Vector Field for Particle Motion in Waves [1]............................................................9-4
Figure 9-4 Wave Power Flux .....................................................................................................9-4
Figure 9-5 Schematic of PG&E WaveConnect System ...........................................................9-10
Figure 9-6 Aquamarine Power Oyster .....................................................................................9-14
Figure 9-7 Illustration of Checkmate Anaconda.......................................................................9-14
Figure 9-8 Expected EMEC Testing Configuration in 2011 .....................................................9-16
Figure 9-9 Wave Star 1:10 Machine with Buoys Raised.........................................................9-17
Figure 9-10 Wave Star 1:2 Machine with Buoys Raised..........................................................9-17
Figure 9-11 Floating Power Plant AS Poseidon.......................................................................9-18
Figure 9-12 Seabased AB Linear Generator ..........................................................................9-19
Figure 9-13 BioPower BioWave..............................................................................................9-20
Figure 9-14 Oceanlinx Oscillating Water Column ...................................................................9-20
Figure 9-15 U.S. Wave Energy Resources.............................................................................9-21
Figure 9-16 NOAA Wave Watch III Grid Resolution ................................................................9-22
Figure 9-17 West Coast Reference Stations ...........................................................................9-23
Figure 9-18 Hawaii Reference Stations (Point Makapuu is CDIP 0098)..................................9-23
Figure 9-19 NOAA Wave Watch III Global Coverage ..............................................................9-24
Figure 9-20 Wave Energy Device Principles ..........................................................................9-28
Figure 9-21 Wave Energy Device Concept..............................................................................9-28
Figure 9-22 Levelized COE Comparison to Wind: Oregon Example with Federal and
State Financial Incentives ................................................................................................9-39
3
Figure 10-1 Major North America Rivers and their Yearly Discharges in km /year..................10-2
Figure 10-2 Steps Affecting Hydrokinetic Turbine Efficiency ...................................................10-4
Figure 10-3 Example of a Hydrokinetic Turbine.......................................................................10-5
Figure 10-4 Water Power Density............................................................................................10-5
Figure 10-5 Hydrokinetic Turbine Deployment from Barge at Ruby, Alaska, on the Yukon
River.................................................................................................................................10-7
Figure 10-6 Village of Eagle, Alaska........................................................................................10-7
Figure 10-7 Village of Eagle during Spring Break-Up of 2009 .................................................10-8
Figure 10-8 Hydro Green’s hydrokinetic turbine and the deployment site at Lock and
Dam #2 on the Mississippi River at Hastings, Minnesota ................................................10-9
Figure 10-9 Preliminary Permit Sites in Arkansas Held by Free Flow Power ........................10-10
10581090
Figure 10-10 Pre-test Inspection of Free Flow Power 3-meter Turbine in Flume at the
USGS Conte Anadromous Fish Laboratory ...................................................................10-11
Figure 10-11 Location of Verdant’s Cornwall Ontario Renewable Energy (CORE) project
in the St. Lawrence River. ..............................................................................................10-12
Figure 10-12 Site of the CORE Project on the St. Lawrence River (left), and Verdant
Turbine (right).................................................................................................................10-12
Figure 10-13 Stream Velocity Profile across the Yukon River at the USGS Gauging
3
Station at Eagle, Alaska for a Discharge Rate of 183,000ft /s .......................................10-13
Figure 10-14 Channel Cross-section at the USGS Gauging Station at Eagle, Alaska, at a
3
Discharge Rate of 183,000ft /s.......................................................................................10-14
Figure 10-15 Velocity Versus Discharge at the USGS Gauging Station at Eagle, Alaska.....10-14
Figure 10-16 Velocity Distribution at the USGS Gauging Station at Eagle, Alaska ...............10-15
Figure 10-17 Average Discharge by Month at the USGS Gauging Station at Eagle,
Alaska ............................................................................................................................10-15
Figure 10-18 Average Velocity by Month at the USGS Gauging Station at Eagle, Alaska ....10-16
Figure 10-19 River In-Stream Energy Conversion Devices ...................................................10-17
Figure 10-20 Experimental configuration of the AHRTA Oscillating Turbine .........................10-20
Figure 10-21 Free Flow Power Turbine Generator ................................................................10-21
Figure 10-22 Free Flow 60 Osprey Turbine in Experimental Test Configuration...................10-21
Figure 10-23 Hydro Green Turbines ......................................................................................10-22
Figure 10-24 Hydro Green Turbine Array Configuration ........................................................10-22
Figure 10-25 Lucid Energy Gorlov Helical Turbine ................................................................10-23
Figure 10-26 Lucid Energy Array Configuration.....................................................................10-23
Figure 10-27 New Energy System Concept...........................................................................10-24
Figure 10-28 New Energy EnCurrent Turbine .......................................................................10-25
Figure 10-29 Ocean Renewable Power Corp OCGen Module ..............................................10-25
Figure 10-30 UEK Prototype Demonstrated in 2000 (Source: UEK) .....................................10-26
Figure 10-31 Verdant Power Free Flow Turbine....................................................................10-27
Figure 10-32 Verdant Power Turbine Lowered into the East River Prior to Mounting on a
Monopile.........................................................................................................................10-27
Figure 10-33 Vortex Induced Vibrations Oscillates Objects in Fluid Currents .......................10-28
Figure 10-34 Ice Jam during Spring Breakup: Upstream View of the Yukon River at
Eagle, Alaska .................................................................................................................10-29
Figure 10-35 Site Location Overview, with Chosen Sites Indicated in Yellow .......................10-31
Figure 10-36 Water Depth Profiles at Three Alaska Sites during Typical River
Discharges .....................................................................................................................10-31
Figure 10-37 Average Water Velocities by Month at Three Sites in Alaska ..........................10-32
Figure 10-38 Average Power Densities by Month at Three Sites in Alaska...........................10-32
Figure 11-1 Different Integration Technologies can Combine to Ease Integration of
Renewable Energy Resources.........................................................................................11-4
xxxi
10581090
Figure 11-2 Ancillary Services are Distinguished by their Deployment Times and
Durations........................................................................................................................11-11
Figure 11-3 Wind Output Varies Diurnally, Seasonally, and with Weather Changes.............11-13
Figure 11-4 Hourly Variation of Solar PV System Output Over a Few Days (Recorded
Data for a 100-kW Site Near Albany, NY, 1998) ............................................................11-13
Figure 11-5 Hourly Load Shapes with and without Wind Generation (from DOE 20%
Wind Energy by 2030 (Report DOE/GO-102008-2567, May 2008) ...............................11-14
Figure 11-6 Distribution of Five-Minute Load Requirement Changes with and without
Wind (from GE Consulting 2008). ..................................................................................11-15
Figure 11-7 Example of Wind Power Plant Output Correlation with Distance (from NREL) ..11-15
Figure 11-8 Example of Interaction between Real and Reactive Power from a PV
Inverter ...........................................................................................................................11-17
Figure 11-9 Loss of Wind Turbine Output Due to Transmission Fault and Reconnect.........11-18
Figure 11-10 Example of Wind Plant Output Variation on Area Control Error (ACE)
(Courtesy of Public Service of New Mexico) ..................................................................11-19
Figure 11-11 Relative Cost of Photovoltaic Electricity Due Only to Resource Variability .....11-20
Figure 11-12 Monthly Solar Energy Variation Over One Year ..............................................11-20
Figure 11-13 Daily Variation of Solar PV Output over One Month........................................11-21
Figure 11-14 Wind Integration Costs .....................................................................................11-25
Figure 11-15 Regulating Reserves from Eastern Wind Integration Study (Source:
EWITS)...........................................................................................................................11-26
Figure 11-16 Today’s Typical Distribution Feeder Topology [1] ...........................................11-27
Figure 11-17 Constant-Speed Induction Machine Wind Turbine Generator with Gearbox....11-32
Figure 11-18 Grid Interface for Peetz Table 30-MW Wind Farm ...........................................11-33
Figure 11-19 Variable-Speed Doubly-Fed Induction Machine-Connected WTG with Gear
Box .................................................................................................................................11-34
Figure 11-20 Variable-Speed, Direct-Drive Synchronous Machine Converter-Connected
WTG (No Gearbox) ........................................................................................................11-35
Figure 11-21 Direct-Drive Synchronous Generator with Simple Rectifier at the WTG and
Voltage-Source Converter Interface with the Utility Grid................................................11-36
Figure 11-22 One Leg, Single-Line Diagram of 12-Pulse HVDC System (Source:
Siemens) ........................................................................................................................11-37
Figure 11-23 Alternative Configurations for Connecting Off-Shore Wind via DC-Link
(courtesy of Barrow Off-shore Wind)..............................................................................11-37
Figure 11-24 A Severe Voltage Sag to Less Than 30% Voltage Remaining Voltage for
100 ms (Upper Trace) and Wind Turbine with LVRT Recovering After the Event
(Lower Trace).................................................................................................................11-39
Figure 11-25 Response of Four Commercially Available Wind Turbine Generators to a
Large Voltage Sag ........................................................................................................11-40
Figure 11-26 Illustration of Reactive Compensation and Voltage Control Mechanisms for
Wind Farm Applications .................................................................................................11-41
Figure 11-27 Power Factor Curves for Example Wind Plant .................................................11-42
10581090
Figure 11-28 Fixed and Switched Capacitor Bank Supporting a 2.8-MW Wind Farm in
India (Source: Crompton Greaves Ltd.) .........................................................................11-43
Figure 11-29 Conceptual Layout of a Dynamic VAR Device and Switched Capacitors
Supporting a Wind Farm (Source: American Superconductor) ......................................11-46
Figure 11-30 DSTATCOM Installation (Source: S&C Electric) ..............................................11-46
Figure 11-31 Simplified Schematic of a Static VAR (SVC) or Distribution Static VAR
Compensator (DSVC) ....................................................................................................11-47
Figure 11-32 Configuration of a Commercially-Available DSVC (Source: S&C Electric).......11-48
Figure 11-33 Distribution Static VAR Compensator (DSVC). The Reactive Compensation
Equipment on the Right Side of the Crestwood Substation Supports the 50-MW
Kumeyaay Wind Energy Project in California. (Courtesy of Alliant Energy) ...................11-48
Figure 11-34 Wind Predictions are Adjusted Using Self-Learning Statistical Methods
(Source: AWS Truepower [26]) ......................................................................................11-50
Figure 11-35 Schematic of Typical Wind Energy Forecasting System (Source: AWS
Truewind [26]) ................................................................................................................11-51
Figure 11-36 Graphical Presentation of Wind Flow Data from a Numerical Weather
Prediction Model (Source: AWS Truewind [26]).............................................................11-51
Figure 11-37 Cross-section of a Pumped Hydro Storage Plant (Courtesy Tennessee
Valley Authority) .............................................................................................................11-54
Figure 11-38 Layout and Components of Typical CAES Plant (Source: Ridge Energy
Storage) .........................................................................................................................11-54
Figure 11-39 Principles of Operation for a Vanadium Redox Flow Battery (Courtesy
Sumitomo Electric Industries) ........................................................................................11-56
Figure 11-40 Present Value and Societal Benefits of Energy Storage Technologies ............11-59
Figure 11-41 Storage Operating Benefits in Different Ancillary Services Markets.................11-60
Figure 11-42 Model for Evaluation of Energy Storage with Wind Power ...............................11-67
Figure 11-43 Distributed Controller Results are Aggregated to Manage Area Power and
System Voltage Profiles .................................................................................................11-69
Figure 11-44 Illustration of Cascaded Restoration of Distributed Generators.......................11-70
Figure 11-45 Concept of Distribution Microgrids of Various Sizes and Levels Allowing
Reliability Islands and Grid Tie Operation......................................................................11-72
Figure 11-46 Distributed Controller Must be Integrated with Overall Distribution Control
Systems to Maximize System Value and Reduce Capacity Requirements ...................11-73
Figure 11-47 FRT Limiting Curves Proposed in the New German Grid Codes for
Connecting PV Systems to the Medium-Voltage Power Grid. .......................................11-74
Figure 12-1 Levelized CO2 Emissions Offset vs. % Fossil Fuel Generation ............................12-6
Figure 12-2 Levelized CO2 Emissions Control Cost vs. Cost Premium and CO2
Emissions Offset ..............................................................................................................12-6
Figure 12-3 Levelized CO2 Emissions Control Cost vs. Cost Premium and Fossil Fuel
Mix at 50% Generation from Fossil Fuels ........................................................................12-7
Figure 12-4 Levelized CO2 Emissions Control Cost at 100-MW California Wind Plant vs.
CO2 Emissions Offset and System Generation Cost (4th Quarter 2010 $) ......................12-9
xxxiii
10581090
Figure 12-5 Levelized CO2 Emissions Control Cost vs. Location and System Generation
Cost (4th Quarter 2010 $) ................................................................................................12-9
Figure 12-6 Levelized CO2 Emissions Control Cost vs. System Generation Cost for
100% Biomass-Repowered and Bubbling Fluidized Bed Plants and 10% Biomass
Co-fired Coal Plant (Incremental Cost vs. Coal for Cofired Plant, 3rd Quarter 2010
$,Source: EPRI [7]) ........................................................................................................12-11
Figure 12-7 Levelized CO2 Emissions Control Cost vs. System Generation Cost for 50-
MW Solar Photovoltaic Plants Las Vegas, Nevada (Top) and Columbus, Ohio
(Bottom) (December 2010 $, Source: EPRI [8]).............................................................12-13
Figure 12-8 Levelized CO2 Emissions Control Cost for 50-MW Flash-Steam and Binary-
Cycle Power Plants vs. Base System Generation Cost (December 2010 $, Source:
EPRI [9]).........................................................................................................................12-15
Figure 12-9 Levelized CO2 Emissions Control Cost for Four Solar Thermal Technologies
(December 2010 $, Source: EPRI [10]) .........................................................................12-17
10581090
LIST OF TABLES
Table 1-1 U.S. Energy Consumption by Energy Source, 2004-2008 (in Quadrillion Btu)
(Source: U.S. EIA, 2009 [15]).............................................................................................1-2
Table 1-2 Installed Renewable Energy Generation Capacity (2010) .........................................1-4
Table 2-1 Calculation of Annual Fixed Charge Rates ................................................................2-2
Table 2-2 Calculating Levelized Annual Charges ......................................................................2-4
Table 2-3 Wind Power Generation—Nominal Dollar Terms ......................................................2-6
Table 2-4 Wind Power Generation—Constant Dollar Terms .....................................................2-7
Table 2-5 Sources and Uses of Construction Funds ...............................................................2-15
Table 2-6 Example Income Statement ($1000) .......................................................................2-18
Table 2-7 Example Cash Flow Statement ($1000) ..................................................................2-23
Table 2-8 Example Sensitivity Analysis ...................................................................................2-28
Table 2-9 Confidence Rating Based on Cost and Design Estimate ........................................2-31
1
Table 2-10 Accuracy Range Estimates for RETG Cost Data (Ranges in Percent) .................2-32
Table 2-11 TPC-TPI Adjustment Factors (AF).........................................................................2-34
Table 3-1 Operating Wind Generation Capacity: End of Year 2003–End of Year 2009
(Source: BTM Consult ApS, March 2010) ..........................................................................3-4
Table 3-2 Forecast Wind Generation Capacity Additions: 2010–2014 (Source: BTM
Consult ApS, March 2010) .................................................................................................3-6
Table 3-3 Selected Onshore Wind Turbine Suppliers and Turbine Models over 2.3 MW........3-40
Table 3-4 Areas of Potential Technology Improvement ...........................................................3-43
Table 3-5 Selected Offshore Wind Turbine Suppliers..............................................................3-46
Table 3-6 Wind Plant Site Assumptions...................................................................................3-49
Table 3-7 Wind Plant Design Assumptions..............................................................................3-49
Table 3-8 Assumed Project Site Weather Conditions..............................................................3-50
Table 3-9 Assumed Project Site Wind Regimes ......................................................................3-50
th
Table 3-10 Total Capital Requirement Estimates (4 Quarter 2009 $) ....................................3-52
th
Table 3-11 Operation and Maintenance Cost Estimates (4 Quarter 2009 $) ........................3-54
Table 3-12 Levelized Cost of Electricity With and Without 30% Investment Tax Credit
th
(4 Quarter 2009) .............................................................................................................3-56
Table 3-13 Performance and Cost Estimate Summary for Horizontal-Axis Wind Turbine
Power Plants (4th Quarter 2009 $) ..................................................................................3-58
Table 3-14 Ten Largest Wind Turbine Markets: 2007–2009 ..................................................3-76
Table 3-15 Growth Rates of Installed Wind Capacity in the Top 10 Markets During 2008 ......3-77
xxxv
10581090
Table 3-16 Annual Average Nameplate Ratings (kW) of Wind Turbines Installed by
Country.............................................................................................................................3-78
Table 3-17 Commercial Megawatt-Class Wind Turbine Generators .......................................3-78
Table 3-18 Global Wind Power Development Forecast: 2009-2013........................................3-82
Table 3-19 Global Participants in the Wind Power Industry.....................................................3-83
Table 3-20 Global Market Shares of Wind Turbine Manufacturers in 2006, 2007 and
2008 .................................................................................................................................3-84
Table 4-1 Biomass Energy Consumption in the U.S. Economy by Sector and Type, 2007
(Exajoules) .........................................................................................................................4-2
Table 4-2 Refuse Derived Fuel Properties [45]........................................................................4-12
Table 4-3 Proximate and Ultimate Analyses for Typical Woody Biomass Fuels......................4-13
Table 4-4 Proximate and Ultimate Analyses for Typical Herbaceous Biomass Fuels .............4-14
Table 4-5 Proximate and Ultimate Analyses for Typical Manures ...........................................4-14
6
Table 4-6 Nitrogen and Ash Concentrations in Biomass Fuels (Values in lb/10 Btu) .............4-16
Table 4-7 Ash Analyses of Various Biomass Fuels Compared to Pittsburgh #8 Coal .............4-17
Table 4-8 Ash Analyses of Switchgrass and Other Herbaceous Crops...................................4-17
Table 4-9 Slagging and Fouling Index for Selected Biomass Fuels ........................................4-18
Table 4-10 Ranges in Concentrations of Trace Metals in Woody Biomass (mg/kg in Dry
Wood)...............................................................................................................................4-19
Table 4-11 Ranges in Concentrations of Trace Metals in Woody Biomass Burned at a
Pulp Mill in the Pacific Northwest (mg/kg in Dry Wood) ...................................................4-19
Table 4-12 Volatility Measures for Representative Biomass Fuels..........................................4-20
Table 4-13 Devolatilization Kinetic Parameters for Biomass Fuels and Reference Coals
(400-1700°C)....................................................................................................................4-22
Table 4-14 Chemical Fractionation Analyses of Various Biofuel Ashes ..................................4-35
Table 4-15 Chemical Fractionation of Potassium in Switchgrass for Four Samples................4-36
Table 4-16 Complete Chemical Fractionation of a Switchgrass Sample .................................4-36
Table 4-17 Composition and Characteristics of the Wastewater Treatment Gas
Produced at the T.E. Maxson Treatment Facility in Memphis, Tennessee ......................4-37
Table 4-18 Representative Biomass Cofiring Tests and Demonstrations................................4-40
Table 4-19 Capital Cost of the Willow Island Cofiring Demonstration......................................4-46
Table 4-20 Capital Cost Summary for the Bailly Generating Station Cofiring System.............4-46
Table 4-21 Estimated Capital Cost of the Albright Cofiring Demonstration if Constructed
as a New Facility ..............................................................................................................4-52
Table 4-22 Pre-Treatment Options for Addressing Cofiring Constraints [46]...........................4-57
Table 4-23 Typical Water Rates for Condensing Turbines in Wood Waste-Fired Plants ........4-64
Table 4-24 Typical Water Rates for Backpressure Turbines in Wood Waste-Fired Plants......4-65
Table 4-25 Plant Design Assumptions for Biomass Repowering Base Case (Source:
EPRI [48]).........................................................................................................................4-77
Table 4-26 Raw and Dried Biomass Fuel Composition ...........................................................4-78
10581090
Table 4-27 Plant Performance Summary for PC Coal-fired and Repowered Biomass-
fired Bubbling Fluidized Bed Plants (Source: EPRI [43]) .................................................4-79
Table 4-28 100% Biomass Repowering Performance and Cost Estimates (3Q 2010$,
Source: EPRI [43]) ...........................................................................................................4-80
Table 4-29 LCOE Estimates for 100% Biomass Repowering (3Q 2010 $, Source: EPRI
[48]) ..................................................................................................................................4-81
Table 4-30 Summary of Biomass Cofiring Base Case (Source: EPRI [48]).............................4-82
Table 4-31 Woody Biomass Fuel Quality: Heat Content, Proximate, and Ultimate
Analyses (Source: EPRI [48]) ..........................................................................................4-83
Table 4-32 Performance Estimates for Cofiring Coal and Undried Biomass (Source:
EPRI [48]).........................................................................................................................4-84
Table 4-33 Performance Estimates for Cofiring Coal and Dried Biomass (Source: EPRI
[48]) ..................................................................................................................................4-84
Table 4-34 Performance Estimates for Cofiring Co-Milled Coal and Torrefied Biomass
(Source: EPRI [48]) ..........................................................................................................4-85
Table 4-35 Biomass Cofiring Performance and Cost Summary (3Q 2010$, Source: EPRI
[48]) ..................................................................................................................................4-86
Table 4-36 Levelized Cost of Electricity Estimate for Cofiring Coal and 10% Undried
Biomass (3Q 2010 $, Source: EPRI [48]) ........................................................................4-87
Table 4-37 Design Assumptions for Bubbling BFB Boiler Plants.............................................4-88
Table 4-38 Plant Performance Estimates for BFB Boiler Plants (Source: EPRI [48])..............4-89
Table 4-39 Performance and Cost Estimates for Bubbling Fluidized Bed Boiler Plants
(3Q 2010$, Source: EPRI [48]) ........................................................................................4-90
Table 4-40 Levelized Cost of Electricity Estimates for Biomass-Fired Bubbling Fluidized
Bed Plants (3Q 2010 $, Source: EPRI [48]) .....................................................................4-92
Table 4-41 Design Assumptions for Stoker Boiler Plant ..........................................................4-93
Table 4-42 Plant Performance Estimates for Stoker Boiler Plant (Source: EPRI [42] [43]) .....4-94
Table 4-43 Performance and Cost Estimates for Stoker Boiler Plant (3Q 2010$, Source:
EPRI [42] [43])..................................................................................................................4-95
Table 4-44 Levelized Cost of Electricity Estimates for 50-MW Stoker Boiler Plant Fired
by Woody Biomass (3rd-Quarter 2010 $, Source: EPRI [47] [48])...................................4-97
Table 4-45 Performance and Cost Estimates for Landfill Gas Combustion.............................4-98
Table 5-1 PV Technology Summary ........................................................................................5-11
Table 5-2 Photovoltaic Site Assumptions ................................................................................5-11
Table 5-3 First-Year Electricity Generation and Capacity Factors ...........................................5-12
Table 5-4 Utility-Scale Solar Photovoltaic Power Plant Total Capital Requirement
Estimates (4th Quarter 2009 $)........................................................................................5-14
Table 5-5 Utility-Scale Solar Photovoltaic Power Plant Operation and Maintenance Cost
Estimates (4th Quarter 2009 $)........................................................................................5-15
Table 5-6 Levelized Cost of Electricity With and Without 30% Investment Tax Credit (4th
Quarter 2009 $)................................................................................................................5-16
Table 5-7 Utility-Scale Solar Photovoltaic Power Plant Performance and Cost Estimate
th
Summary (4 Quarter 2009 $)..........................................................................................5-18
xxxvii
10581090
Table 5-8 Site Data for 27 Locations in the Continental U.S. ..................................................5-22
Table 5-9 Information Included in Clean Power Estimator Databases.....................................5-23
Table 5-10 Rooftop Case-Study Electric-Rate Scenarios ........................................................5-24
Table 5-11 Rooftop Case-Study Utility Rates ..........................................................................5-25
Table 5-12 Rooftop Case-Study System Performance ............................................................5-25
Table 5-13 Rooftop Case-Study Capital Costs ........................................................................5-26
Table 5-14 Rooftop Case-Study Economic-Analysis Results ..................................................5-27
Table 5-15 Solar PV Power Plant Capital, O&M, and Lease Cost Projections for Fixed
Flat-Plate Thin-Film Photovoltaic Power Plants (Dec-09 $) .............................................5-29
Table 5-16 Solar PV Power Plant Capital, O&M, and Lease Cost Projections for One-
Axis Tracking Crystalline Silicon Flat-Plate Photovoltaic Power Plants (Dec-09 $) .........5-29
Table 5-17 Solar PV Power Plant Capital, O&M, and Lease Cost Projections for Two-
Axis Tracking High Concentration Photovoltaic Power Plants (Dec-09 $) .......................5-30
Table 5-18 Solar PV Power Plant Capital and O&M Cost Scaling Factors vs. Rated Plant
Output Relative to a 5-MWac Plant....................................................................................5-30
Table 5-19 Utility-Scale PV Power Plant O&M Cost Estimates ...............................................5-37
Table 6-1 Proposed Geothermal Resource Temperature Classification....................................6-4
Table 6-2 Installed Geothermal Power Worldwide [28, 34]......................................................6-11
Table 6-3 US Installed Capacity as of March 2009 [35]...........................................................6-12
Table 6-4 Estimated Costs for a 50-MW Geothermal Plant (Dec-2009 $) ...............................6-38
Table 6-5 Estimated Geothermal Development Costs (Dec-2009 $).......................................6-38
Table 6-6 Design Assumptions for Geothermal Plants ............................................................6-40
Table 6-7 Performance and Total Capital Requirement Estimates for Geothermal Power
Plants (December 2010 $, Source: EPRI [41]) ................................................................6-41
Table 6-8 Levelized Cost of Electricity Estimates for Geothermal Power Plants (30-year
Project Life, Constant December 2010 $, Source: EPRI [41]) .........................................6-42
Table 6-9 Land Area Requirements for Current Renewable Technologies .............................6-43
Table 6-10 Averages of Four Significant Pollutants, as Emitted from Geothermal and
Coal Facilities, are Listed in the Table Below (Source: GEA 2005 [27]) ..........................6-45
Table 7-1 Central Receiver Demonstration Projects................................................................7-12
Table 7-2 Technology Monitoring Guide Solar Thermal Power Plants ....................................7-22
Table 7-3 Technology Process Development Map: Solar Thermal Power Plants ...................7-23
Table 7-4 Comparison of the Solar Two and Gemasolar Projects...........................................7-27
Table 7-5 Performance and Cost Estimates for Solar Thermal Technologies (December-
2010 $, Source: EPRI [19]) ..............................................................................................7-34
Table 7-6 Selected Data for Solar Trough Systems ................................................................7-37
Table 7-7 Peak Central-Receiver System Efficiency Components ..........................................7-45
Table 7-8 Annual Average Central Receiver System Efficiency Components.........................7-46
Table 8-1 Offshore In-Stream Tidal Energy Conversion Device Developers...........................8-27
Table 8-2 Current Issued and Pending FERC Tidal In-Stream Preliminary Permits (as of
October 4, 2010) ..............................................................................................................8-30
10581090
Table 8-3 Cost and Performance Estimates for Selected Tidal Power Plants .........................8-32
Table 8-4 Cost Estimates for Selected U.S. Feasibility Evaluation Sites (Dec-2009 $)...........8-33
Table 8-5 Installed and Planned U.S. Tidal Energy Capacity ..................................................8-36
Table 9-1 Wave Energy Conversion Device Developers as of November 2008......................9-29
Table 9-2 FERC Active and Pending Preliminary Permits .......................................................9-34
Table 10-1 Manning Coefficients for Representative River Substrates ...................................10-3
Table 10-2 In-Stream River Energy Conversion Device Developers .....................................10-19
Table 10-3 Cost and Performance Estimates for Alaska In-Stream River Power Plants.......10-33
Table 10-4 Cost Estimates for Three Feasibility Evaluation Sites .........................................10-34
Table 10-5 Major Environmental Issues and Mitigation Recommendations ..........................10-35
Table 10-6 Installed and Planned U.S. River In-Stream Power Capacity (MW) ....................10-36
Table 11-1 Comparison of Non-Thermal Renewable Generation Technologies .....................11-2
Table 11-2 Functions and Services Provided by Generation...................................................11-6
Table 11-3 Comparison of Output Controllability for Various Generation Technologies
(EPRI, T. Key, November 2010) ......................................................................................11-8
Table 11-4 Distributed Power System Performance Expectations at Various Connection
Points in the Electric System..........................................................................................11-22
Table 11-5 Summary of Integration Technical Impacts .........................................................11-24
Table 11-6 Summary of Wind Integration Cost Impacts ($/MWh)..........................................11-25
Table 11-7 Grid Penetration Scenarios and Changing Role of Distribution Generation .......11-30
Table 11-8 Types and Functions of Custom Power Devices per Draft IEEE Standards
Project 1409 ...................................................................................................................11-45
Table 11-9 Advantages and Disadvantages of Candidate Energy Storage Technologies
to Support Wind Power Energy Variability .....................................................................11-63
Table 11-10 Advantages and Disadvantages of Candidate Energy Storage
Technologies to Support Wind Power Energy Intermittency and Output Fluctuations
(Continued) ....................................................................................................................11-64
Table 11-11 Energy Storage Characteristics by Application (Megawatt-scale) ....................11-65
Table 11-12 Energy Storage Characteristics by Application (Kilowatt-scale) ........................11-71
Table 12-1 Lifetimes in the Atmosphere and Relative Infrared Absorption Strengths of
the Greenhouse Gases ....................................................................................................12-1
Table 12-2 Fossil Carbon Intensity of Coal, Oil, Natural Gas and Wood Fuels .......................12-2
Table 12-3 Carbon Intensity of Generation Technologies .......................................................12-3
Table 12-4 Wind Plant Performance and Cost vs. Location (4th Quarter 2010 $, Source:
EPRI [6])...........................................................................................................................12-8
Table 12-5 Levelized Cost of Electricity for 50-MW Biomass-Fired Stoker and Fluidized
Bed Boiler Power Plants ($/MWh, 3rd Quarter 2010 $, Source: EPRI [7]).....................12-10
Table 12-6 Performance and Cost Estimates for 10-MW Solar Photovoltaic Power Plants
(4th Quarter 2010 $, Source: EPRI [8]) ..........................................................................12-12
Table 12-7 Levelized Cost of Electricity for 50-MW Flash-Steam and Binary-Cycle
Geothermal Power Plants (December 2010 $, Source: EPRI [9])..................................12-14
xxxix
10581090
Table 12-8 Constant-Dollar Levelized Cost of Electricity for Solar Thermal Technologies
(December 2010 $) ........................................................................................................12-16
10581090
1
INTRODUCTION
1.1 Background
Renewable energy technologies are those that utilize inexhaustible or naturally replenished
resources to generate electricity with minimal environmental impact. Energy sources addressed
in this update of the Renewable Energy Technology Guide (RETG) (formerly known as the
Renewable Energy Technical Assessment Guide (TAG-RE)) include wind, biomass, solar
photovoltaic, geothermal, solar thermal, ocean tidal, wave energy, and river instream energy
conversion.
Passage of the Public Utilities Regulatory Policies Act (PURPA) in 1978 was a turning point for
domestic renewables. PURPA was drafted in response to the oil crisis of the early 1970s and
guaranteed a market at some price for any energy an alternative provider could produce. Solar
and wind power in particular made great strides during this period. However, to some extent,
lower oil prices during the 1980s and 1990s lessened interest in, development of, and
government support for renewable technologies.
Recent concerns about the long-term impacts of rising concentrations of carbon dioxide and
other greenhouse gases have brought renewed vigor to the industry. The 1992 United Nations
Framework Convention of Climate Change (UNFCCC) set an overall framework for
intergovernmental efforts to tackle the challenges posed by global climate change and has been
ratified by 192 countries. The UNFCCC next adopted the Kyoto Protocol on Climate Change in
1997, in which industrialized nations (except for the United States and a few other countries)
agreed to set binding emission targets for six greenhouse gases. The Kyoto Protocol has been a
critical driver of subsequent funding for renewable energy projects in the European Union and
elsewhere, to a lesser degree. As a result, Germany, Japan, the United Kingdom, and other
European Union countries have set aggressive goals for meeting increasingly large percentages
of their energy needs with renewable energy technologies.
In December 2007, the UNFCCC met in Indonesia and adopted the Bali Roadmap for
developing a new negotiating process to tackle climate change with the goal of completing this
by 2009. Part of the strategy for the Bali agreement is to wait to finalize the accord until a new
administration is in place in Washington, D.C. In any case, renewable energy is anticipated to be
a major component of any solution agreed to by international negotiators and implemented by
international governments. In January 2008, in accordance with its goals to reduce greenhouse
gas emissions, the EU adopted country-specific goals for achieving overall 20% renewable
energy usage by the year 2020. And in North America, governments – from California to the
eastern states that make up the Regional Greenhouse Gas Initiative (RGGI) – are taking actions
to require reductions of greenhouse gases from the electric power sector and others.
1-1
10581090
Introduction
New concerns about energy security and independence in the wake of terrorist attacks have also
renewed interest in developing domestic energy resources in the United States, including non-
fossil renewable options. Indeed, as of November 2010, 29 individual states and the District of
Columbia had passed renewable energy portfolio standard (RPS) laws that mandate energy
providers to sell a certain percentage of power to be generated from eligible renewable resources.
Another six states had goals for increasing renewable electricity’s contribution, with some states
provided financial incentives to utilities to achieve their goals.
For a variety of regulatory, social, economic, and environmental reasons, interest in renewable
energy continues to grow. The main factors are the targeted government subsidies such as the
Production Tax Credit (PTC), Investment Tax Credit (ITC), and state renewable portfolio
standards. As renewable energy technologies evolve and become more cost-competitive, and
their environmental benefits become more valuable, they are being deployed with increasing
frequency and public support.
Table 1-1 and Figures 1-1 and 1-2 show the relative contributions of renewable energy to total
U.S energy supply and electricity generation during 2005 through 2009, based on data reported
by the U.S. Department of Energy (DOE) Energy Information Administration (EIA) [1, 2].
During 2009, renewable energy contributed just over 8% of total energy and 10% of total
electricity generation, of which most was contributed by hydroelectric and biomass energy.
Table 1-1
U.S. Energy Consumption by Energy Source, 2004-2008 (in Quadrillion Btu)
(Source: U.S. EIA, 2009 [15])
Energy Source 2004 2005 2006 2007 2008 2009

Fossil Fuels
Coal 22.466 22.797 22.447 22.749 22.385 19,761
Net Coal Coke 0.138 0.044 0.040 0.036 0.049 -0.024
1
Natural Gas 22.909 22.561 22.224 23,702 23.791 23.362
2
Petroleum 40.292 40.391 39.955 39.769 37.279 35.268
Subtotal 85.805 85.793 84.687 86.246 83.496 78,368
Nuclear Electric 8.222 8.161 8.215 8.455 8.427 8.349
Power
Renewable Energy
Hydroelectric 2.690 2.703 2.869 2.446 2.511 2.682
Geothermal 0.341 0.343 0.343 0.349 0.360 0.373
Biomass 3.010 3.117 3.277 3.503 3.852 3.883
3
Solar 0.065 0.066 0.072 0.081 0.097 0.109
Wind 0.142 0.178 0.264 0.341 0.546 0.697
Subtotal 6.247 6.406 6.824 6.719 7.366 7.744
Imports 0.039 0.084 0.063 0.107 0.112 0.117
U.S. TOTAL 100.313 100.445 99.790 101.527 99.402 94.578
1. Natural gas data include supplemental gaseous fuels.
2. Petroleum products supplied, including natural gas plant liquids and crude oil burned as fuel.
3. Solar only include solar thermal electric and not solar thermal heating or photovoltaics.
1-2
10581090
Introduction
94.60 Quadrillion Btu/year 7.74 Quadrillion Btu/yr
Coal
Natural Gas
23%
23%
Solar 0.1%
Renew able Biom ass 4.2%
Energy Geotherm al 0.4%
8.2% Hydroelectric 2.8%
Wind 0.7%
Nuclear
Pow er
8.8%
Petroleum
35%
Figure 1-1
Contributions of Renewable Energy and other Sources to Total U.S. Energy Supply during
2009 (U.S. EIA, 2010 [1])
3,814 Billion kWh/yr 379 Billion kWh/yr

Coal
46%
Other
0.3%
Solar 0.02%
Biom ass 0.7%
Renewable Geotherm al 0.4%
Energy Hydro 6.9%
9.9% Wind 1.9%
Natural Gas Nuclear Power

22% Petroleum 21%
0.9%
Figure 1-2
Contributions of Renewable Energy and other Sources to Net U.S. Electricity Generation
during 2009 (U.S. EIA, 2009 [2])
Table 1-2 provides estimates of the installed capacity by generation source, including renewable
technologies in the U.S. as of the end of 2009 (unless otherwise stated). Note that some of these
sources are growing rapidly. For example, global wind capacity is estimated to have climbed
from 59 GW at the end of 2005 to 94 and 120.8 GW by end of 2007 and 2008, respectively.
1-3
10581090
Introduction
Renewable energy resources are broadly available across the United States and throughout the
world. Some resources are almost universally available; others are limited to particular areas.
Locations with high annual insolation are obviously best suited to solar power. Solar energy can
still be usefully and economically applied in locations with lower quality resources including,
in some cases, extreme northern or southern latitudes. Wind energy is generally deployed along
coastlines or mountain passes with reliable sustained winds. The upper plains of the American
Midwest also host some of the world’s best wind resources and a number of the newest and
largest wind power projects. In addition, the Midwest is an excellent source of biomass, which
can be grown through managed agricultural programs to provide a continuous supply of biomass
fuels. Geothermal resources are concentrated in geologically active areas such along the Pacific
“Ring of Fire” that includes the Philippines, Indonesia, and California. The resources for small
hydroelectric systems exist in most countries.
The benefits of using renewable energy are many and extend beyond issues of abundance and
environmental impact. Diversifying energy resources provides security against economic or
political events that may impact one particular resource. They offer some protection against risks
associated with fluctuating fossil-fuel prices and supplies. Investment in renewable energy can
contribute to local economic growth and employment. Many renewable energy plants can be
built in a modular fashion proportionate to load growth patterns and local needs.
Table 1-2
Installed Renewable Energy Generation Capacity (2010)
Worldwide
U.S.
Technology (Including U.S.)
(MW)
(MW)
Wind 36,698 Est. 180,000

Biomass (2008) 11,940 55,000
1
Geothermal 3,086 10,7151
Solar Photovoltaics2 2,000 21,500
Solar Thermal:
Parabolic Trough 421 950
Dish/Engine <1 2.5
Central Receiver <1 45
Compact Linear Fresnel Reflector 5.0 9.4
Low-Impact and Emerging Hydro:

Ocean Tidal Energy < 0.1 3
Ocean Wave Energy < 0.1 <1

1. Geothermal capacity data are for Mid-2010.
2. PV data reflect module sales; installed capacity is likely similar.
1-4
10581090
Introduction
The value of renewable energy is determined by a number of factors, including its ability to meet
mandated renewable energy capacity requirements, respond to customer demands for renewable
energy, and differentiate a power provider from its competitors. Renewables can also offer
significant environmental benefits such as offsetting carbon dioxide, NOX, and SO2 emissions.
Some renewable energy technologies such as biomass and geothermal (and to a lesser extent
some solar thermal technologies) are dispatchable, making them valuable generation assets.
The output of non-dispatchable technologies such as solar PV can correspond well with summer
afternoon load peaks. And EPRI’s ongoing efforts to develop accurate forecasting models will
improve the value of wind energy in a company’s generation portfolio.
1.2 EPRI’s Role
Since its inception in the 1970s, EPRI has conducted programs to assess, evaluate, and advance
renewable power technologies. EPRI has published periodic updates to the Technical Assessment
Guide—TAG-RE®) since the 1970s, a document containing the most up-to-date technical and
®
economic data available on generation technologies. The EPRI TAG —now known as the EPRI
RETG—has become the recognized industry standard for information essential in preliminary
planning for generation technology evaluation and application. Renewable energy technologies
were addressed in the 1997 and earlier editions of the EPRI TAG® as well as the 1997 joint
EPRI-U.S. DOE report Renewable Energy Technology Characterization [3].
EPRI has also published several editions of its Technical Assessment Guide – Distributed
Resources (TAG-DR) [4], which provides a consistent and credible set of cost and performance
data involving distributed resources—defined as decentralized energy generation, storage,
distribution, and end-use technologies in locations that benefit the electric system, specific
customers, or both. The TAG-DR also addressed renewable technologies.
EPRI project managers concluded in 1999 that renewable energy technology had grown
sufficiently and was developing and changing so rapidly that it would be worthwhile to prepare
a dedicated Renewable Energy Technical Assessment Guide (TAG-RE) and update it annually,
beginning in 2000. The inaugural edition has been updated and expanded annually and
subsequent editions were issued each year through 2009 ([5] through [15]). Beginning with the
2009 edition, the TAG-RE was renamed the Renewable Energy Technology Guide (RETG).
The 2010 RETG update includes new performance and cost data for goethermal and biomass
power technology and describes other new studies and project results for each of the renewable
energy technologies, gathered during the past year.
1.3 Objective
This update of the RETG is based on the 2009 EPRI RETG, which was issued in December 2009
It was later revised and reissued in June 2010 in order to update the wind and solar photovoltaics
cost information [5]. The RETG fulfills two roles. First, it serves as a one-stop information
source for a wide variety of renewable energy technologies evaluated on a consistent basis.
EPRI’s experience and expertise, along with its reputation for objectivity and credibility, make
RETG a uniquely valuable reference. The various renewable technologies have been evaluated
1-5
10581090
Introduction
with sufficient technical depth to enable planners to conduct an accurate assessment of the
renewable energy options that are available to them.
Second, the RETG provides information to educate all interested parties in the key technology
concepts, language, and issues related to renewables. It offers not only technical depth but also
breadth of coverage across many disciplines including project planning, resource availability and
management, regulatory processes, operating and maintenance requirements, market potential,
and future developments.
The information summarized in the RETG encompasses more than three decades of EPRI
research on renewable technologies and draws upon a large foundation of knowledge built by the
U.S. DOE and other national and international experts. This information has been distilled and
synthesized to provide the readers of RETG the most up-to-date, accurate, practical intelligence
available that they can use to investigate, plan, and deploy the latest renewable energy
technologies.
1.4 Definitions and Units
Unless otherwise stated, all dollar figures cited in the RETG are assumed to be U.S. dollars
corrected for inflation to end-of-year 2010 values.
Throughout the RETG, most notably the Biomass and Solar Thermal chapters, the subscripts “e”
(e.g., kWe or MWe) and “t” (e.g., kWt or MWt) are used to differentiate between electrical and
thermal power, respectively. In the Solar Photovoltaic chapter, the terms kWp or MWp may be
used to denote “peak” PV cell or module capacity.
1.5 Scope
The RETG consists of 12 chapters. Following Chapter 1: Introduction, they are:
Chapter 2: Economic Methodology and Assumptions provides a detailed discussion of the basis
upon which data were gathered and presented.
Chapter 3: Wind Power covers one of the most widespread and economically competitive
renewable energy technologies. Wind power systems progressed substantially through the past
decades as a result of government incentives, with a steady trend of cost reduction. Wind power
is particularly popular in Europe, due partly to aggressive government goals and public support,
and is considered to be a commercially-established and competitive grid-power technology.
Chapter 4: Biomass Energy focuses on biomass fuel and combustion properties, direct
combustion, gasification, and cofiring with coal in utility boilers. This chapter addresses resource
availability, technology status, performance and cost history and projections, operating and
maintenance labor requirements, environmental emissions, installed capacity, and cost and
economic issues.
1-6
10581090
Introduction
Chapter 5: Solar Photovoltaics (PV) addresses power systems that convert sunlight directly into
electricity. These solid-state electronic devices have no moving parts, no fluids, no noise, and no
emissions of any kind. These benefits have positioned PV to be the preferred power technology
for many remote applications. In addition, developments on the horizon promise to significantly
reduce the costs of PV and open up new markets for distributed applications, such as innovative
building-integrated technologies, and increasingly, MW-scale projects.
Chapter 6: Geothermal provides the latest information on systems that generate electricity by
tapping underground steam reservoirs. Although available in relatively few areas, geothermal
power has been a practical reality in California, Italy, and Asia for decades. This chapter also
addresses the potential in both developed and developing countries for the more common
geothermal hot water and liquid-dominated hydrothermal resources.
Chapter 7: Solar Thermal addresses systems that use concentrated sunlight to heat a working
fluid and generate electricity in a thermodynamic cycle. Such systems range in scale from
25-kW reflector dish-generator systems to 85-MW solar thermal trough plants that concentrate
solar energy using parabolic trough reflectors to heat oil, which is then used to generate steam
and electricity. Another technology option is central receiver technology, which uses hundreds
of mirrors to focus sunlight onto a central receiver. The 10-MW Solar Two demonstration
incorporated thermal storage through the use of a molten salt heat transfer fluid and a large
storage tanks.
Chapter 8: Ocean Tidal Energy addresses low-impact and emerging hydro systems that capture
ocean tidal current energy to generate electricity. These systems are distinct from traditional
hydroelectric installations, which typically generate electricity via a turbine powered by water
impounded behind a dam. Many ocean energy generation technologies are derived from
analogous technologies in the wind turbine industry.
Chapter 9: Ocean Wave Energy addresses low-impact and emerging hydro systems that convert
the kinetic energy of ocean waves into electricity.
Chapter 10: River In-Stream Energy reviews the potential for generating electrical power from
river energy. Technology and developer companies are working to modify tidal energy turbines
for use in river environments.
Chapter 11: Integrating Renewable Energy with the Grid addresses power electronics, energy
storage, and other technologies need to integrate large wind plants and other intermittent
generation technologies into the electricity grid. Energy storage technologies such as batteries,
pumped hydro, compressed air energy storage (CAES), flywheels, and superconducting
magnetic energy storage (SMES) can be selected and designed to absorb short, intermediate, and
longer-term fluctuations of output from seconds to days. Other integration technologies, such as
line compensation, power electronics, integration with hydro, and solar and wind energy
forecasting, address short-term fluctuations.
Chapter 12: Greenhouse Gas Emissions Control discusses the role that renewable energy power
technologies can play in greenhouse gas emissions reduction and includes example calculations
showing how that role can be quantified.
1-7
10581090
Introduction
1.6 References
1. Primary Energy Consumption by Source, Selected Years, 1949-2009, U.S. Energy
Information Administration, Washington, D.C.: 2010:
http://www.eia.gov/emeu/aer/pdf/pages/sec1_9.pdf.
2. Electricity Net Generation: Electric Power Sector, 1949-2009, U.S. Energy Information
Administration, Washington, D.C.: 2010: http://www.eia.gov/emeu/aer/pdf/pages/sec8_9.pdf
3. Renewable Energy Technology Characterizations. EPRI, Palo Alto, CA, and the Office
of Utility Technologies, U.S. Department of Energy, Washington, D.C.: December 1997.
TR-109496.
4. TAG Technical Assessment Guide, Distributed Resources-1997. EPRI, Palo Alto, CA:
November 1997. TR-105124-R1.
5. Renewable Energy Technology Guide – RETG 2009. EPRI, Palo Alto, CA: 2010. 1021379.
6. Renewable Energy Technology Guide—RETG: April 2009 Update. EPRI, Palo Alto, CA:
2009. 1019300.
7. Renewable Energy Technical Assessment Guide – TAG-RE 2008. EPRI, Palo Alto, CA:
December 2008. 1015801.
March 2008. 1014182.
March 2007. 1012722.
December 2005. 1010407.
December 2004. 1008366.
December 2003. 1004938.
December 2002. 1004196.
December 2001. 1004034.
15. Renewable Energy Technical Assessment Guide–TAG-RE 2000. EPRI, Palo Alto, CA:
February 2001. 1000574
1-8
10581090
2
ECONOMIC METHODOLOGY AND ASSUMPTIONS
The economic methodology and assumptions presented here provide a uniform framework to
evaluate the economic feasibility of alternative renewable energy projects on a consistent basis.
Since both regulated utility and unregulated non-utility generators are developing renewable
energy projects, this section presents separate methodologies to apply depending on the
developer type.
The methodology descriptions presented below are excerpted from the 1999 Technical
Assessment Guide, Volume 3, Revision 8, Fundamentals and Methods–Electricity Supply [1].
The methodology is illustrated by a 200-MW wind project example using cost and performance
data from a 2009 EPRI report [2].
2.1 Regulated Utility Power Projects
The total annual cost of a regulated utility power project consists of the annual fixed charge on
the capital investment plus the operation and maintenance (O&M) costs. The latter costs include
operation and maintenance labor and material, fuel, raw materials and utilities, waste disposal
fees, and other costs.
2.1.1 Fixed Charge Rates
A very useful concept is the capital fixed charge rate or fixed charge rate. This rate represents
the annual charges customers would have to pay each year so that the utility recovers its capital-
related revenue requirements, and is expressed as a percentage of the booked cost of a plant.
Fixed charge rates are typically measured in one of three ways:
• Annual fixed charge rate
• Nominal levelized fixed charge rate
• Real levelized fixed charge rate.
Subsections 2.1.2 through 2.1.4 address these concepts and their derivation as well as the
components and calculation of annual fixed charge rates [1].
2.1.1.1 Annual Fixed Charge Rates
Annual fixed charge rates express the annual capital revenue requirements as a percentage of the
booked cost. In Table 2-1, for example, the total present value of the annual capital charge or
booked cost is $604.27 million, and the annual capital revenue requirements based on that
2-1
10581090
Economic Methodology and Assumptions
booked cost are shown in Column (1). The annual fixed charge rates are the ratio of the annual
capital revenue requirements to the booked cost of $604.27 million, and are shown in Column
(2). The annual fixed charge rates decline over time as the annual capital revenue requirements
decline. So, for example, in the first year, the annual fixed charge rate is 19.09% and declines to
5.32% by the end of the book life. The assumed debt/equity ratio in Table 2-1 is 1.5.
Table 2-1
Calculation of Annual Fixed Charge Rates
Nominal Annual
Capital Revenue
Fixed Charge Rates
Requirements
End of Year (Percentage of
($ Million, Nominal)
Booked Cost)
(1) (2)
2011 $115.34 19.09%
2012 112.38 18.60%
2013 108.34 17.93%
2014 104.46 17.29%
2015 100.74 16.67%
2016 97.16 16.08%
2017 93.72 15.51%
2018 90.41 14.96%
2019 87.20 14.43%
2020 84.02 13.90%
2021 80.83 13.38%
2022 77.65 12.85%
2023 74.46 12.32%
2024 71.28 11.80%
2025 68.09 11.27%
2026 64.91 10.74%
2027 61.72 10.21%
2028 58.54 9.69%
2029 55.35 9.16%
2030 52.17 8.63%
2031 48.98 8.11%
2032 46.49 7.69%
2033 44.70 7.40%
2034 42.90 7.10%
2035 41.11 6.80%
2036 39.32 6.51%
2037 37.52 6.21%
2038 35.73 5.91%
2039 33.94 5.62%
2040 32.14 5.32%
Booked Cost $604.27
($ millions)
2-2
10581090
2.1.1.2 Nominal Levelized Annual Charges and Nominal Levelized Fixed Charge Rates
The second type of fixed charge is the nominal levelized fixed charge rate, which translates
booked costs into a constant annual nominal dollar charge with the same present value as the
actual annual capital revenue requirements. Calculating the nominal levelized annual charge can
be thought of as a two-step process (see Example 2-1).
Example 2-1:
The first step is to calculate the present value of the annual capital revenue requirements.
Column (1) of Table 2-2 below shows the annual capital revenue requirements in Table 2-1.
Since the annual revenue requirements are in nominal terms, they must be discounted using the
nominal discount rate. Column (2) shows the annual present value factors used to bring the costs
from end-of-year to present value. Column (3) lists the real or constant-dollar revenue
requirements that have inflation factored out of the values (e.g., $115.34/(1.03)1 = $111.98 for
the year 2007). Column (4) lists the annual present value factors based on the real discount rate.
Once the present value has been calculated, the levelized annual charges and the fixed charge
rates can be derived as noted in the table.
It is important to note that the levelized fixed charge rate depends on the capital structure and
cost of money, tax depreciation, construction length, book life, and taxes.
2.1.1.3 Real Levelized Annual Charges and Real Levelized Fixed Charge Rates
The third type of fixed charge rate is real levelized fixed charge. It is the fraction of the booked
cost, expressed in constant dollar terms, that customers would have to pay annually over the
book life of a plant to cover capital revenue requirements. The real levelized annual charge and
fixed charge rate are calculated the same way as their nominal counterparts.
However, real levelized fixed charges have an important advantage over their nominal
counterparts: they do not reflect general inflation. The nominal levelized charge reflects
assumptions about future inflation. If a power contract is specified in nominal terms, there are
risks associated with expected inflation. If inflation is less than expected, the buyer will pay too
much and the seller will earn greater-than-anticipated profits. The opposite would be true if
inflation were greater than anticipated. On the other hand, if a contract is specified in real terms,
the risks to the buyer and seller can be avoided. The real levelized charge can be adjusted for
actual inflation over the term of the contract.
How are levelized fixed charge rates used? Generation alternatives are often compared on the
basis of levelized costs. With the appropriate levelized fixed charge rate, the booked cost of
alternatives can be converted to levelized annual cost terms. The EPRI financial model computes
levelized fixed charge rates, which can be used to evaluate generating and other utility
investment options.
2-3
10581090
Table 2-2
Calculating Levelized Annual Charges
Nominal Revenue Requirements Real Revenue Requirements
Capital Revenue Capital Revenue
End of Present Value Present Value
Requirements (2) Requirements (4)
Year (1) Factors (3) Factors
($ Million) ($ Million)
(1) (2) (3) (4)
2011 $115.34 0.9193 $111.98 0.9469
2012 112.38 0.8451 105.93 0.8966
2013 108.34 0.7769 99.14 0.8490
2014 104.46 0.7142 92.81 0.8039
2015 100.74 0.6566 86.90 0.7612
2016 97.16 0.6036 81.37 0.7208
2017 93.72 0.5549 76.21 0.6825
2018 90.41 0.5101 71.37 0.6462
2019 87.20 0.4690 66.83 0.6119
2020 84.02 0.4311 62.52 0.5794
2021 80.83 0.3964 58.39 0.5486
2022 77.65 0.3644 54.46 0.5195
2023 74.46 0.3350 50.70 0.4919
2024 71.28 0.3079 47.12 0.4658
2025 68.09 0.2831 43.71 0.4411
2026 64.91 0.2603 40.45 0.4176
2027 61.72 0.2393 37.34 0.3954
2028 58.54 0.2199 34.38 0.3744
2029 55.35 0.2022 31.57 0.3546
2030 52.17 0.1859 28.88 0.3357
2031 48.98 0.1709 26.33 0.3179
2032 46.49 0.1571 24.26 0.3010
2033 44.70 0.1444 22.65 0.2850
2034 42.90 0.1328 21.11 0.2699
2035 41.11 0.1221 19.63 0.2556
2036 39.32 0.1122 18.23 0.2420
2037 37.52 0.1032 16.89 0.2291
2038 35.73 0.0948 15.62 0.2170
2039 33.94 0.0872 14.40 0.2054
2040 32.14 0.0801 13.24 0.1945
Present Value of Annual ($ million) $901.81 $901.81
Revenue Requirements1
Escalation Rate 3.00% 3.00%
Discount Rate2 8.78% 5.61%
Capital recovery factor (A/P, 8.78%, 30) (A/P, 5.61%, 30)
notation
Capital recovery factor value 0.0954 0.0696
Levelized Annual Charges3 ($ million) 86.05 62.80
Booked Cost ($ million) $604.27 $604.27
Levelized Annual Fixed 14.24% 10.39%
Charge Rates4, 5
Notes:
1. The present value is at the beginning of 2010 and results from the sum of the products of the annual present value factors
times the annual requirements.
2. The discount rate for real-dollar analysis has inflation factored out: 1.0878/1.03-1.
3. The levelized annual charges (end-of-year) result from the present value times the (A/P, i, n) factor.
4. The levelized annual fixed charge rate results from the levelized annual charges divided by the booked cost.
5. The real levelized annual charge is an end-of-year payment expressed in beginning-of-year 2008 real dollars.
2-4
10581090
2.1.2 Example Annual Fixed Charge Rates for a 200-MW Wind Power Plant
To illustrate how plant characteristics affect annual fixed charges, the 1999 TAG report presents
several examples, including transmission; distribution; and renewable, nuclear, combined-cycle
combustion turbine, simple-cycle combustion turbine, and fossil fuel generation. The examples
differ with respect to construction period, tax life, and book life.
Tables 2-3 and 2-4 address the example of a 200-MW wind power plant. The tables present the
capital cost of the plant, the construction period, book life, tax life, and, based on these data, the
annual, cumulative, and levelized fixed charges. The fixed charges in Table 2-3 are expressed in
nominal terms, while those in Table 2-4 are expressed in constant-dollar terms.
2.1.3 Other Capital Expenses—Capital Additions
Capital additions are investments in a plant after it is in service. Because they are treated as
capital investments, the above discussion of capital-related revenue requirements applies to
capital additions as well. The annual revenue requirements associated with specific capital
additions depend on relevant book and tax depreciation schedules. For many analyses, it is
sufficient to treat them as expenses, in which case customers are assumed to pay for them on an
as-you-go basis. For the same level of capital additions, expensing them yields a smaller present
value of revenue requirements than capitalizing them; capitalizing would include a tax
component whereas expensing would not.
2.1.4 Calculating Costs per Kilowatt-Hour
Calculating costs on a per-kilowatt-hour basis provides a relatively easy way to compare new
generating alternatives and their costs with those of one or more existing resources.
2.1.4.1 Levelized Costs per Kilowatt-Hour
Comparing costs on a levelized or a present-value-of-revenue-requirements basis is appropriate

under the following conditions:
• When the benefits or revenues are the same between alternatives. For example, if the
revenues to a utility will be the same for alternatives that are all designed to produce 300
MWnet. Therefore, the most economical plant can be selected by minimizing the present value
of revenue requirements.
• When alternatives have different benefits but the benefits remain the same for each year. For
example, a 300-MWnet plant can be compared with a 200-MWnet plant by comparing ratios of
levelized cost per kilowatt-hour (often referred to as levelized busbar costs).
2-5
10581090
Table 2-3
Wind Power Generation—Nominal Dollar Terms
Instantaneous Construction Cost ($/kW at mid-year of first year of construction)
Total Plant Cost 2,045
AFUDC (interest during construction) 60
Capital Cost ($/kW at commercial operation date)
Total Plant Investment 2,105
Due Diligence, Permitting, Legal, Development 150
Total Booked Cost (total capital requirement) 2,255
Time Frames and Economic Factors
Plant Construction Period (years) 1
Tax Recovery Period (years) 5
Book Life (years) 30
Apparent escalation rate during construction1 3.00
(%/year)
AFUDC Rate, nominal (%/year) 6.00
Discount Rates (%/year)
Nominal 8.78
Real 5.61
Fixed Charges
Cumulative
End of Year Annual Levelized2
Present Value
1 0.1920 0.1765 0.1920
2 0.1788 0.3277 0.1857
3 0.1595 0.4515 0.1777
4 0.1467 0.5563 0.1709
5 0.1378 0.6468 0.1653
6 0.1289 0.7246 0.1605
7 0.1230 0.7929 0.1564
8 0.1201 0.8541 0.1530
9 0.1171 0.9091 0.1503
10 0.1142 0.9583 0.1479
11 0.1113 1.0024 0.1458
12 0.1083 1.0419 0.1439
13 0.1054 1.0772 0.1422
14 0.1024 1.1087 0.1406
15 0.0995 1.1369 0.1392
16 0.0965 1.1620 0.1379
17 0.0936 1.1844 0.1366
18 0.0906 1.2043 0.1355
19 0.0877 1.2220 0.1344
20 0.0847 1.2378 0.1334
21 0.0818 1.2517 0.1325
22 0.0788 1.2641 0.1316
23 0.0759 1.2751 0.1308
24 0.0729 1.2848 0.1300
25 0.0700 1.2933 0.1293
26 0.0671 1.3008 0.1286
27 0.0641 1.3075 0.1280
28 0.0612 1.3133 0.1273
29 0.0582 1.3183 0.1268
30 0.0553 1.3228 0.1262
Note: The Total Plant Cost figure used in the above illustrative example is likely low for end-of-2006 wind energy installation
costs.
2-6
10581090
Table 2-4
Wind Power Generation—Constant Dollar Terms
Instantaneous Construction Cost ($/kW at mid-year of first year of construction)
AFUDC (interest during construction) 60
Capital Cost ($/kW at commercial operation date)
Total Plant Investment 2,105
Total Booked Cost (total capital requirement) 2,255
Time Frames and Economic Factors
Plant Construction Period (years) 1
Tax Recovery Period (years) 5
Book Life (years) 30
1
Apparent escalation rate during construction 3.00
(%/year)
AFUDC Rate, nominal (%/year) 6.00
Discount Rates (%/year)
Nominal 8.78
Real 5.61
Fixed Charges
End of Year 2
Annual Cumulative Present Value Levelized
1 0.1920 0.1765 0.1864
2 0.1736 0.3277 0.1777
3 0.1503 0.4515 0.1677
4 0.1342 0.5563 0.1591
5 0.1224 0.6468 0.1519
6 0.1112 0.7246 0.1455
7 0.1030 0.7929 0.1401
8 0.0976 0.8541 0.1354
9 0.0925 0.9091 0.1314
10 0.0875 0.9583 0.1278
11 0.0828 1.0024 0.1246
12 0.0782 1.0419 0.1216
13 0.0739 1.0772 0.1189
14 0.0697 1.1087 0.1164
15 0.0658 1.1369 0.1141
16 0.0620 1.1620 0.1119
17 0.0583 1.1844 0.1099
18 0.0548 1.2043 0.1080
19 0.0515 1.2220 0.1062
20 0.0483 1.2378 0.1045
21 0.0453 1.2517 0.1029
22 0.0424 1.2641 0.1014
23 0.0396 1.2751 0.1000
24 0.0370 1.2848 0.0987
25 0.0344 1.2933 0.0974
26 0.0320 1.3008 0.0963
27 0.0297 1.3075 0.0951
28 0.0275 1.3133 0.0941
29 0.0254 1.3183 0.0931
30 0.0235 1.3228 0.0921
Notes:
1. The apparent escalation rate represents the total annual rate of change of capital costs during construction. For this
example, the inflation rate equals 3% and the real escalation rate equals 0%.
2. Levelized over the period from the first year through the current year. For example, the levelized fixed charge rate for a
25-year book life is 0.0974; for a 30-year book life, it is 0.0921. Values where the book life is less than the tax life should
be considered as approximations only.
3. Total Plant Cost figure used in above illustrative example is likely low for end-of-2008 wind energy installation costs.
2-7
10581090
Levelized busbar costs are calculated by the following steps:

• Calculate the present value of revenue requirements.
• Calculate the levelized annual revenue requirements by multiplying the present value by the
capital recovery or (A/P,i,n) factor.
• Calculate the levelized busbar cost by dividing the levelized annual cost by the uniform
annual generation in kilowatt-hours.
Frequently, alternatives need to be compared in which the amount of generation changes each
year. Sometimes analysts calculate alternative measures that may present analytical problems,
such as:
• Levelizing the annual cost of generation. The annual cost of generation can be calculated by
dividing the annual revenue requirement by the annual generation. Levelizing these rates
(calculating a present value of the ratios and multiplying the result by an A/P factor) may
lead to mathematical problems because it does not properly account for the amount of
generation in each year.
• Dividing the present value of revenue requirements by the present value of kilowatt-hours.
This can result in erroneous results under certain mathematical conditions.
For the case where annual generation varies, other analytical methods must be used. These
methods include purchasing replacement power so that the generation remains constant over the
years, applying more sophisticated approaches included in production cost models, or
performing a net present value (or profitability) analysis that discretely evaluates the variations
in revenue.
2.2 Non-Utility Generator Power Projects

Electricity from non-utility generators (NUGs) has played an increasingly important role in the
electric utility industry since the passage of the Public Utility Regulatory Policies Act of 1978
(PURPA). As electric utility industry restructuring continues, the role of non-utility generation is
likely to increase. The Energy Policy Act of 2005 modified some of PURPA’s mandatory
purchase requirements that obligate utilities to purchase power from qualifying small power
facilities.
One of the key goals of restructuring was to shift risk from electricity customers to both utility
and non-utility power producers. Under the current regulatory structure, customers bear the
risks associated with and benefit from prudent utility investment decisions. While giving up the
potential for higher-than-expected profits as well as the potential for losses, electric utilities are
guaranteed a return on their prudent investments. With restructuring, the risk associated with the
generation portion of electricity production has shifted somewhat to power producers. The extent
to which risks are shifted depends on the specifics of industry restructuring, which varies among
states and regulatory jurisdictions.
2-8
10581090
Merchant plants are generally characterized as those that have some commodity risk for
electricity sales (i.e., the sale of electricity is not fully committed to long-term power sales
agreements). However, most renewable energy power plants being built remain either utility-
owned or, more commonly, privately-owned, with their output pre-sold to utilities at
contractually agreed-upon rates.
The key differences between utility and non-utility power producers are:
• Obligation to Serve: Electric utilities have traditionally had an obligation to serve and to
provide reliable electricity service. However, the obligation to serve is becoming less clear as
the industry restructures the regulation of generating and distribution companies. Non-utility
power producers develop projects for their potential economic rewards and have the option to
sell their power on a wholesale basis to a utility, on a retail basis to a customer, or directly to
a power pool.
• Rates/Prices: Rates for utilities are usually set using the revenue requirements approach.
Non-utility power producers typically attempt to set prices as high as the market will allow.
• Risks and Benefits: Customers of electric utilities bear the risks associated with prudent
investments. Since customers, not utilities, bear the risk, utilities earn a lower return on
investments associated with a monopoly. Non-utility power producers, or utilities providing
generation in competitive power markets, bear the risks associated with their investments but
can mitigate them to the extent they can negotiate contracts for fuel purchases and energy
sales.
The following subsections focus on the economic evaluation of non-utility or independent power
projects. Subsection 2.2.1 provides a brief discussion of the various types of non-utility power
producer projects.
Subsection 2.2.2 describes the development of an economic pro forma for an “example”
merchant plant.
Subsection 2.2.3 summarizes a sensitivity analysis for the example merchant plant pro forma and
discusses the major risks associated with financing merchant plants.
Subsection 2.2.4 provides a statement of caution on the limitations of the examples that are
discussed.
2.2.1 Types of Non-Utility Generators
An understanding of the economic evaluation methods for non-utility projects requires an

understanding of the potential economic rewards. Non-utility projects can be classified into the
following types:
• PURPA Cogenerators and Small Power Producers
• Wholesale Generators Exempt from the Energy Policy Act of 1992
• Merchant Power Plants
2-9
10581090
2.2.1.1 PURPA Cogenerators and Small Power Producers
Title II of PURPA created a class of power producers called qualifying facilities (QFs) for which
utility ownership is limited to 50%. Qualifying facilities—which can include cogenerators and
small power producers—are exempt from certain state regulations, FERC regulations, and the
recently repealed Public Utility Holding Company Act (PUHCA). However, the Energy Policy
Act of 2005 (EPAct05) amended provisions within PURPA that affect cogeneration and small
power production facilities, including eliminating the ownership limitation on utilities.
Cogenerators produce two forms of energy: electricity and steam (or heat). Two types of
cogenerators have emerged. The first type comprises facilities designed primarily to meet
the thermal requirements of an industrial host, with electricity generation being a secondary
consideration. The second type has become known as “PURPA machines,” which are designed
primarily for the purpose of selling electricity to electric utilities while satisfying the thermal
requirements under Title II of PURPA.
In 2006, FERC, in compliance with EPAct05, issued a decision that did not significantly change
the definition of a qualifying facility. FERC decided not to dramatically tighten standards,
impose new and higher efficiency rules, or restrict access to QF status. Instead, the Commission
stated that combined heat and power (CHP) cogeneration continues to be intrinsically more
efficient than separate heat and power production, that current efficiency standards are
appropriate, that case-by-case evaluation is called for rather than any one-size-fits-all numeric
rule, that smaller cogenerators in particular present no risk of being “sham” projects to foist
overpriced power onto utilities, and that cogenerators should be allowed to continue self-
certifying as QFs. Small power producers continue to be qualifying facilities that use biomass,
waste, geothermal resources or renewable resources such as wind, solar energy, or water. There
are other non-utility generating units in addition to QFs, but they are regulated under the Federal
Power Act.
Power sales for qualifying facilities were typically based on a utility’s avoided cost rates.
Avoided cost rates were generally specified as the cost of a utility’s next plant to be built. By
purchasing power from a non-utility generator, the utility typically paid the non-utility generator
the avoided cost for not building this plant. Consequently, ratepayers would be indifferent as to
who supplied the power. Financing a qualifying facility was relatively straightforward because
the credit worthiness of the project was tied closely with that of the utility (or power purchaser).
Power sales usually consisted of a capacity charge and an energy charge. The capacity charge
frequently comprised a debt service portion that was not subject to escalation and a fixed
operating and maintenance cost that may have been subject to inflation over time. The energy
charge was typically based on the cost of fuel and variable operating costs and was subject to a
negotiated inflation or escalation rate over time.
The economic incentive for qualifying facilities was reduced in the late 1980s and early 1990s
when rates for some qualifying facilities were reduced below the local utility’s avoided cost. In
some states a qualifying facility’s rates had to be low enough to win a competitive bid for a
power solicitation.
2-10
10581090
The economic incentive for QFs was further reduced by competition from exempt wholesale
generators, which resulted from the Energy Policy Act of 1992. Many long-term power purchase
agreements for qualifying facilities have turned out to be significantly more expensive than the
market price of power. There are efforts in some states to restructure or buy out these contracts.
2.2.1.2 Exempt Wholesale Generators
The Energy Policy Act of 1992 (EPAct92) created another group of non-utility facilities called
Exempt Wholesale Generators (EWGs), which was unaffected by EPAct05. The major
characteristics of EWGs include:
• Ownership and operation is for wholesale delivery of electricity.
• Ownership by utilities is allowed to be 100%.
• There is no requirement for a thermal host.
• Wholesale rate approval is required from FERC.
• Exemptions from the Securities and Exchange Commission (SEC) regulations under PUHCA
are allowed.
Other major provisions of the EPAct92 include:

• FERC is empowered to order transmission access to third parties (such as EWGs).
• FERC is prohibited from ordering retail wheeling. However, states may choose to order retail
wheeling, which could further encourage the construction of non-utility generators.
EPAct92 also allowed EWGs to develop large central power stations that have economy-of-scale
advantages over smaller cogeneration plants. In addition, EPAct92 opened the way for states to
consider retail wheeling and the establishment of merchant power plants.
2.2.1.3 Merchant Power Plants
Merchant plants, which have arisen in the United States as electric wholesale competition has
taken hold in many regions, are generally characterized as those that have some commodity risk
for electricity sales (i.e., the sale of electricity is not fully committed to long-term power sales
agreements). The power will either be sold on a spot market basis to the power pool or under
contracts with varying terms to utilities. “Pure” merchant power more precisely refers to power
sales that are not covered by conventional long-term agreements. The term “hybrid” merchant
power refers to plants selling power under a combination of conventional power sales
agreements and spot market sales where the power is at risk for prevailing market conditions. For
the purposes of this report, the term “merchant power plants” will generally refer to plants in
which a “substantial” portion of the power sales (sometimes referred to as the “offtake”) is at
risk.
Merchant power plants include existing, repowered, and greenfield units. Most currently
operating merchant plants are existing plants that have recently been sold by electric utilities.
2-11
10581090
These existing plants are either currently economical or represent a site that could be repowered
to be economical (e.g., by adding a new combustion turbine at an existing site).
Greenfield power plants are those that are built at a new site. A large number of greenfield
merchant power plants are being developed in areas where they will be able to produce power
less expensively than existing older generation or in areas where additional power will be
required to meet growing demand.
The economic analysis in this report centers on greenfield merchant power plants, but is broadly
applicable to repowering projects and to other non-utility power plants.
2.2.2 Development of an Economic Pro Forma for a Merchant Plant
The revenue requirements method (as discussed in Section 2.1) is generally not applicable to the
evaluation of non-utility projects. While there are a variety of methods to evaluate non-utility
power projects, all methods depend on calculating cash flows. The cash flows represent all of the
revenues from the sale of electricity less the sum of all expenses, debt service, and income taxes.
The net cash flows represent cash available to equity holders. The major differences between the
two types of analyses are summarized below and followed by an example pro forma economic
analysis for a non-utility plant.
2.2.2.1 Differences Between Regulated Utility and Non-Utility Power Projects
2.2.2.1.1 Pricing
Regulated utilities are restricted in how they can set prices to recover the costs of investments.
Recovery of the total capital-related costs is determined using the revenue requirements approach
described in Section 5 of the 1999 EPRI TAG report [1]. In particular, the recovery of capital-
related costs depends on return on equity, interest on debt, book depreciation, income taxes, and
property taxes and insurance. But taxes, in turn, depend on capital-related costs, interest on debt,
tax depreciation schedules, and income tax rates. Thus, capital-related revenues are determined
by the underlying costs.
Non-utility power producers have more flexibility. They are not bound to the revenue
requirements approach for determining how their costs will be recovered. Such flexibility is
necessary for non-utilities to respond to changes in market power prices. Whereas under the
revenue requirements approach, capital-related revenues decline over the life of a project, this
need not be the case for non-utility power projects. The prices of power for non-utility projects
can rise, fall, or remain constant over the contract period depending on the needs of the purchaser
and the internal requirements of the non-utility power producer.
2.2.2.1.2 Income Taxes and Depreciation
There are two types of depreciation for regulated utilities: book and tax (see Section 2.1). Under
the revenue requirements method, book depreciation is used to determine revenues, while tax
depreciation is used to calculate income taxes. For non-utility power producers, there is no such
2-12
10581090
distinction. Tax depreciation is calculated to reduce income taxes, which directly affect cash
flow. However, book depreciation (which may be needed for accounting purposes) does not
affect cash flow. Cash flow drives the economic viability of non-utility projects.
2.2.2.1.3 Capital Expenditures
Revenue requirements reflect the cost of constructing a plant as well as the allowance for funds
used during construction (referred to as “interest during construction” for non-utility projects).
For non-utility power producers, only the debt portion of construction expenditures is typically
capitalized. The equity portion of the interest during construction is recovered in the net cash
flows and is not capitalized.
2.2.2.1.4 Debt Financing
Debt is a major source of financing for construction projects. Regulated utilities tend to borrow
against their corporate balance sheet. Corporate stability and borrowing capability become
important issues for utility financing.
Non-utilities finance projects through both corporate debt and project finance (as discussed later
in this section). Corporate debt is becoming more common as the industry consolidates to fewer
and larger companies through mergers and acquisitions. Nevertheless, there is a significant role
for limited recourse project finance.
2.2.2.1.5 Economic Methodology
Regulated utilities typically evaluate potential projects using the present worth of revenue
requirements methodology. Financial structure (debt and equity proportions) is implicitly
included in the analysis by using an after-tax discount rate.
Non-utilities typically evaluate potential projects using a discounted net cash flow analysis. For
project finance opportunities, specific financial parameters are included in the costs of the
project. The after-tax net cash flows are then discounted with the owner’s minimum acceptable
rate of return.
2.2.3 Conceptualizing the Analysis
This section illustrates cash flow calculations using the example of a 200-MW merchant wind
plant. General parameters for the example are described, followed by a discussion of income and
cash flow statements. Economic results for evaluating the cash flows are also discussed.
In this 200-MW wind project example, the Total Plant Cost is assumed to be $2,120/kW. This
represents a significant increase from the $1,000/kW used in the 2004 and earlier editions of the
RETG, and accounts for the recent cost increases for wind turbines and other components caused
by high demand and higher steel prices.
2-13
10581090
A merchant plant needs to sell its power at a price sufficient to cover fuel costs, operating and
maintenance costs, income and property taxes, debt service, working capital, and return on and
recovery of the equity investment. This “price” is usually calculated at the busbar of the plant
(essentially at the plant boundary on the high-voltage side of the transformer) and may be
separated into capacity and energy components. Some regions of the country have a shortfall of
generating capacity and, consequently, have both the capacity and energy components. Other
markets only have an energy component. These payments can be further complicated by various
risk-hedging business deals (from options to tolling agreements, which are not considered in this
report) with power marketers, gas companies, or utilities.
For this 200-MW wind plant example, we have assumed the following parameters and costs:
• Time Frames
– The durations of project development and permitting can vary widely and are not
explicitly included here. Reasonable cost allowances for these activities are discussed
below.
– Duration of engineering, procurement, and construction: 12 months.
– Commercial operation date: January 1, 2011.
– Service life of plant: 20 years.
• Technical and Operating Parameters
– Capability: 200 net MW at rated wind speed.
– Operation: intermittent.
– 200-MW wind plant composed of 100 x 2.0-MW wind turbines.
2.2.3.1 Total Capital Requirement
The major categories of total capital requirement include construction, development, and
financing costs. Table 2-5 displays general categories for the uses of the capital funds, as well as
their sources, as described below. Capital costs can vary significantly as a function of variations
in many items such as equipment type and cost, interconnection, site difficulties, and financing
structure. This example is designed to illustrate a reasonable magnitude of “all-in” costs.
2.2.3.1.1 Construction Cost
The largest component of capital cost is usually engineering, procurement, and construction
(EPC), which are included in the term “total plant cost” (TPC). TPC includes detailed design,
equipment and materials procurement, and installation and construction costs. Typically, an EPC
contractor supplies these services. TPC costs generally range from 60% to 90% of the total
capital requirement depending on development difficulty and financing costs.
2-14
10581090
Other costs that are usually included with the total capital requirement include:
• Land (excluded for wind plant as land is leased).
• Operating and maintenance mobilization (or startup costs): This example assumes one month
of operating labor and 0.5 months of full-capacity fuel consumption as approximate cost
allowances (no fuel is used by wind plant).
• Spare parts. Spare parts costs can range widely depending on the maintenance philosophy
and remoteness for the plant. This example assumes approximately 2% of the EPC cost for
spare parts.
Table 2-5
Sources and Uses of Construction Funds
Commercial Operation:
(Example assumes a 100% construction loan and 60% debt/40% equity January 1, 2011
financing at plant startup)
Total Capital Proportion
Description Units 2008 2009 2010 $1,000 $/kW of Total
SOURCES OF FUNDS
Equity $1,000 0 194,112 194,112 40.0%
Debt $1,000 0 291,168 291,168 60.0%
Total Project $1,000 0 485,280 485,280 100.0%
USES OF FUNDS
Land $1,000 0 0 0 0 0 0.0%
Total plant cost $1,000 0 0 409,000 409,000 2,045 84.3%
Operating and maintenance mobilization $1,000 0 0 137 137 1 0.0%
Spare parts $1,000 0 0 8,180 8,180 41 1.7%
Subtotal Construction Cost $1,000 0 0 417,317 417,317 2,087 86.0%
Owner's Cost $1,000 0 0 2,045 2,045 10 0.4%

Financial fees $1,000 0 0 5,823 5,823 29 1.2%
Interest during construction $1,000 0 0 12,520 12,520 63 2.6%
Reserves $1,000 0 0 20,450 20,450 102 4.2%
Subtotal Owner's and Financing $1,000 0 0 40,838 40,838 204 8.4%
Working capital $1,000 0 0 6,260 6,260 31 1.3%

Contingency $1,000 0 0 20,866 20,866 104 4.3%
Total Capital Requirement $1,000 0 0 485,280 485,280 2,426 100.0%
Notes:
Construction escalation = 0.00% per month, 0.00% per year
Construction interest = 0.50% per month, 6.00% per year
2-15
10581090
2.2.3.1.2 Financing Cost
Merchant plants can be financed either as a limited recourse project or through general corporate
debt. Limited recourse project financing refers to those projects that are dependent on the
specific cash flows of the project as defined in the project agreements. Qualifying facilities and
exempt wholesale generators were commonly financed on a limited recourse project finance
basis. Some merchant plant projects are also being financed on a limited recourse project basis.
However, there is a trend for financing merchant plant projects with corporate debt (as discussed
later in this section of the report). For this example, debt is assumed to be financed with limited
recourse project financing. Assumed financing costs and parameters for a project-financed
transaction include:
• Construction Loan:
– 100% debt: Frequently, construction loans may require equity on a pro rata basis with
the construction costs. However, a 100% debt construction loan is assumed to keep this
example conceptually simple.
– Interest rate during construction of 8.0%.
– Financial fees of approximately 2% of the loan amount.
– Debt service reserve fund of approximately six months of debt service. The lending
institution may require that a debt reserve fund be included in the capital cost of a project.
The fund is set up to pay the debt service for the term loan if unexpected conditions
adversely affect the project revenues.
• Term Loan:
– 60% debt/40% equity
– Commercial bank loan rate of 8.0% for a 10-year term
2.2.3.1.3 Development Cost
Development costs include a variety of costs that a non-utility generating company incurs
to develop a project. Examples include security deposits, permitting (including construction
permits and environmental permits), owner’s engineering and general and administrative costs,
development fees, legal fees, and easements and rights-of-way. These costs can vary widely
depending on the complexity of the site, the financing, and the terms of contracts with the EPC
contractor and equipment suppliers. For this example (Table 2-5), an approximate cost allowance
of 5% of the TPC price was assumed for total development costs.
2.2.3.1.4 Other Costs
A variety of other costs are frequently included in the total capital requirement, including:
• Working capital, which includes requirements for cash and inventories, and for this example
is assumed to equal one month of account receivables.
• Contingency, assumed for this example to be 5% of the total cost.
2-16
10581090
2.2.3.2 EPRI Capital Cost Definitions
Section 4 of the 1999 EPRI TAG report [1] provides capital cost definitions for regulated
utilities. The three measures of capital cost are Total Plant Cost, Total Plant Investment, and
Total Capital Requirement.
The Total Plant Cost (TPC) is the sum of process facilities capital, general facilities capital,
engineering and home office overhead, and project and process contingencies. It is usually
expressed in constant dollars and assumes overnight construction,—i.e., the construction of the
facility occurs instantaneously. When expressed in constant dollars, the total plant cost is also
referred to by several terms: real direct capital cost, overnight capital costs, or overnight
construction costs.
The Total Plant Investment (TPI) adjusts the Total Plant Cost to account for the fact that
construction occurs over a period of years, and hence the costs are distributed over the same
period and tend to escalate over time. In addition, the project allows for interest accrued on each
expense from the date of the expense until the completion and commissioning of the facility. The
accrued interest is called allowance for funds used during construction (AFUDC) and is also
referred to as interest during construction. Total Plant Investment is the sum of the Total Plant
Cost, an adjustment for the escalation of capital costs during construction, and AFUDC. The
1999 EPRI TAG Report [1] provides formulas for calculating escalation and interest during
construction as a function of the escalation rate, cost of money, and construction expenditure
profile.
The Total Capital Requirement (TCR) is the sum of the Total Plant Investment and owner’s
costs such as land and property tax, insurance, pre-production, startup, and inventory costs.
These terms are maintained for consistency in this report. However, a project that is “project
financed” typically organizes its capital costs into construction costs (sometimes referred to as
“hard costs”) and financing and development costs (sometimes referred to as “soft costs”). EPRI
values for project and process contingencies are typically included within the TPC numbers that
are guaranteed by the EPC contractor. An overall project contingency (based on development,
financing, permitting, and the potential for EPC contractor overruns) is normally included by the
banks and is calculated right before the total capital requirement is calculated.
Since the industry convention for summarizing capital costs varies somewhat from a regulated
utility to a market-based plant, the cost components are identified in Table 2-5 for comparison
with other cost breakdowns. For example, EPRI’s value for total plant investment usually
includes the sum of the total plant costs and the interest during construction. TPI is not shown in
the table, but can be easily calculated.
2.2.3.3 Income Statement
The income statement summarizes the revenues and expenses for each year of the project. Table
2-6 presents an example income statement, which is discussed below.
2-17
10581090
Table 2-6
Example Income Statement ($1000)
A B C D E F G H I J K L M N O P Q R S T U V
1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
2 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
3 Revenues
4 Capacity Payments 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
5 Energy Payments 53,425 54,761 56,130 57,533 58,971 60,446 61,957 63,506 65,093 66,721 68,389 70,098 71,851 73,647 75,488 77,375 79,310 81,293 83,325 85,408
6 Federal Production Tax Credit 13,199 13,529 13,867 14,214 14,569 14,934 15,307 15,690 16,082 16,484 0 0 0 0 0 0 0 0 0 0
7 Total Revenues 66,624 68,290 69,997 71,747 73,541 75,379 77,264 79,195 81,175 83,204 68,389 70,098 71,851 73,647 75,488 77,375 79,310 81,293 83,325 85,408
8 Avg. Electricity Rev. (cents/kWh) 8.7 8.9 9.2 9.4 9.6 9.9 10.1 10.4 10.6 10.9 11.2 11.4 11.7 12.0 12.3 12.6 12.9 13.3 13.6 13.9
9
10 Fuel Costs
11 Variable Fuel Costs 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
12 Fixed Fuel Costs 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
13 Total 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
14
15 Variable Operating Expenses
16 Supplies and Consumables 307 314 322 330 338 347 356 364 374 383 392 402 412 423 433 444 455 467 478 490
17 Water & Water Treatment 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
18 Other 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
19 Land Laease 1,603 1,643 1,684 1,726 1,769 1,813 1,859 1,905 1,953 2,002 2,052 2,103 2,156 2,209 2,265 2,321 2,379 2,439 2,500 2,562
20 Total 1,909 1,957 2,006 2,056 2,108 2,160 2,214 2,270 2,326 2,385 2,444 2,505 2,568 2,632 2,698 2,765 2,834 2,905 2,978 3,052
21
22 Fixed Operating Expenses
23 Operating labor 728 746 765 784 804 824 844 865 887 909 932 955 979 1,004 1,029 1,054 1,081 1,108 1,135 1,164
24 General and administrative 146 149 153 157 161 165 169 173 177 182 186 191 196 201 206 211 216 222 227 233
25 Maintenance labor and matl 3,400 3,485 2,858 3,002 3,154 3,314 3,482 3,658 3,843 4,038 4,242 4,457 4,683 4,920 5,169 5,430 5,705 5,994 6,298 6,616
26 Insurance 1,456 1,492 1,530 1,568 1,607 1,647 1,688 1,731 1,774 1,818 1,864 1,910 1,958 2,007 2,057 2,108 2,161 2,215 2,271 2,327
27 Property Taxes 971 995 1,020 1,045 1,071 1,098 1,126 1,154 1,183 1,212 1,242 1,273 1,305 1,338 1,371 1,406 1,441 1,477 1,514 1,552
28 Other 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
29 Total 6,700 6,868 6,325 6,556 6,797 7,048 7,309 7,581 7,864 8,159 8,467 8,787 9,121 9,469 9,832 10,210 10,604 11,016 11,445 11,892
30
31 Total Operating Expense 8,609 8,825 8,331 8,612 8,905 9,208 9,523 9,850 10,190 10,544 10,911 11,292 11,689 12,101 12,529 12,975 13,439 13,921 14,422 14,944
32 Earnings Before Interest, Taxes,
33 Deprec. and Amortization (EBITDA) 58,015 59,465 61,666 63,135 64,636 66,171 67,741 69,345 70,985 72,661 57,478 58,806 60,162 61,546 62,959 64,400 65,871 67,372 68,902 70,464
34
35 Pro Forma Income Tax Calculations
36 Tax Depreciation 81,800 130,880 78,528 47,117 47,117 23,558 0 0 0 0 0 0 0 0 0 0 0 0 0 0
37 Earnings before Interest + Taxes (23,785) (71,415) (16,862) 16,018 17,519 42,613 67,741 69,345 70,985 72,661 57,478 58,806 60,162 61,546 62,959 64,400 65,871 67,372 68,902 70,464
38 Interest Paid 23,293 21,686 19,949 18,073 16,048 13,860 11,498 8,946 6,190 3,214 0 0 0 0 0 0 0 0 0 0
39 Interest Received (at 5.00% per year) 250 250 250 250 250 250 250 250 250 250 0 0 0 0 0 0 0 0 0 0
40
41 Net Operating Loss 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
42 Taxable Earnings (46,829) (92,850) (36,561) (1,806) 1,721 29,002 56,493 60,649 65,044 69,697 57,478 58,806 60,162 61,546 62,959 64,400 65,871 67,372 68,902 70,464
43 State/Local Tax (2,341) (4,643) (1,828) (90) 86 1,450 2,825 3,032 3,252 3,485 2,874 2,940 3,008 3,077 3,148 3,220 3,294 3,369 3,445 3,523
44 Federal Tax (15,571) (30,873) (12,156) (600) 572 9,643 18,784 20,166 21,627 23,174 19,111 19,553 20,004 20,464 20,934 21,413 21,902 22,401 22,910 23,429
45 Total Tax Obligation (17,912) (35,515) (13,984) (691) 658 11,093 21,608 23,198 24,879 26,659 21,985 22,493 23,012 23,541 24,082 24,633 25,196 25,770 26,355 26,952
46
47 Net Earnings after Taxes (28,917) (57,335) (22,576) (1,115) 1,063 17,909 34,884 37,451 40,165 43,038 35,493 36,313 37,150 38,005 38,877 39,767 40,675 41,602 42,547 43,511
2-18
10581090
2.2.3.3.1 Revenues
Forecasting revenues over the service life of a merchant power plant is one of the most critical
aspects of the economic analysis. Revenues for a utility generator are set by the revenue
requirements method and approved by a state commission. However, there is no guarantee of
revenues with a merchant plant because the price of power is set by the electricity markets.
Section 2 of the 1999 EPRI TAG report [1] provides additional information.
A forecast of market prices for a deregulated power market varies with time of day and time of
year, as shown for a historical time period in Figure 2-1. Detailed forecasting of the market price
for power typically requires the following steps:
• Project the cost of fuel for the generating units in the region.
• Evaluate the cost of generation from existing units.
• Project the cost of generation from planned units
• Project the dispatch order of units that will comprise the market.
• Include variations in electricity consumption as a function of time-of-day, time-of-year, and
weather.
• Develop a price duration curve for each year of the service life.
Figure 2-1
Day-Ahead Prices for the California Power Exchange
2-19
10581090
In an annual market price duration curve, data are arranged to show the fraction of a year that the
market price is at or above a given level. An example of a price duration curve for 1998 is shown
in Figure 2-2. For the California Power Exchange, the price was greater than or equal to
approximately $29/MWh for 20% of the year. The integrated price duration curve shows the
cumulative average for any given portion of a year. For example, a market utilization of 20% of
the year would have resulted in an average price of $45/MWh. The average for the entire year
was $24.4/MWh.
This information (appropriately adjusted for transmission costs) can then be used in the
calculation of revenues for the merchant plant.
Figure 2-2
Example Price Duration Curve
For the example (not related to the California Power Exchange data), the revenue assumptions
include:
• For simplicity, only an energy component is assumed for the price of electricity (see line 8 of
Table 2-6). The energy component is based on average power sales and assumed to be
approximately 8.5 cents/kWh in January 2010, escalating at 2.5% per year. For some
projects, there would also be a capacity component (line 7 of Table 2-6) of revenues.
• In addition, it is assumed that the 200-MW wind plant qualifies for the 10-year Federal
Production Tax Credit, which was increased to 2.1 cents/kWh by the Emergency Economic
Stabilization Act of 2008 and is assumed to escalate at 2.5%/yr for 10 years.
• Average annual capacity factor: 35%.
2-20
10581090
2.2.3.3.2 Operating and Maintenance Expenses
Operating and maintenance costs need to be projected for each year of the service life. For this
example, the operating cost assumptions (included in lines 13 through 32 of Table 2-6) include:
• Average annual capacity factor: 35%
• Average annual net plant heat rate (lower heating value): N/A for wind plant
• Delivered price of natural gas: N/A for wind plant
• Variable operating costs: 0.05 mills/kWh (at Jan. 1, 2011)
• Annual land lease cost: 3% of electricity sale revenue
• Labor: five employees at an average fully burdened wage rate of $35/hour (at Jan. 1, 2011)
• General and administrative expenses: 20% of operating labor
• Maintenance labor and material: $17/kW-yr (at Jan. 1, 2011)
• Insurance: 0.5% of the total capital requirement
• Property taxes: 1.25% of the total capital requirement
• All costs escalate at 2.5% per year (for simplicity)
• Income statements often include a book depreciation component for the calculation of book
income. However, this income statement calculates earnings before interest, taxes,
depreciation and amortization (BETIDE) to focus on the calculation of net cash flows (see
the next subsection).
2.2.3.4 Income Taxes
Income taxes are included on a conceptual basis. For the example, key assumptions include:
• All of the depreciable capital investment qualifies for 20-year MACRS tax depreciation (line
36, Table 2-6) (for simplicity)
• Interest paid on the debt service is tax deductible (Line 38, Table 2-6)
• Interest received from the debt service reserve increases the taxable earnings (line 39, Table
2-6)
• A state income tax rate of 5% (line 43, Table 2-6)
• Marginal federal income tax rate of 35% (line 44, Table 2-6)
• This example assumes that the project structure includes a single purpose corporation. It is
important to note that some projects are structured as partnerships. Although partnerships are
not taxed directly under federal and most state laws, a tax computation is needed so that
investors can compute their individual cash return from the project and also to compare it
with other opportunities. In this case, the tax information needed by the partners is computed,
but tax payments are excluded from the project’s cash flow forecast.
2-21
10581090
2.2.3.5 Cash Flow Statement
The cash flow statement calculates the after-tax net cash flow for the project. Table 2-7 presents
an example cash flow statement.
The cash flow statement begins with the EBITDA (line 5 in Table 2-7) as brought forward from
line 33 of Table 2-6 and includes the following adjustments (lines 9-25, Table 2-7):
• Less income taxes
• Less debt service (principal and interest payments for the loan)
• Plus interest received from the debt reserve fund
• Less any new contributions to reserves (sometimes additional reserves may be required)
• Plus return of the reserves at the end of the debt service term (the reserve fund is returned if it
has never been drawn down)
• Less any adjustments to working capital (growth in working capital requirements)
• Less equity investment during construction
The net cash flows represent the cash flows associated with the equity investor (owner of the
project). During the construction period, the cash flows are negative. Throughout the service life
of the plant, the net cash flows are initially negative and become positive in the out years.
The net present value (line 31, Table 2-7) and the internal rate of return (“IRR” on line 32, Table
2-7) are economic measures of the project that reflect the present worth of profit over the service
life and the profitability of the project, respectively. The cumulative IRR reflects the IRR for the
relevant service life. For example, the IRR for a 20-year project life is projected to be 12.8%,
while the IRR for the first 19 years of net cash flows is projected to be 12.2%. The meaning of
these measures is discussed in the following section.
Figure 2-3 shows the variation of the components of busbar electricity sales price vs. year. The
fixed costs represent income taxes, debt service, and working capital. The fixed costs decline
when the debt service for the loan ends (a 10-year loan in this example). The O&M component
represents the fixed and variable operating and maintenance component. The net cash flow
represents the capital recovery for the equity investor.
2-22
10581090
Table 2-7
Example Cash Flow Statement ($1000)
A B C D E F G H I J K L M N O P Q R S T U V W X Y
1 Commercial Operation: January 1, 2010
2 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
3 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
4
5 EBITDA 58,015 59,465 61,666 63,135 64,636 66,171 67,741 69,345 70,985 72,661 57,478 58,806 60,162 61,546 62,959 64,400 65,871 67,372 68,902 70,464
6 0
7 Taxes Paid (17,912) (35,515) (13,984) (691) 658 11,093 21,608 23,198 24,879 26,659 21,985 22,493 23,012 23,541 24,082 24,633 25,196 25,770 26,355 26,952
8
9 Cash Flow From Operations 0 75,927 94,980 75,651 63,825 63,978 55,078 46,132 46,147 46,105 46,002 35,493 36,313 37,150 38,005 38,877 39,767 40,675 41,602 42,547 43,511
10
11 Debt Service 43,393 43,393 43,393 43,393 43,393 43,393 43,393 43,393 43,393 43,393 0 0 0 0 0 0 0 0 0 0
12 Interest Received 1,023 1,023 1,023 1,023 1,023 1,023 1,023 1,023 1,023 1,023 0 0 0 0 0 0 0 0 0 0
13 Contribution to Reseserves 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
14 Disbursement (return) of Reserves 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
15
16 Additions to Working Capital
17 Accounts Receivable 0 182 187 191 196 201 206 211 216 222 (1,082) 200 205 210 216 221 227 232 238 (9,763)
18 Fuel Stocks 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
19 Spare Parts 0 205 210 215 220 226 231 237 243 249 255 262 268 275 282 289 296 304 311 (12,758)
20
21 Capitalized Refurbishments
22
23 Contributed Capital 0 0 194,112
24
25 Net Cash Flow Before Tax 0 0 (194,112) 15,645 16,709 18,900 20,358 21,850 23,374 24,933 26,526 28,155 29,820 58,304 58,344 59,688 61,061 62,461 63,890 65,348 66,836 68,353 92,985
26 Cumulative 0 0 (194,112) (178,467) (161,759) (142,859) (122,501) (100,651) (77,276) (52,343) (25,817) 2,338 32,158 90,462 148,806 208,494 269,555 332,016 395,906 461,255 528,090 596,443 689,428
27
28 Net Cash Flow After Tax 0 0 (194,112) 33,557 52,224 32,884 21,049 21,191 12,281 3,325 3,328 3,276 3,161 36,319 35,851 36,676 37,519 38,379 39,257 40,153 41,066 41,998 66,032
29 Cumulative 0 0 (194,112) (160,555) (108,332) (75,447) (54,398) (33,207) (20,926) (17,601) (14,273) (10,998) (7,837) 28,482 64,333 101,009 138,528 176,908 216,165 256,317 297,384 339,381 405,414
30
31 Net Present Value (at time = 0) 66,512
32 Cumulative IRR on Net Cash Flow After Tax -38.8% -21.2% -13.0% -6.7% -3.8% -3.1% -2.4% -1.8% -1.2% 3.2% 5.7% 7.5% 8.8% 9.8% 10.6% 11.3% 11.8% 12.2% 12.8%
33
34 Coverage Ratios (before tax) Minimum Average
35 Total Debt Service 1.34 1.51 1.34 1.37 1.42 1.45 1.49 1.52 1.56 1.60 1.64 1.67 na na na na na na na na na na
36 Debt Service Including Reserves 1.67 1.93 1.81 1.84 1.89 1.93 1.96 2.00 2.03 2.07 2.11 1.67 na na na na na na na na na na
37
38 Reserves
39 Contribution to Reserves 0 20,450 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
40 Reserve Balance 0 20,450 20,450 20,450 20,450 20,450 20,450 20,450 20,450 20,450 0 0 0 0 0 0 0 0 0 0 0
41 Disbursement of Reserves 0 0 0 0 0 0 0 0 0 20,450 0 0 0 0 0 0 0 0 0 0
42 Operating Income with Reserves 78,465 79,915 82,116 83,585 85,086 86,621 88,191 89,795 91,435 72,661 0 0 0 0 0 0 0 0 0 0
43
44 Working Capital Balances
45 Accounts Receivable 7,282 7,464 7,651 7,842 8,038 8,239 8,445 8,656 8,873 9,095 8,013 8,213 8,419 8,629 8,845 9,066 9,293 9,525 9,763 10,007
46 Fuel Stocks 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
47 Spare Parts 8,180 8,385 8,594 8,809 9,029 9,255 9,486 9,723 9,967 10,216 10,471 10,733 11,001 11,276 11,558 11,847 12,143 12,447 12,758 13,077
2-23
10581090
200
150
Busbar Cost ($/MWh, end-of-year $)
100
Federal PTC
Net Cash Flow
Fixed Costs
50
O&M
Land Lease
Fuel
0
(50)
(100)
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Figure 2-3
Components of Busbar Cost for Example 200-MW Wind Generation Plant
2.2.3.6 Economic Measures
Several common measures address the economic viability of a project. The busbar cost includes
capital, O&M, fuel, and other costs of generating electricity. The profitability of the net cash
flows is typically addressed by calculating the net present value (NPV) and the internal rate of
return (IRR). Other common measures that address the riskiness of a project include the payback
period and debt coverage ratios.
2.2.3.6.1 Busbar Cost
The busbar cost is the unit cost of generating one unit of electricity (e.g., 1 MWh), and is equal
to the total annual costs minus credits received, divided by the total annual electricity generation.
For the 200-MW wind plant example, the cost components include fixed and variable O&M cost,
other fixed costs including income taxes and debt service, and other net cash flow. The credit is
the 10-year federal Production Tax Credit (PTC), which is assumed to be 2.1 cents/kWh in 2009,
escalates at 2.5%/yr for 10 years, and is zero after 10 years. Figure 2-3 illustrates how the various
cost components and the PTC vary over the 20-year operating period of the wind plant. The 10-
year PTC is shown as a negative contribution to the busbar cost during the first 10 years of
operation, after which it is eliminated. The fixed cost component decreases after 10 years
because the 10-year loan has been retired and the debt service component of fixed cost is
eliminated. The average annual busbar cost increases from 8.7 cents/kWh in 2009 to
13.9 cents/kWh in 2028.
2-24
10581090
2.2.3.6.2 Net Present Value (NPV)
The net present value represents the present value (or present worth) of profit using the time
value of money principles described in Section 3 of the 1999 EPRI TAG report [1]. This
calculation results from discounting the net cash flows at the minimum acceptable rate of return
for an equity investor. The method is also referred to as the discounted cash flow method.
The net present value must be defined at a certain point in time. Frequently, the NPV is
calculated at the commercial operation date. In this case, the total capital requirement (at the
commercial operation date) is subtracted from the net cash flows that are discounted or brought
back to the same date. For the example provided, the total capital requirement is considered at
the commercial operation date, since the construction loan is based on 100% of the capital
investment and the interest during construction essentially includes the future worth of the
construction expenditures at the commercial operation date.
The NPV may also be calculated at the beginning of the construction time period if equity is
required to fund construction (as opposed to a 100% construction loan). In any event, the NPV
can be calculated at the point in time that makes the most sense for the evaluation.
In addition to determining the present value of cash flows for potential projects, the discounted
cash flow method can be used in other ways as well. Up to this point, revenues from the project
were assumed to be known. With an assumed rate of return, the discounted cash flow approach
can then be used to compute the annual revenues, or rate structure, necessary to make the project
a viable one. Essentially, this is what is done in the revenue requirements method.
Another application would be to use the discounted cash flow method to negotiate with a
potential project constructor. If all the other variables are known (or estimated), the discounted
cash flow approach could be used to compute the initial cash outlay that would make the project
feasible. Alternatively, the discounted cash flow method can be used to project values for
liquidated damages if a contractor fails to meet specific targets on plant performance or capacity.
Other types of studies can include an assessment of capital/energy tradeoffs, such as an
evaluation of the potential benefits from spending capital dollars to reduce heat rate. In many of
the previously discussed cases, solutions are found using trial-and-error algorithms, and
spreadsheets can find solutions for even complex problems very quickly.
Finally, a contract may cover a period that does not extend over the entire economic life of a
project. There is, however, value to the project for the period beyond the life of the contract,
which is referred to as its terminal value. Additionally, there may be a salvage value or a
dismantling and decommissioning cost at the end of the service life. Analysts often omit these
values or costs because they are subject to uncertainty since they occur far in the future, have
relatively little impact on the results, are hard to estimate, and make the analysis slightly more
conservative. However, there may be some cases in which it is appropriate to include the
terminal value in calculating discounted cash flows.
2-25
10581090
2.2.3.6.3 Internal Rate of Return (IRR)
The internal rate of return addresses the profitability of a project. Mathematically, the IRR is
defined as the discount rate that sets the present worth of the net cash flows over the service life
equal to the equity investment at the commercial operating date.
An IRR of 20% does not necessarily mean that net cash flows will represent 20% of the equity
investment for each and every year of service life. However, an IRR of 20% does mean that the
equity investor will earn the equivalent of 20% of the outstanding equity balance each year. That
balance will be reduced in some fashion over the life of the plant.
Many companies have a minimum acceptable rate of return that must be met before a potential
project is seriously considered. This minimum acceptable return is known as the hurdle rate,
which can be used to screen potential projects based on their internal rates of return. In the
example above, the project would be viable if its 20% internal rate of return is greater than the
hurdle rate.
There are several caveats to be aware of when calculating the IRR:

• The IRR solution is a trial-and-error solution that is typically solved by a convergence
routine available in spreadsheet software.
•
th
The solution is based on solving an n -degree polynomial that may have multiple real,
positive roots. More than one change in the sign of the coefficients of the net cash flows is an
indication of possible multiple roots. The net cash flow line in the example in Table 2-7 has
only one real positive root due to only one change in sign of the coefficients (negative net
cash flow at the commercial operating date changing to positive net cash flows for the
service life). A standard engineering economics textbook should be consulted for situations
where multiple roots are suspected.
• Changes in the IRR are not scalar, and a small change in the cash flows can have a large
effect on the IRR.
• Comparing IRR values may be misleading. While the internal rate of return allows investors
to rank options based on their potential rate of return, it does not take into account a project’s
size. For example, it does not allow an analyst to compare a $1 million project with a 25%
internal rate of return with a $10 million alternative having a 20% internal rate of return. An
incremental analysis may be required. Again, a standard engineering economics textbook
should be consulted for these situations.
2.2.3.6.4 Payback Period
The payback period represents how long it will take the net cash flows to recover the capital
investment (i.e., how many years for the cumulative net cash flows to become positive). There
are two versions of the payback period measure:
• The simple payback period typically represents the number of years for the net cash flows to
pay back the capital investment. The calculation requires summing the net cash flows for the
preceding years. The time value of money is not considered in the calculation.
2-26
10581090
• The discounted payback period represents the number of years for the present worth of net
cash flows to recover the capital investment. The time value of money is considered in the
calculation.
The discounted payback period is the preferred measure since it provides a proper economic
signal for the importance of the net cash flows. The simple payback period can be misleading
since cash flows in future years have the same apparent value as cash flows in the current year.
The payback period is a measure of secondary economic importance since it does not consider
all of the net cash flows over the life cycle of a project. Significant cash flows after the payback
period are not included in the measure number. Nevertheless, the payback period remains a
popular indicator of the risk level of the project.
2.2.3.6.5 Debt Coverage Ratios
For projects that are debt financed, debt coverage ratios address the level of risk for the project
from the perspective of the lending institution. The debt coverage ratio is calculated as the
earnings before interest, taxes, depreciation, and amortization (EBITDA), divided by the annual
debt service. EBITDA represents the revenues minus expenses (not including book depreciation)
on a before-income-tax basis. The debt service represents the principal and interest payments to
service the loan. The ratio is designed to address the ability of the project’s cash flows to service
the debt; the higher the ratio, the lower the level of risk for the project.
The minimum debt coverage ratio should always be greater than 1.0. Depending on the risk of
the project, banks may require that the minimum never go below a certain threshold such as 1.3
or 1.5.
Lending institutions also calculate the average debt coverage ratio over the life of the loan. This
measure reflects the ability to service the loan over its term. Depending on the level of risk, the
lending institution may require that the average over the life of the loan never drop below a
certain threshold such as 1.7 or 2.0.
2.2.3.7 Sensitivity Analysis
Sensitivity analysis is important for understanding the key drivers for the economic pro forma.
An economic analysis typically begins with specifying a base case as defined in the previous
example. Sensitivity analysis is based on re-evaluating the economic analysis by changing input
parameters by reasonable amounts. The analyst can then study the results to determine the key
drivers in the pro forma model. An analyst can also determine what reasonable variations in
input parameters may cause the project to miss its financial goals.
Table 2-8 presents a sensitivity analysis for the previous example. The effects of reasonable
variations in key input parameters on key economic results are shown. For example, a 20%
decrease in the total plant cost (TPC) would cause the IRR to increase from 12.8% for the base
case to 18.8% (the pro forma assumes that certain development costs are based on the TPC
cost—these would decrease accordingly). Debt coverage ratios and payback periods would
improve accordingly.
2-27
10581090
Table 2-8
Example Sensitivity Analysis
Simple Discounted
Net Present Value2 Senior Debt Payback Payback
Input Data After Tax $ million (Dec 2008) Coverage Ratios3 Period Period
1
Parameter % Change Value Case IRR Result % vs. Base Minimum Average (years) (years)
CAPITAL COST $ thousand, December 2009

Total Plant Cost 20% 490,800 2 8.6% 1,917 4% 1.10 1.25 13 20
Base 409,000 1 12.8% 66,512 122% 1.34 1.51 11 14
-20% 327,200 3 18.8% 131,108 240% 1.69 1.90 5 8
FINANCING
Debt Proportion 40% 4 11.1% 53,097 97% 2.01 2.27 9 15
Base 60% 1 12.8% 66,512 122% 1.34 1.51 11 14
80% 5 16.8% 80,037 146% 1.00 1.13 12 13
Loan Term Years

8 6 9.3% 12,690 23% 1.15 1.29 13 20
Base 10 1 12.8% 66,512 122% 1.34 1.51 11 14
12 7 15.4% 103,401 189% 1.50 1.69 6 11
Interest Rate Rate

6.0% 8 13.8% 81,166 148% 1.47 1.65 8 13
Base 8.0% 1 12.8% 66,512 122% 1.34 1.51 11 14
10.0% 9 11.7% 51,172 94% 1.22 1.38 11 16
REVENUES
cents/kWh, January 1, 2010
Power Revenue 20% 10.2 10 17.1% 138,989 254% 1.58 1.77 5 11
Base 8.5 1 12.8% 66,512 122% 1.34 1.51 11 14
-20% 6.8 11 8.1% (5,964) -11% 1.10 1.24 13 >20
Capacity Factor 40% 12 16.5% 126,394 231% 1.55 1.75 6 11

Base 35% 1 12.8% 66,512 122% 1.34 1.51 11 14
30% 13 8.9% 6,631 12% 1.12 1.27 13 20
Federal Production cents/kWh, January 1, 2010

Tax Credit 3.0 14 14.6% 91,769 168% 1.47 1.65 6 12
Base 2.1 1 11.9% 54,667 100% 1.29 1.45 11 15
0.0 15 8.9% 7,580 14% 1.03 1.17 13 20
Notes
1. Internal rate of return on discounted net cash flow after tax.
2. Discount rate is 8.5%.
3. Senior debt coverage ratios are based on operating income (before tax, not including reserves) divided by debt service.
As can be seen from the table, the key drivers in the analysis are the power sales revenue,
capacity factor, federal Production Tax Credit, and total plant cost. A 20% decrease in electricity
revenues would cause the IRR to decrease from 12.8% to 8.1%, and the debt coverage ratios to
be low and possibly unacceptable depending on the goals of the project.
The sensitivity analysis provides a basis for beginning the probabilistic risk analysis that is
described in Sections 8 and 9 of the 1999 EPRI TAG report [1].
2-28
10581090
2.2.3.8 Project Risks and Financing
Non-utility generating plants can be financed as limited recourse projects or through general
corporate debt.
Limited recourse project financing refers to those projects that are dependent on the specific cash
flows of the project as defined in specific project agreements. Limited recourse financing usually
requires the formation of a special-purpose company to protect the exposure of the parent
company. The financial exposure of the owner is typically limited to the equity invested in the
physical and intangible assets of the project. However, the interest of the owner in the assets is
junior to that of the lenders. This type of financing requires acceptable risk allocation through
proper contract structuring with financially strong contractual parties, performance guarantees,
and liquidated damages (intended to “backstop” guarantees provided by the parties to the various
project contracts and agreements). The backstop for the guarantees can be assigned to the
appropriate original equipment manufacturers.
Qualifying facilities (under PURPA) were commonly financed on a limited recourse project
finance basis because these projects had a revenue stream that was secured by long-term power
sales agreements. The ultimate risk of power sales became a function of the creditworthiness of
the utility that had contracted to buy the power. Most utilities have been creditworthy because of
their regulated monopoly status.
The regulatory restructuring of the industry has allowed the development of merchant power
plants that sell power on a commodity basis (as previously shown in Figure 2-1). Limited
recourse financing for a merchant plant requires more stringent assurances that the project can
meet its goals. The amount and type of limited recourse debt financing will be a function of:
• Properly structured project ownership. An appropriate legal structure and experienced and
well-managed sponsors are fundamental requirements. Experienced contractors are required
for engineering, procurement and construction.
• The proportion of the plant’s output that will be under contract. It is expected that this
percentage will vary widely between developers with different risk strategies. A sound
security package in terms of the project agreements will secure the appropriate cash flows
with a low degree of volatility.
• The competitiveness of a project in its region. The project will have to be one of the lowest-
cost producers. However, being one of the lowest-cost producers in a commodity-driven
market has several implications:
– Maximizing the capacity factor will become relatively less important than maximizing
the profitability of the plant production.
– The economics of a project may be driven by capacity payments in some areas of the
country, while in other regions they will be driven by energy payments.
– Different plant types will occupy a niche based on the time of day or season of the year.
For example, some natural gas combined-cycle units may be most profitable in the
“shoulder” or intermediate peaking times of day. Simple-cycle natural-gas-fired plants
and pumped storage plants are typically most economical for the time of day and season
2-29
10581090
of the year when power is at peak demand. The economic viability of run-of-river hydro
plants will depend on the hydrologically wet seasons.
• The capabilities for understanding and forecasting market pricing.
• The ability of the project cash flows to cover the risks of power sales. Many projects will
have structural enhancements that take the following forms:
– Methods to shift risk, such as subordinating fuel expenses to debt service payments, or to
utilize a tolling agreement for fuel supply and power output
– Methods to share risk, such as higher equity investments, limited parent guarantees, or
subordinated debt provided by vendors and/or EPC contractors
– Methods to account for the timing of risks, such as cash sweeps, reserve accounts and
commodity swaps, and remarketing of unused fuel.
Achieving an investment-grade rating for limited recourse project financing is difficult when
power sales are subject to “commodity” risk. The volatility of commodity prices leads to risk that
the plant may not be able to meet its financial goals. Consequently, lenders will require that
projects have a greater equity proportion of financing to absorb the volatility risk (i.e., the risk
that the cash flows will be insufficient to pay the debt service).
A typical merchant plant project that is project financed can be characterized as follows:
• Equity proportion of the capital structure of 30% to 50% depending on the risk of the project.
• Average annual debt coverage ratios of at least 1.75 to 2.0 (probably closer to 2.0).
• Debt terms of 10 to 20 years with cash sweep provisions.
• Relatively higher debt reserves than were required for qualifying facilities because of the
seasonality of electricity pricing. (Debt reserves of 6 to 12 months of debt service payments
may be required, depending on the level of risk. Debt service reserves can take the form of
cash that is financed, letters of credit, or other financial commitments.)
• Other financial structural enhancements such as subordinated debt, limited parent guarantees
and commodity swaps.
• A well-defined legal structure.
• An internal rate of return on after tax equity that is typically at least 15%.
As more equity is required for limited recourse project financing, the financing complexity and
costs increase. Essentially, a premium is paid for limited recourse debt even though the higher
levels of equity are designed to absorb the risk of volatility in the cash flows. Consequently,
large firms with healthy balance sheets and diversified assets may have an advantage in
financing merchant power plants through general corporate debt. Corporate debt allows a
company to avoid burdensome contractual arrangements and project security arrangements that
are required for project financing and, in some cases, allows the company to tolerate more
commodity-based risk.
2-30
10581090
The advantage of financing with corporate debt is one of the factors driving the industry to
consolidate into larger companies. Large balance sheets (sometimes formed as a result of
mergers and acquisitions) allow for financing through corporate debt. Large companies with
diversified portfolios of power plant assets can also balance downside risks that occur in one
regional market with upside risks that occur in another. Financing with corporate debt may
provide funding that is more flexible, has shorter lead-time, and often has reduced costs.
In conclusion, the financing of merchant power plants is evolving along several parallel paths. A
framework for limited recourse project financing is developing in the financial community that
requires significant proportions of equity investment. Additionally, there is an increasing trend
for financing with corporate debt as the industry consolidates to larger firms.
2.2.4 Limitations of Examples
This section has outlined methods for evaluating power projects developed by non-utility
producers and provided comparisons to evaluations for utility projects. Many simplifying
assumptions have been applied that may not be appropriate for an actual project. The appropriate
method for any entity to use in making economic decisions depends on many factors including
its strengths and weaknesses, its strategic aim, its competitors, and the economic and legal
environment. In addition, as competition increases, industry participants will assess their
missions and their risk profiles, and the distinctions between regulated and non-utility entities
can be expected to blur.
2.3 Guidelines for Economic Evaluation of Renewable Energy Projects
2.3.1 Design/Cost Estimate
The rating system shown in Table 2-9 indicates the level of effort involved in the design and
cost estimate.
Table 2-9
Confidence Rating Based on Cost and Design Estimate
Letter Rating1 Key Word Description

A Actual Data on detailed process and mechanical designs
Historical data from existing units
B Detailed Detailed process design
(Class III design and cost estimate)
C Preliminary Preliminary process design
(Class II design and cost estimate)
D Simplified Simplified process design
(Class I design and cost estimate)
E Goal Technical design/cost goal
(or value developed from literature data)
1. If complete design and cost-estimating methods and assumptions are not available, the rating is reduced by one unit.
2-31
10581090
2.3.2 Accuracy Ranges
The accuracy of cost estimates has been discussed in detail in many texts and papers on the
subject. Table 2-10 presents estimates of the range of accuracy for the cost data presented in this
report.
Table 2-10
Accuracy Range Estimates for RETG Cost Data (Ranges in Percent)1
Design & Cost

A B C D E&F
Estimate
Mature Commercial Demo Pilot Lab & Idea
Rating
A. Actual 0 – – – –
B. Detailed -5 to + 5 -10 to + 10 -15 to + 20 – –
C. Preliminary -10 to + 10 -15 to +15 -20 to +20 -25 to +30 -30 to +60
D. Simplified -15 to + 15 -20 to +20 -25 to +30 -30 to +30 -30 to +80
E. Goal – -30 to + 70 -30 to +80 -30 to +100 -30 to +200

1. This table indicates the overall accuracy of cost estimates. Accuracy is a function of the level of cost-estimating
effort and the degree of technical development of the technology. The same ranges apply to O&M costs.
2.3.3 Definitions of Economic Terms
The following briefly defines various economic terms used throughout this RETG report:
Plant Size: Net electrical capacity of the unit that forms the basis for the cost estimates.
Available for Commercial Orders: Estimate of the earliest date that a commercial-size unit could
be ordered with normal performance guarantees, if the technology is not already commercially
available. This date assumes an orderly development program, with success at each step. Since
most technology development programs have unexpected delays and unsuccessful steps that
must be repeated, the actual commercial order date is likely to be later than the one shown.
Furthermore, the very first commercial units will be somewhat more expensive (in constant
dollars) than the mature unit designs that form the basis for the cost estimate.
First Commercial Service: Estimate of the earliest date that a commercial-size unit could be put
into operation. The first commercial service date estimate is usually based on adding the
idealized construction time plus an allowance for design time to the “available for commercial
orders” date.
Hypothetical In-Service Year: Year in which the plant is assumed to be placed into service. In
most cases this is the same as the “first commercial service.”
Plant Capital Cost: Basis for the cost estimate in terms of the number of units included in the
plant.
2-32
10581090
General Facilities and Engineering Fee: General facilities capital is the total construction cost of
the general facilities, including roads, office buildings, shops, laboratories, etc. The engineering
fee is the engineering and home office overhead and fee paid to the architect/engineering
company, and is considered representative of the type of generation or storage unit.
Project and Process Contingency: Project contingency is a capital cost contingency factor
covering the cost of additional equipment or other costs that are not apparent in the preliminary
design and that would result from a more detailed design of a definitive project. Process
contingency is a capital cost contingency factor applied to a new technology in an effort to
quantify the uncertainty in the technical performance and cost of the commercial-scale
equipment.
Total Plant Cost (TPC): Total of all direct and indirect construction costs at a single point in
time, excluding the time value of money (interest, etc.) during the construction period. Capital
cost estimates expressed at a single point in time are often referred to as instantaneous or
overnight construction costs.
Total Cash Expended (TCE): Amount of money that would have been spent for all labor,
materials, and indirect costs at the time the unit goes into service. TCE is expressed in mixed-
year dollars accumulated over the construction period.
AFUDC (Allowance for Funds Used During Construction): Interest paid on the cash expended
during the construction period. This interest is accumulated and compounded until the plant goes
into service.
Total Plant Investment (TPI): Sum of TCE and AFUDC. TPI is expressed in mixed-year dollars
accumulated at the same point in time as total plant cost. For cases where the expenditure rate is
uniform, total plant investment (TPI), total cash expended (TCE), and allowance for funds used
during construction (AFUDC) can be calculated by multiplying total plant cost (TPC) by the
appropriate adjustment factor (AF) shown in Table 2-11. That is:
TPI = TPC x AF (TPI)

TCE = TPC x AF (TCE)
AFUDC = TPC x AF (AFUDC)
2-33
10581090
Table 2-11
TPC-TPI Adjustment Factors (AF)
Idealized Adjustment Factors1

Construction
Time (Years) AF (TPI) AF (TCE) AF (AFUDC)
1 1.0 1.0 0
2 1.025 0.980 0.045
3 1.050 0.961 0.089
4 1.076 0.942 0.134
5 1.103 0.924 0.179
6 1.131 0.906 0.225
7 1.160 0.889 0.271
8 1.189 0.872 0.317
1. Based on 4.1% cost escalation and 9.2% AFUDC interest rate.
Owner Costs: This includes prepaid royalties, startup costs, inventory capital, catalyst and
chemicals, and land.
Total Capital Requirement: All capital costs that would go into the rate base on a hypothetical
in-service date. Since the technology may not be available until a much later point in time, the
total capital requirement should be multiplied by the appropriate cost escalation factor for a
future in-service date.
Total Capital Replacement: Capital cost that would be incurred by employing a particular
technology; the total capital cost over the lifetime of the unit. This represents major replacement
or refurbishing of equipment such as replacement of blades in a combustion turbine.
Operation and Maintenance (O&M) Costs: Costs are expressed at the same point in time as the
total capital requirement and represent the costs for a normal year of operation. The extra costs
associated with plant start-up in the first year are included in the total capital requirement.
Incremental costs are broken down into variable and consumable components. Consumables are
computed directly from the plant material balance. The variable component primarily represents
variable maintenance and is estimated from an algorithm incorporating the units expected
capacity factor.
Unit Availability: Fraction of time a generating unit is able to supply power at various capacity
levels. Estimates are based on mature plants and are likely to be significantly more optimistic
than the initial experience with first-of-a-kind plants.
Minimum Load: Estimate of the minimum load from which the unit could be ramped up to full
load in a reasonable time (one hour or less).
Pre-Construction, Licensing, and Design Time: Time periods for pre-construction activities
depend on local regulations and conditions. The time period is intended to be a reasonable
approximation, but actual times may range from one-half to two times the estimated time.
2-34
10581090
Idealized Plant Construction Time: The idealized construction time is intended to represent the
minimum construction time required to build a plant under ideal conditions. Actual construction
times will likely be somewhat longer due to weather, regulatory problems, labor strikes, etc.
Unit Life: Estimate of the book life of the plant. Maintenance costs include sufficient funds to
replace minor equipment that wears out before the unit life shown.
Technology Development Rating and Design/Cost Estimate Rating: Rating systems designed
to indicate the quality of the information used to develop the data.
2.4 References
1. TAG® Technical Assessment Guide, Volume 3, Rev. 8: Fundamentals and Methods-
Electricity Supply. EPRI, Palo Alto, CA: November 1999. TR-100281-V3R8.
2. Wind Power Technology Status and Performance and Cost Estimates - 2009. EPRI, Palo
Alto, CA: 2009. 1020362.
3. Technical Assessment Guide, Electricity Supply-1993, Volume 1, Rev. 7. EPRI, Palo Alto,
CA: June 1993. TR-102275-V1R7.
4. U.S. DOE Energy Information Administration, Cost and Quality of Fuels for Electric Utility
Plants 1994, July 1995, pp. 15, 27.
5. U.S. DOE Energy Information Administration, Annual Energy Outlook 1996, January 1996,
p. 79.
6. U.S. DOE Energy Information Administration website, Status of State Electric Industry
Restructuring Activity, http://www.eia.doe.gov/cneaf/electricity/chg_str/regmap.html, July
2003.
2-35
10581090
10581090
3
WIND POWER
Installed Capacity • 180,000 MW estimated worldwide.

(September 31, 2010)
• 36,698 MW MW1 in the United States
Technology Readiness • Mature, commercial.
• High confidence in cost estimates and projections.
Environmental Impact • Minimal during manufacturing and operation.
• New technologies minimize avian mortality.
• Sight and sound impacts must be considered in planning and
operation.
Economic Status • Competitive in many markets.
• Tax and cash incentives still desirable to facilitate most projects.
Policy Status • Encouraged by RPS and energy policies in most jurisdictions.
• Encounters occasional resistance from the public, environmental
groups, and others concerned primarily with visual impact of turbines.
Trends to Watch • Near-term global capacity growth of approximately 25% per year.
• Larger individual turbines (taller towers, longer blades).
• Larger aggregate wind farms.
• Advanced materials and controls.
• Rapid expansion into high wind resource areas (U.S. plains, North
Atlantic, Asia) as costs becomes competitive in more markets.
• Small, direct-drive turbines hybridized with gensets and energy
storage.
• Entrance into low-wind-speed environments.
3.1 Introduction
Wind is generated by small regional differences in atmospheric pressure caused by solar heating
of the Earth’s surface, radiation cooling at night, the passage of air over warm or cold ocean
water, the passage of fronts and storms, and other complex meteorological phenomena. Since the
1
Source: American Wind Energy Association, www.awea.org
3-1
10581090
Wind Power
dawn of civilization, people have relied on the wind to propel sailing vessels and power grain-
grinding mills, saw mills, water pumps, electric generators, and other devices.
In recent years, the modern wind turbine has been developed and commercialized for electricity
generation, principally in Europe and the United States. Wind is now one of the fastest growing
forms of electricity generation in the world. During 2009 alone, more than 38,000 MW of new
wind capacity was installed worldwide. As of September 2010, the installed wind generation
capacity was 36,698 MW in the United States and approximately 180,000 MW worldwide.
This section of the RETG addresses worldwide historical and projected installed wind generation
capacity, wind turbine technology, likely development pathways, wind turbine performance and
project cost estimates, operating and maintenance labor requirements, grid integration, onshore
and offshore project development process and market, environmental issues, potential for
greenhouse gas reduction and evolving markets for equipment and services.
3.2 Installed Wind Capacity
The worldwide installed wind generating capacity continues to grow at a rapid rate. Figure 3-1
shows the historical installed wind generating capacity through 2009 and the projected capacity
from 2010 through 2015 in the United States, Europe and the rest of the world. During 2009,
installed capacity increased by about 30%, from 122,158 MW to 160,084 MW, and is forecast to
nearly triple by 2014 to 447,689 MW.
Historical and Projected Installed Wind Capacity 1991- 2014

500,000
Europe U.S. Rest of the World
450,000
400,000
Installed Capacity (MW)
350,000
300,000
250,000
200,000
150,000
100,000
50,000
0
1991 1994 1997 2000 2003 2006 2009 2012
Year
Figure 3-1
Historical (1991-2009) and Projected (2010-2014) Installed Wind Generation Capacity in the
U.S., Europe, and Remainder of the World (Source: National Renewable Energy
Laboratory, Wind Power Monthly, and BTM Consult ApS, March 2010)
3-2
10581090
Wind Power
Table 3-1 presents the installed capacity by country at the end of each year during the period
2003 through 2009. As of the end of 2009, the United States led the world in installed wind
capacity (35,159 MW), followed by the People’s Republic of China (25,853 MW), Germany
(25,813 MW), Spain (18,784 MW). China installed the most wind capacity during 2009 with
13,750 MW, followed by the United States (9,922 MW), Spain (2,331 MW), Germany (1,917
MW), and India (1,172 MW).
The U.S. market has traditionally fluctuated from year to year depending on the status of the
federal production tax credit (PTC). The federal PTC expired on December 31, 2003 and was
extended to December 2005 by tax legislation passed by Congress in September 2004. Since
then Congress has extended the PTC four times, to December 2007, 2008, 2009 and 2012. As a
result of the PTC extensions, wind capacity additions have soared in the United States.
Furthermore, the enactment of the American Recovery and Reinvestment Act (ARRA) in
February 2009 introduced a greater degree of project finance flexibility that is expected to spur
wind development. In addition to extending the PTC to 2012, ARRA also allows all PTC-eligible
projects to qualify for the investment tax credit (ITC). The stipulation essentially grants owners
of non-solar renewable energy facilities the option to irrevocably choose the ITC in lieu of the
PTC and, in turn, earn an up-front tax credit equal to 30% of a project’s capital costs. The option
remains in effect for the current period of the PTC—through 2012 for wind energy facilities and
through 2013 for other qualified renewable energy systems.
But, perhaps more importantly, ARRA also offers a new federal cash grant program, which
provides grants in amounts equal to 30% of project costs in lieu of investment tax credits. The
stipulation now applies to wind projects now that PTC-eligible projects can also qualify for the
ITC. As a result, developers now have the option to forgo the tax equity portion of their
traditional financing packages and instead receive an equivalent amount of money in the form of
a government cash grant for at least until 2011.
Table 3-2 presents the forecast annual capacity additions for 2009 through 2013 and the
projected installed capacity at the end of 2013 by country. As discussed later, much of the new
wind capacity in Europe will be installed at offshore, shallow water sites. From 2009 through
2013, the United States is expected to install the most new wind generation capacity (52,000
MW), followed by P.R. China (42,800 MW), India (15,850 MW), Germany (11,300 MW), UK
(11,100 MW), and Spain (11,000 MW). By the end of 2013, the countries with the largest
projected installed wind generation capacity are expected to be the United States (77,237 MW),
P.R. China (54,921 MW), Germany (35,233 MW), Spain (27,453 MW), and India (25,505 MW).
Figure 3-2 compares the operating wind capacity in different nations and regions of the world at
the end of 2007 and 2008, indicating both overall capacity growth and shifting shares of the total.
3-3
10581090
Wind Power
Table 3-1
Operating Wind Generation Capacity: End of Year 2003–End of Year 2009
(Source: BTM Consult ApS, March 2010)
Operating Wind Capacity by Country (MW)

Region/Country 2003 2004 2005 2006 2007 2008 2009
Europe
Germany 14,612 16,649 18,445 20,652 22,277 23,933 25,813
Spain 6,420 8,263 10,027 11,614 14,714 16,453 18,784
Italy 922 1,261 1,713 2,118 2,721 3,731 4,845
France 274 386 775 1,585 2,471 3,671 4,775
UK 759 889 1,336 1,967 2,394 3,263 4,340
Denmark 311 585 1,087 1,716 2,150 2,829 3,474
Portugal 3,076 3,083 3,087 3,101 3,088 3,159 3,408
Netherlands 938 1,081 1,221 1,557 1,745 2,222 2,226
Greece 428 478 554 571 789 1,024 1,537
Sweden 538 587 705 862 987 1,102 1,198
Ireland 230 339 498 748 807 1,015 1,187
Austria 415 607 820 966 983 997 997
Rest of Europe (Other East
na 75 132 263 374 671 987
European and Baltic countries)
Turkey 20 20 20 76 225 512 984
Poland 55 55 65 170 313 472 849
Norway 78 106 177 222 297 385 605
Belgium 101 158 275 328 355 385 390
Finland 53 83 85 89 113 113 117
Luxembourg 16 12 12 12 12 21 21
Switzerland 6 8 11 11 11 13 17
Total Europe 29,252 34,725 41,044 48,627 56,824 65,971 76,554
North America
United States 6,361 6,750 9,181 11,635 16,879 25,237 35,159
Canada 351 444 683 1,459 1,845 2,371 3,321
Total North America 6,712 7,194 9,864 13,094 18,724 27,608 38,480
Asia
China 571 769 1,264 2,588 5,875 12,121 25,853
India 2,125 3,000 4,253 6,228 7,845 9,655 10,827
Taiwan 8 16 72 118 224 369 411
Rest of Asia: Indonesia, N Korea,
Malaysia, Philippines, Thailand, na 4 28 28 28 28 56
Vietnam, etc.
Total Asia 2,704 3,789 5,617 8,963 13,973 22,174 37,147
3-4
10581090
Wind Power
Table 3-1 (continued)

Operating Wind Generation Capacity: End of Year 2003–End of Year 2009
Operating Wind Capacity by Country (MW)

Region/Country 2003 2004 2005 2006 2007 2008 2009
South & Central America
Brazil 22 31 31 31 231 392 687
Mexico 3 3 3 3 86 86 332
Other Americas na 50 54 54 56 79 153
Costa Rica 71 71 71 79 79 79 104
Argentina 26 30 30 31 31 31 33
Total S. & Central America 122 185 189 198 493 667 1,309
Pacific Region
Japan 384 761 991 1,159 1,457 1,681 2,033
Australia 103 239 421 717 796 972 1,587
New Zealand 37 55 167 167 170 321 325
South Korea 8 21 69 89 194 235 311
Pacific Islands na na na 6 11 11 16
Total Pacific Region 532 1,081 1,653 2,132 2,628 3,220 4,272
Middle East & Africa
Egypt 123 146 180 231 310 384 552
Morocco 54 54 64 122 124 206 254
Middle East: Jordan, Iran, Iraq,
28 28 28 28 28 62 160
Israel, Saudi Arabia, Syria, etc.
Tunisia 72 101 101 101 101 101 101
Rest of Africa: Algeria, Cape
5 6 6 6 6 44 77
Verde, Ethiopia, Libya, S Africa, etc.
Total Middle East & Africa 282 335 379 488 569 797 1144
Total World 40,216 47,885 59,234 74,283 94,005 122,158 160,086
3-5
10581090
Wind Power
Table 3-2
Forecast Wind Generation Capacity Additions: 2010–2014
Cum
Cum MW Forecast Wind Capacity Additions (MW)
MW
Total End
Region/Country End 2009 2010 2011 2012 2013 2014
Added 2014
Europe
Austria 997 50 100 150 250 250 800 1,797
Belgium 605 430 150 350 200 100 1,230 1,835
Bulgaria 131 75 100 100 150 250 675 806
Czech Republic 160 50 100 150 150 200 650 810
Denmark 3,408 350 100 500 150 150 1,250 4,658
Estonia 131 75 75 100 100 150 500 631
Finland 117 100 200 200 200 200 900 1,017
France 4,775 1,600 2,100 2,500 2,700 3,000 11,900 16,675
Germany 25,813 2,000 2,400 2,500 3,000 3,500 13,400 39,213
Greece 1,198 200 200 250 300 300 1,250 2,448
Hungary 229 100 100 200 250 300 950 1,179
Ireland (Republic) 1,187 200 250 400 350 400 950 2,787
Finland 4,845 1,300 1,500 1,500 1,500 1,500 7,300 12,145
Italy 31 50 50 75 75 100 350 381
Latvia 103 50 75 100 100 150 475 578
Lithuania 2,226 300 250 400 400 500 1,850 4,076
Netherlands 390 300 500 700 1,000 1,000 3,500 3,890
Norway 849 300 500 700 1,000 1,200 3,700 4,549
Poland 3,474 1,000 1,000 1,000 1,200 1,000 5,200 8,674
Portugal 129 100 200 200 200 500 1,200 1,329
Rumania 18,784 2,000 2,500 2,000 2,500 2,000 11,000 29,784
Spain 1,537 600 750 750 900 1,200 4,200 5,737
Switzerland 17 25 25 50 75 100 275 292
Turkey 984 400 500 500 600 500 2,500 3,484
United Kingdom 4,340 1,600 2,200 2,500 3,000 2,500 11,800 16,140
Rest of Europe: Luxembourg,
Cyprus, Malta, Iceland, Balkan 94 50 75 150 150 200 625 719
States, etc.
Total Europe 76,554 13,305 16,000 18,025 20,500 21,250 89,080 165,634
3-6
10581090
Wind Power

Forecast Wind Generation Capacity Additions: 2010–2014
Cum
Cum MW Forecast Wind Capacity Additions (MW)
MW
Total End
Region/Country End 2009 2010 2011 2012 2013 2014
Added 2014
Americas
Brazil 935 300 400 500 500 700 2,400 3,335
Canada 3,321 1,200 1,500 2,000 2,500 3,000 10,200 13,521
Mexico 453 200 300 400 400 600 1,900 2,353
United States 35,159 8,000 10,000 15,000 15,000 17,000 65,000 100,159
Other Americas 483 300 400 500 500 800 2,500 2,983
Total Americas 40,351 10,000 12,600 18,400 18,900 22,100 82,000 122,351
Asia
P.R. China 25,853 14,000 15,000 15,500 16,500 18,000 79,000 104,853
India 10,827 2,500 2,500 3,500 4,000 4,000 16,500 27,327
Taiwan 411 150 250 400 600 800 2,200 2,611
Rest of Asia: Indonesia, North
Korea, Malaysia, Philippines, 56 50 100 200 200 300 850 906
Thailand, Vietnam, etc.
Total Asia 37,147 16,700 17,850 19,600 21,300 23,100 98,550 135,697
OECD-Pacific
Australia 1,886 500 550 650 800 800 3,300 5,186
Japan 2,208 300 400 500 700 700 2,600 4,808
New Zealand 467 150 250 250 250 250 1,150 1,617
South Korea 311 250 300 450 600 700 2,300 2,611
Total OECD-Pacific 4,872 1,200 1,500 1,850 2,350 2,450 9,350 14,222
Other Countries
Egypt 552 250 300 500 700 700 2,450 3,002
Morocco 254 125 150 150 200 200 825 1,079
Other African Countries: Algeria,
Cape Verde, Ethiopia, Libya, 209 200 200 250 400 450 1,500 1,709
Tunisia, South Africa, etc.
Middle East: Jordan, Iran, Iraq,
101 100 150 200 300 500 1,250 1,351
Saudi Arabia, etc.
Transition Economies: Russia,
White Russia, Ukraine, 28 150 300 500 750 900 2,600 2,628
Uzbekistan, Kazakhstan, etc.
Rest of the World 18 0 0 0 0 0 0 18
Total Other Countries 1,162 825 1,100 1,600 2,350 2,750 8,625 9,787
Total World 160,084 42,030 49,050 59,475 65,400 71,650 287,605 447,689
3-7
10581090
Wind Power
Operating Wind Capacity by Country or Region

End of 2008 End of 2009
25%
Total Capacity
22.0% 2008: 122,100 MW
20.9%
20.7% 2009: 160,100 MW
Percentage of Total World Capacity
20.0%
19.6%
20%
16.1% 16.1%
15%
13.5%
11.7%
9.9%
10%
7.9%
6.8%
5%
3.5%
3.1%
1.9% 2.1%
1.1% 1.2%
0.7% 0.7%
0.3% 0.3%
0%
United Rest of P.R. China Germany Spain India Pacific Canada South and Middle Rest of
States Europe Region Central East and Asia
America Africa
Country or Region
Figure 3-2
End-of-Year Operating Wind Capacity by Country and Region: 2008 and 2009 (Source: BTM
Consult ApS, March 2010)
Historically, the U.S. market fluctuates from year to year depending on the status of the federal
production tax credit (PTC). The federal PTC expired on December 31, 2003, and was extended
to December 2005 by tax legislation passed by Congress in September 2004, and then to
December 2007 by the 2005 Energy Policy Act. It was further extended to December 2008 by
Congress when it passed a tax and trade policy bill in December 2006. It was then extended until
December 2009 when Congress passed an energy improvement act in October 2008. And it was
more recently extended to 2012 by ARRA. As a result of the PTC extensions in 2005, 2006,
2007, and 2008, wind capacity additions in those years exploded in the United States, totaling
about 2,430, 2,454, 5,244, and 8,358 MW, respectively. Meanwhile, stipulations embedded in
ARRA—the extension of the PTC to 2012, the option of project owners to choose the ITC in lieu
of the PTC, and the cash grant program—were expected to bolster wind energy capacity
additions in 2010 and beyond. However, the American Wind Energy Association (AWEA)
reports that only 395 MW of wind capacity was installed during third quarter 2010, and 1634
MW through the third quarter, which were the lowest levels since 2007 and 2006. The slowdown
is the result of the economic downturn and the difficulty of arranging for financing, low power
value, and the uncertainty regarding passage of a federal renewable energy standard (RES). After
the November 2010 elections, it is unlikely that there will be a federal RES in the near future.
3-8
10581090
Wind Power
Figure 3-3 shows the distribution of U.S. wind generation capacity among the states as of June
2009. At the end of 2008, 23 states contributed approximately 50 MW or more of wind power.
Recent additions to this list since 2006 include Hawaii and South Dakota. Meanwhile, as of the
end of 2008, two states—Minnesota and Iowa—derive over 7% of their power needs from wind
and 13 others get more than 2%.
Forecasts of future U.S. wind capacity vary depending on different assessments of how planned
projects will proceed. As shown in Table 3-2, BTM Consult ApS forecasts that cumulative U.S.
wind capacity will reach approximately 43,000 MW in 2010. However, since the base of new
wind plant construction has slowed, the end-of-year installed capacity will be lower.
Installed Megawatts of Wind Power as of July 20, 2010 (Source: AWEA)
Washington
1914 Montana ME
North
386 Dakota 200
Minnesota
Oregon 1222 1797 VT
1920 Idaho 6 NH
South 26
164
Idaho Dakota Wisconsin
449 New York
11 Wyoming 412 Michigan 1274 CT 68
1101 143 MA 15
Iowa Penn. NJ 8
Nebraska
Nebraska 3670 748 RI 2
20
153 Ohio DE 2
Utah IN
10
223 Illinois 1127
Colorado WV
California 1848
1248 Kansas 414
2739 Missouri
1026
457
Oklahoma Tennessee
Arizona 1130 29
New Mexico
63 597
Texas
9707
Alaska
9
Hawaii
63
Existing Power Capacity (in MW) 0-10 11-100 101-1000 1001-10,000
Wind Power Capacity State by State-July 2010, Total: 36,698 MW
Figure 3-3
Installed Wind Generation Capacity in the United States by State as of July 2010
(Source: American Wind Energy Association)
3-9
10581090
Wind Power
3.3 Wind Energy Principles

Capturing the wind’s kinetic energy depends on the efficiency of the conversion process as well
as the power of the wind, which can be expressed by the equation:
Pw = 1/2 ρA V3
where ρ = density of the air, A = swept area of the rotor, and V = velocity of the wind.
It can be seen that the power in the wind varies proportionally with the cube of the wind speed,
which has important bearing on the design and siting of wind turbines. As a result, even a small
increase in wind speed can substantially boost the power available in the wind. For example, a
25% increase in wind speed approximately corresponds to a doubling in the power contained in
the wind, which illustrates the importance of accurate resource assessment to a project’s success.
Because of the strong link between wind speed and power, and because of the variable nature of
the resource, the direct use of average wind speeds to estimate energy generation can lead to
gross inaccuracies. Therefore, it makes sense to talk about the wind resource in statistical terms.
For sites where resource measurements have not been taken over a significant period of time and
actual distributions are unavailable, meteorologists have found that the Weibull probability
function provides a good approximation of the distribution of wind speeds over time. The
Weibull function is:
f(v) = (k/Vc)*(V/Vc)(k-1)*exp((-V/Vc)k)
where k is a shape parameter between 1 and 3, and Vc is the characteristic velocity defined as the
weighted average wind speed at hub height. The Rayleigh distribution, shown in Figure 3-4, is a
special case of the Weibull function in which k=2. It requires only the characteristic wind speed
to define an approximate distribution.
Although the greatest power production occurs at higher wind speeds, most sites do not
experience those high wind speeds with great frequency. Due to laws of conservation of mass
and momentum, there is a theoretical limit to the fraction of energy in the wind that can be
extracted by the rotor, called the Betz Limit. The theoretical limit of rotor energy collection
efficiency is 59%. Since every other system in the turbine contributes losses as well, only a
fraction of the power in the wind can be extracted by a practical wind turbine. The actual
mechanical power output is:
Pm = Cp (1/2 ρ A V ) = Cp Pw
3
where Cp is the coefficient of performance and represents overall system efficiency.
3-10
10581090
Wind Power
Figure 3-4
Rayleigh Probability Density Function for Wind Speed
The coefficient of performance, Cp, varies with wind speed and blade tip speed, and is often
plotted as a function of the ratio of the blade tip speed to the wind speed because it clearly shows
the optimum operating point and is applicable to both constant-speed and variable-speed wind
turbines. Figure 3-5 illustrates a typical curve for the coefficient of performance vs. the tip
speed/wind speed ratio.
Figure 3-5
Coefficient of Performance (Cp) vs. Tip Speed/Wind Speed Ratio
3-11
10581090
Wind Power
3.4 Developments in Wind Turbine Technology
Most utility-scale wind turbines currently available from established turbine manufacturers
utilize the “Danish concept” turbine configuration. This configuration uses a three-bladed rotor,
an upwind orientation (blades positioned upwind of the tower), and an active yaw system to keep
the rotor oriented into the wind. Figure 3-6 illustrates the major components of a typical utility-
scale wind turbine design. The nacelle contains the drive train, which usually consists of a low-
speed shaft connecting the rotor to the gearbox, a two-or three-stage speed increasing gearbox,
and a high-speed shaft connecting the gearbox to the generator. Each turbine is equipped with a
transformer to step up the generator voltage to the on-site collection system voltage. Sometimes
this transformer is mounted within the nacelle, and sometimes pad-mounted transformers are
used near the base of the tower. The current status and recent developments of each of the main
wind turbine components are described in the following sections. The main sources of the
information are the 2010 EPRI report, Advanced Wind Turbine Technology Assessment–2010
[67] and the 2009 report, Wind Power Technology Status and Performance and Cost Estimates–
2009 [68].
Controller
Wind Speed &
Direction Sensors Main Shaft Blades
Hub
Nacelle
Cooling
System
(Radiator)
Tower
Generator
Access Door
Mechanical
Gearbox Yaw Motor
Brake
Figure 3-6
Components of a Typical Wind Turbine (Source: Danish Wind Industry Association,
www.windpower.org)
3.4.1 Generators & Power Electronics
In the early days of wind energy, wind plants were small relative to conventional sources of
generation. Consequently, wind plants were not required to help stabilize the power grid and
were even required to disconnect from the grid during transient events like grid faults. In recent
years, the power capacity of wind plants has grown significantly such that annual capacity
additions are comparable to those of conventional power plants. The penetration of wind energy,
which is a ratio of generation from wind to the total electrical load of the grid, has also increased.
This growth has caused a dramatic shift in the grid interconnection requirements imposed on
wind energy. Wind plants are increasingly being required to behave more like conventional
generation sources.
3-12
10581090
Wind Power
Grid integration requirements vary from region to region, but most follow the same basic
construct. Wind plants are required to contribute to the stability of the grid voltage by controlling
the flow of reactive power. Providing voltage support in this way is required during normal grid
operation, as well as during grid faults. There are limits to how quickly the power output of wind
plants can change over the course of fractions of seconds to one hour. Meeting these new, more
stringent grid-integration requirements, in addition to reducing structural loading and improving
energy capture, affects the design of the wind turbines, especially the generator topology.
3.4.1.1 Squirrel Cage Induction Generator
In the early 1990s, the standard wind turbine topology consisted of a constant-speed
asynchronous squirrel cage induction generator (SCIG) connected directly to the grid through a
transformer. Constant-speed wind turbines with SCIGs are relatively simple. The rotational
speed of the generator is essentially controlled by the frequency of the grid, and the power and
torque of the turbine are controlled by pitching the turbine blades. However, the reaction time of
the blade pitch system limits the wind turbine’s ability to limit torque and power, which results
in significant torque and power spikes during wind gusts. These spikes increase fatigue loading
and adversely affect the grid.
The stability of the grid voltage is maintained by controlling the flow of reactive power. SCIGs
always consume reactive power, and their reactive power consumption is not controllable. Power
factor correction capacitors are typically used to compensate for the reactive power consumed by
SCIGs, but SCIGs are unable to contribute to the stability of the grid. Most manufacturers have
traded the simplicity of the fixed-speed SCIG for the controllability available in more complex
generator configurations.
3.4.1.2 Variable-Speed Turbine
The rotational speed of many contemporary wind turbines is decoupled from the grid frequency
in order to achieve variable-speed operation. Variable-speed operation has two primary
advantages over fixed-speed operation. Variable-speed turbines are able to limit power and
torque by letting the rotational speed of the turbine rotor vary during wind gusts. The improved
control of torque reduces mechanical loading, and the improved control of power makes the
output electricity more palatable to the grid. The second advantage of variable-speed operation is
improved aerodynamic efficiency, which leads to increased energy capture in the lower wind
speed range. However, as discussed below, sophisticated power electronics are required to
enable variable-speed operation. Due to the parasitic electrical load associated with power
electronics, the net improvement in energy capture is reduced.
A patent held by General Electric (GE) has historically limited the use of variable-speed
technology in North America by other manufacturers. Recently GE has licensed the technology
to several other manufacturers; however, it is unclear if the license includes the capability to
provide reactive power at each turbine, which is held under a separate patent owned by GE. With
the imminent expiration of GE’s variable-speed patent, manufacturers who already utilize
variable-speed technology in Europe are likely to implement the technology in the United States.
3-13
10581090
Wind Power
3.4.1.3 Power Electronics
Power electronics are solid-state devices (semiconductors) that convert the form of electric
power by changing the frequency and/or voltage and current. Power electronics play a significant
role in the generators of many wind turbines by allowing variable-speed operation and improving
controllability. One of the most common solid-state devices utilized in wind turbine generators is
the insulated-gate bipolar transistor (IGBT). IGBTs function as fast switches that can withstand
several thousand volts and conduct current on the order of 1000 amps. In wind turbine
applications, power electronics are primarily used to rectify (convert ac to dc) and/or invert
(convert dc to ac) currents and voltages.
3.4.1.4 Squirrel Cage Induction Generator with Full Power Conversion
The controllability and operating range of the SCIG can be appreciably improved by connecting
the SCIG to the grid through a full-load frequency converter. A full-load frequency converter is a
power electronics package that rectifies the variable-frequency ac output of the generator and
then inverts it at the frequency of the grid. The addition of the full-load frequency converter
allows high bandwidth control of both the active and reactive power of the turbine. This enables
the wind turbine to contribute to the stability of the grid during normal grid operation and
extreme grid events such as faults. The full-load frequency converter also makes variable-speed
operation possible, which has benefits in reducing mechanical loading.
3.4.1.5 Doubly-Fed Induction Generator
The doubly-fed induction generator (DFIG) is an asynchronous machine that is currently the
most common wind turbine generator type. The stator of the DFIG is connected directly to the
grid through a transformer, similar to a SCIG. However, the generator rotor of the DFIG is
connected to a power electronics package, called a partial-load frequency converter, through slip
rings. In this configuration, a fraction of the power exchanged between the wind turbine and the
grid flows through the power electronics frequency converter, and the remaining power flows
through the stator. Similar to the full-load frequency converter previously mentioned, the partial-
load frequency converter allows active and reactive power to be controlled and permits variable-
speed operation. The dynamic range of the DFIG is slightly less than generator topologies with
full-load frequency converters, but the cost of the power electronics is less for partial-load
conversion than for full-load conversion.
3.4.1.6 Synchronous Generator
There are several types of synchronous generators used in wind turbines; however, all of them
have an electrical or mechanical means of decoupling the speed of the turbine rotor from the
frequency of the grid. As opposed to induction generators, synchronous machines appear very
stiff to the mechanical components of the drive train. Wind turbines with synchronous generators
must have variable-speed turbine rotor operation in order to prevent severe mechanical loading
from wind gusts. The most common means of providing this decoupling is to connect a full-load
frequency converter between the generator and the grid. The operation and benefits of the full-
load frequency converter previously discussed for the SCIG with full power conversion are true
3-14
10581090
Wind Power
for synchronous generators. Another, less common, means for decoupling the turbine rotor speed
from the grid frequency is through the use of a variable-speed gearbox. In this configuration, the
synchronous generator is connected directly to the grid, and a variable-speed gearbox is
connected between the generator and the turbine rotor.
The synchronous generator can either be a permanent magnet generator (PMG) or a wound rotor
synchronous generator (WRSG).
3.4.1.7 Direct-Drive Low-Speed Wound Rotor Generators
Some turbine manufacturers have chosen to avoid the reliability issues that persistently plague
gearboxes by eliminating them from the design. These direct-drive designs connect the rotor
shaft directly to a low-speed generator. Since the generator speed is inversely proportional to
pole count, direct-drive generators have multiple poles and are of large diameter. Usually the
generator housing is integrated into the nacelle frame structure.
Figure 3-7 is a schematic of an Enercon direct-drive turbine. This first-generation low-speed

generator is exemplified by the doubly-fed, asynchronous induction generator with a wound
rotor. The large diameter of the generator influences the design of the nacelle.
Figure 3-7
Schematic of Enercon Direct Drive Turbine (Source: Enercon website,
http://www.enercon.dk/en/_home.htm. Retrieved September 10, 2010)
3.4.1.8 Direct-Drive Low-Speed Permanent Magnet Generators
What are sometimes termed second-generation direct-drive turbines often employ full-
conversion permanent magnet (PM) generators. Using permanent magnets to produce the rotor
field simplifies the mechanical design for, by example, eliminating the need for slip rings. In
3-15
10581090
Wind Power
addition, PM generators are generally more efficient, particularly at partial load. Other
advantages include the higher magnetic field strength and lighter weight compared to a wound
rotor. [70] Figure 3-8 shows a 1.5-MW prototype PM generator tested at NREL’s National Wind
Technology Center.
Figure 3-8
Direct-Drive 1.5-MW Generator under Testing at NREL’s National Wind Technology Center
(Source: Lee, Fingersh. NREL Photographic Information eXchange via
http://www.nrel.gov/data/pix/Jpegs/14689.jpg)
One disadvantage of PM generators, however, is the higher material cost for the permanent
magnets, which are made from alloys of the rare earth elements neodymium or samarium-cobalt.
In addition, since 95% of the world’s PM material is currently supplied by China, the rapid and
widespread deployment of low-speed PM generators could lead to supply problems that might
drive up prices [71]. On the other hand, 45% of the known sources of rare earth elements are
found outside of China, and some projections show that annual demand of rare earth magnets
will amount to less than 0.2% of known global supply by 2015. Other projections suggest the
global supply may be as much as 200 million metric tons.
3.4.1.9 Superconducting Low-Speed Generators
For Europe to meet its goal of installing 300 GW of wind power by 2030, the high population
density onshore means that 120 GW will need to be installed offshore. For the United States to
reach its goal of 20% wind energy by 2030, 50 GW of offshore wind power will be needed. Asia
is also developing an offshore market. To come anywhere close to meeting these combined
offshore market demands, offshore turbines rated up to 10 MW and possibly larger will most
likely be necessary. Turbines of that size can take advantage of technologies that are not
economically feasible at a smaller scale.
3-16
10581090
Wind Power
One of the technologies emerging for direct-drive turbines larger than 5 MW is the application of
high-temperature superconductors (HTS) in the generator [72, 73]. High-temperature
superconducting generators may have the potential to produce comparatively smaller, lighter and
more efficient machines for the 5- to 10-MW turbines currently being designed for the offshore
market. NREL is sponsoring American Superconductor Corp. (ASC) in conducting research into
the economic impact and costs of a direct-drive, superconducting 10-MW generator that uses
high-temperature superconducting tape instead of copper wire for the generator’s rotor [74]. The
use of superconducting tape for windings results in a significantly smaller generator for the same
power rating. A smaller generator will be lighter and, because it is superconducting, its losses
will be lower than those of a conventional turbine. According to ASC, the weight of the 10-MW
superconducting generator they are building is approximately 120 metric tons. In contrast,
conventional designs of that power rating weigh 300 metric tons. Figures 3-9 shows an exploded
view of a superconducting generator.
HTS race-track coils
Rotor support
Rotor iron
Back iron
Cu stator
Figure 3-9
Example of an Eight-Pole, Air-Cooled Superconducting Generator (Source: Abrahamsen, A
B, et. al. [73], with permission)
3.4.2 Blades and Rotor
The purpose of the blades and rotor is to convert the kinetic energy of the wind to mechanical
rotational energy that can then be converted into electricity. Airfoils along the blade are oriented
so that a component of the lift vector is directed into the rotor plane of rotation, producing torque
about the low-speed shaft at the rotor center. The remaining component of lift is perpendicular to
the plane of rotation, and produces bending loads on the blade as well as thrust loading, which
must be accommodated by the structural load path (shafts, bearings, tower-top connection, tower
and foundation). General design goals for the blade and rotor include maximizing aerodynamic
efficiency (energy production) while minimizing loads on the blades and turbine structure. For
example, the use of thicker airfoils, particularly at the blade root, has resulted in improved
structural integrity while maintaining aerodynamic performance.
3-17
10581090
Wind Power
Figure 3-10 shows the cross-section of a typical blade construction. Aerodynamic loading is
primarily carried by the “spar caps” on the upper and lower airfoil locations. A sandwich-style
shell provides the aerodynamic shape and resists panel buckling. Shear webs span the upper and
lower shells. In the case of Figure 3-10, shear webs are at the front and rear of the spar caps,
effectively forming a box-beam; however, several variants on the shear web layout are in
widespread use.
Figure 3-10
Cross-Section of Wind Turbine Blade (Source: DNV-GEC)
Blade technology is advancing in the areas of design methods, load-mitigating technology,

materials and manufacturing. A few examples are addressed in the following paragraphs. The
overarching objectives of these advancements are to enable increased energy capture with
reduced or mitigated loads, weight, and cost.
Design practices are continually being refined to take advantage of developments in computing
speed and cost. Three-dimensional modeling of blade shapes and finite element analysis are now
the industry standard. Computational fluid dynamics is increasingly being used, although the
majority of design work is still performed using more simplified modeling approaches.
Several technologies are being developed to reduce aerodynamic loading in blades associated
with turbulence or other transient changes in wind. Advanced control methods are being
developed to independently pitch the blades. Several manufacturers and research institutions are
working on small fast-acting devices to control the aerodynamic condition of the blade via active
control technology. Passive load control is also being investigated. One of the major areas of
interest is in “bend-twist coupling,” where an increase in aerodynamic loads will cause the blade
to twist in such a way as to reduce the loading. This is shown schematically in Figure 3-11,
where the orientation of the fibers is used to achieve the desired twist-coupling. At least one
manufacturer has developed a prototype rotor that uses geometry to realize bend-twist coupling.
Measures must be taken to ensure sufficient clearance between the blade tip and tower when the
blade is under aerodynamic loading. These measures may include a combination of increased
rotor overhang distance, an upward tilting of the nacelle, or a coning angle of the hub. LM
3-18
10581090
Wind Power
Glasfiber has a patented approach of adding “pre-curve” to the blade so that the tip clearance is
increased prior to loading.
Figure 3-11
Bend-Twist Coupling Achieved by Fiber Orientation (Source: Laird, Daniel. Sandia National
Laboratories 2006 Wind Turbine Blade Workshop)
The growth in blade weight with increased length is non-linear. Generally, the weight increases
faster than the blade length. This means that, without material innovations, the average rating of
a utility-scale wind turbine will reach an upper limit unless the growth of blade weight can be
controlled. Advancements in materials are ongoing to reduce blade weight and cost. Per unit
weight, carbon fibers are significantly stronger and stiffer than glass fibers; however, carbon
fibers are much more expensive. As a result, the majority of blade manufacturers use fiberglass.
Notable exceptions are Vestas and Gamesa, which use carbon fiber in the spar caps of their
largest blades.
Blades are typically constructed from fiber-reinforced plastic (FRP) materials. Strength-per-
weight and stiffness-per-weight are both improved when the amount of load-carrying fibers is
increased relative to the plastic matrix. Newer manufacturing processes have been introduced
to increase the fiber content of the blade laminate. Nearly all new production is done with either
“vacuum assisted resin transfer molding” (VARTM) or the use of preimpregnated fabrics
(prepreg), rather than the traditional “wet layup” process. One manufacturer has developed an
innovative process in which a complete blade is produced in a single VARTM process, rather
than separate manufacture of skins and shear webs followed by a secondary assembly step. The
approach eliminates the need for bond lines, reducing weight and increasing structural integrity.
Several approaches have been considered to facilitate the transport and erection of multi-
megawatt-scale blades. Several manufacturers are evaluating or developing a mid-span joint in
the blade so it can be manufactured and shipped in parts prior to assembly at the site.
3-19
10581090
Wind Power
3.4.2.1 Passive Aerodynamic Control
Passive aerodynamic devices have been used by manufacturers for many years to fine tune the
performance of rotor blades. They include vortex generators, shown in Figure 3-12, leading edge
slats tall strips, spoilers and winglets [75]. Vortex generators and leading edge slats affect what is
known as the boundary layer over an airfoil section, which in turn affects lift and drag
characteristics. The general effect of this control method is to delay the onset of trailing-edge
separation and subsequent stall, and thereby increase lift.
Figure 3-12
Vortex Generators (Counter-Rotating Array)
Fences and winglets would be used to mitigate the effects of three-dimensional flow along the
blade, thus improving predictability and performance. Fences would be deployed on inboard
sections where three-dimensional flow is strongest and can potentially interfere with the section
further outboard. Winglets mitigate tip loss effects by blocking the flow around the tip from the
high-pressure surface of the blade to the low-pressure side. Winglets may have an impact on
acoustic emissions as well.
Applying passive aerodynamic devices on the inboard sections of the blade could yield increases
in annual energy output on the order of 1% to 3% without adding much to the cost of the blade.
3.4.2.2 Active Aerodynamic Control
As with passive aerodynamic devices, active devices also fall into two major categories:
boundary layer control and camber modification. The technologies that fall under this heading
include leading edge slats, “active” vortex generators, plasma actuators, and boundary layer
suction/blowing. Circulation control is a technology that does not strictly fit in the category of
3-20
10581090
Wind Power
“boundary layer control,” but has a similar effect on lift characteristics. The feasibility of using
active boundary layer modification of local airfoil sections to control turbine loads appears less
advantageous.
3.4.2.3 “Stealth” Rotor Blade
Aviation radar interference is an increasingly common issue that developers face when planning
and developing new wind projects. Wind turbines can reflect radar signals, creating false returns
and blind spots that can sometimes appear on the operator’s screens. Aviation, defense, and
weather radar systems—particularly older systems—can be impacted. This issue has the
potential to hold up the installation of hundreds of megawatts of wind development, both in the
United States and in Europe. Manufacturers are investigating ways to reduce the interference
“footprint” caused by wind turbines and wind farms. Vestas and technology services company
QinetiQ are working to develop a “stealth” rotor blade that will be less visible to radar by
applying a 5-mm coating to the blades that absorbs incoming radar signals, similar to the
coatings used on stealth aircraft. [76][77] However, since the coating adds 1,200 kg to the weight
of a large turbine blade, Vestas is working to incorporate the coating into the fiberglass-
reinforced epoxy and plastic foam of the rotor blade’s composite structure.
Alternative approaches to reducing the impacts of wind turbines on radar involve developments
in the signal-producing algorithm used by the radar system, as well as transitioning from analog
radar stations to more modern digital equipment. Such upgrades will occur at some point in the
future, but it is unclear when or who will ultimately bear the upgrade costs.
3.4.2.4 “Smart” Blades
“Smart” blades, also known as intelligent rotors, are wind turbine blades that are instrumented for
active load measurement, control and damage detection. The idea is to continuously monitor and
control the performance and loads of a wind turbine at the source where the wind meets the rotor.
This is a significant area of basic research at organizations such as Sandia National Laboratories.
In general, the use of sensors for active measurement and control is increasing. Potential benefits
of “smart” blade technology include the ability to provide real-time information to the control
system to prevent catastrophic wind turbine damage. “Smart” blade systems are complementary to
active aerodynamic control systems, providing real-time sensor data about the state of the rotor
needed to properly adjust control surfaces such as flaps or micro-tabs.
Merging a “smart” monitoring and control system with active aerodynamic devices such as flaps
or ailerons leads to a “smart” turbine that can adapt and respond in real time to fluctuations in the
wind speed at the rotor blades.
3.4.2.5 Flatback Airfoils
Using a systems approach to simultaneously optimize manufacturing and the structural and
aerodynamic performance of rotor blades, researchers at Sandia National Laboratories have
developed a design that employs flatback airfoils along the inboard portion of the blade. Flatback
3-21
10581090
Wind Power
airfoils are characterized by a blunt trailing edge in place of the conventional trailing edge that
tapers to a point.
Compared to a blade with the same thickness but a sharp trailing edge, blades that incorporate
flatback airfoils are likely to be stronger, lighter, generate more lift along the inboard section and
have less sensitivity to blade soiling. The flatback geometry simplifies blade construction and the
truncated blade width (also known as chord length) makes them easier to transport.
3.4.2.6 Blade Manufacturing Processes
Demand for an increased number of wind turbine blades that are manufactured more quickly has
spurred innovation in the materials and process sector. Current blade manufacturing strategies are
both labor and capital intensive. As a result, many manufacturers are looking for automated
solutions.
Industrial automation companies have been applying their knowledge of aerospace automation to
the wind industry in recent years. Robots are used to lay dry “tows” or strips of woven fiberglass
cloth in preparation for them to be sprayed, vacuum-bagged, resin infused and cured. Using this
process, the Spanish company MTorres has a goal of laying down 2,000 lbs (907 kg) of material
per hour in order to complete one blade every eight hours. MTorres uses proprietary software to
provide instructions to the lamination machine controller, which generates and executes machine
codes for carrying out the lamination process. The automation technology is still in development,
with plans to begin production as early as 2011, possibly in the United States.
3.4.2.7 Blade Transport and Shipping
Several approaches have been considered to facilitate the transport and erection of multi-
megawatt-scale blades. Several manufacturers are evaluating or developing a mid-span joint in
the blade so it can be manufactured and shipped in parts prior to assembly at the site.
3.4.2.8 Blade Test Facilities
With the continuing advancements in blade technology, there is increased interest in blade
research and test facilities in the United States. In collaboration with NREL, two new structural
test faculties are planned: one in Boston, Massachusetts, and one on the Texas Gulf Coast.
3.4.3 Yaw Systems
All horizontal-axis turbines use some type of system to rotate (“yaw”) the rotor into the wind.
Older turbine designs with down-wind rotor configurations, and some small turbines with tail-
fins, rely on the balance of wind forces to passively align the rotor to the wind direction, but
all upwind turbines use an active system. The most common arrangement uses a large-diameter
rolling-element turntable bearing (“slewing ring”) mounted to the top of the tower. The nacelle
frame is mounted to the rotating ring of this bearing.
3-22
10581090
Wind Power
The nacelle is rotated about the tower top with multiple yaw drives, which are mounted in the
nacelle frame and engage gear teeth in the fixed bearing ring. Most yaw drives are made up of
multi-stage planetary or worm-gear units driven by electric motors, although some turbines
utilize hydraulic motors instead. The number of yaw drives required varies with the size of the
turbine, from two for a typical sub-megawatt-class turbine to as many as eight for the largest
multi-megawatt-class turbines.
The speed of yaw rotation is very slow, on the order of tens of minutes for one full rotation.
Yaw alignment is detected using the wind vane mounted to the top of the nacelle; when the rotor
is not aligned directly into the wind, the vane will be offset from the rotor axis of rotation,
indicating a yaw error. The control system then starts and stops the yaw motors in the correct
direction until the yaw error is below a specified threshold. To keep the turbine from constantly
yawing in response to short-term variations in the wind speed, algorithms are used to filter the
yaw error measurement.
The aerodynamic forces on the rotor can cause very large yaw torques. Early turbine designs
used brakes on the yaw motors to counteract this torque and keep the rotor in position.
Experience has shown that these forces change rapidly, and over time can damage the yaw and
cause wear on the yaw gear teeth. Modern designs incorporate friction brakes on the nacelle
frame itself. These are often hydraulically-actuated so that the friction can be at least partially
released when the yaw drives are active.
An alternative arrangement for the yaw system uses a sliding bearing instead of a rolling-element
bearing. The sliding elements are low-friction polymer pads that are arranged about a machined
steel flange at the tower top. This arrangement provides both support for the nacelle frame and
friction to hold the rotor in position, and the yaw drives have to be over-sized to overcome this
friction.
Since electric yaw drive motors are started and stopped with contactors, the yaw movement
is sudden and can subject the yaw drives to very high forces. At least one manufacturer is
supplying variable-speed yaw drives to allow a smooth start and controlled forces. Another
common challenge with the conventional arrangement is wear of the teeth on the stationary yaw
gear. Some manufacturers are installing automatic greasing systems to ensure that the gear teeth
are adequately coated with lubricant. One manufacturer has side-stepped this problem entirely,
and has designed a unique hydraulically driven yaw system that uses a series of cylinders and
linkages.
3.4.4 Sensors
The most critical sensors on a wind turbine are the speed sensors on the rotor and generator
shafts. These sensors are the primary defense against excessive turbine over-speed, which can
be extremely destructive. Other critical sensors include the wind speed and direction sensors,
vibration sensors, sensors that detect when the electrical cables from the nacelle in the tower
are wrapped too tightly, temperature sensors on components that are susceptible to temperature
extremes, and pressure and flow sensors for the cooling and lubrication fluid systems. Voltage,
current, and power sensors monitor the electrical system and utility grid.
3-23
10581090
Wind Power
The evolution of wind turbine sensors has followed that of standard industrial controls. Many
modern sensors provide digital communications on standard computer network protocols such as
CANBUS. Other than the evolution from analog to digital sensors, there are only a few
significant advances in the sensors used on wind turbines. One of these advances is the use of
ultrasonic wind measurement sensors. These sensors provide wind speed, direction and
temperature from a single unit with no moving parts and a single serial cable. Although more
expensive than conventional sensors, ultrasonic sensors offer improved reliability and simplicity
of installation. Ultrasonic sensors can be heated more effectively than standard sensors; however,
the data sensors may experience additional fouling in conditions with a high degree of dust or
precipitation.
3.4.4.1 Optical Strain Gages
A sensor that is being introduced into some of the larger, more complex turbines is the optical
strain gage. These sensors are built into the blades for condition monitoring and in conjunction
with blade pitch mechanisms for structural load alleviation. The long-term reliability of these
sensors has not yet been proven, but many major turbine vendors are experimenting with them
and it is possible that some have included them in production turbines.
3.4.4.2 Wind Measurements—LIDAR
Active research projects involving nacelle-mounted wind speed sensors use Light Detection and
Ranging (LIDAR) as one of the primary technologies under consideration. These optical remote
sensing units are designed to analyze the oncoming wind, enabling more informed control
decisions that will optimize power output and minimize structural loads. Typically, the wind
speed and direction are measured using sensors mounted on top of the nacelle, downwind of the
rotor. As a result, the wind speed and direction measurements slightly lag the actual wind
conditions at the rotor. Wind turbine rotor operating efficiency decreases by about 0.5% to 1%
for every degree of misalignment with the oncoming wind. Field testing is underway on LIDAR
systems that use lasers to measure the wind speed and direction up to 300 m in front of the rotor,
providing a 20-second advanced notice of the instantaneous wind speed and direction at hub-
height. This advanced notice provides sufficient time to orient the nacelle and adjust blade angles
to improve energy capture and also reduce peak loads. One LIDAR system manufacturer claims
that addition of the LIDAR system could increase energy capture by 10%. In addition, the
nacelle-mounted LIDAR systems use fiber optics instead mirrors and are less expensive, more
robust, and less susceptible to misalignment.
While the LIDAR forecasting systems do provide valuable wind information to the control
system, they add expense and complexity to the turbine. Nacelle-mounted LIDAR systems are
estimated to cost from $100,000-$200,000, which would be 7% to 10% of the turbine capital
cost, depending on turbine size. In addition, since measuring the inflow specific to one machine
would not be useful for improving rotor performance at a neighboring machine, each wind
turbine requires its own LIDAR system. Also, the addition of the nacelle-mounted LIDAR
hardware adds complexity and could negatively impact both reliability and maintenance costs.
3-24
10581090
Wind Power
3.4.4.3 Condition Monitoring
In an effort to properly maintain wind turbine equipment, minimize plant downtime and maximize
energy generation, modern wind turbines increasingly use condition monitoring (CM) systems that
provide diagnostic data on the physical state of the equipment to identify and prevent potentially
serious failures. Some current CM techniques include bore-scopes, blade monitoring, oil analysis,
vibration monitoring and manual field observation.
Different CM systems can be run while the wind turbine is on-line or off-line. By using on-line
CM systems, operators can monitor real-time operational data. Collected data can be used to
identify the environmental conditions that exist prior to CM events, as well as the resulting actual
loading impacts and specific gear failures. Some on-line CM methods include vibration analysis,
oil particle counting, and integrated gear and bearing analysis tied into a central monitoring
service. Off-line CM techniques require plant downtime, so they can be costly and labor intensive.
Off-line CM activities include strain and structural testing, recording torque and shear
measurements, and sampling oil and other turbine fluids. To keep costs down, when the turbine
off-line non-destructive CM testing techniques such as ultrasound, thermography and infrared
testing are used.
Improvements to CM systems are continuously being developed. Many of the new technologies
include remotely monitoring and responding to alarms, developing new algorithms to identify
flaws, and automating even more of the O&M activities. Additional information is provided in
the 2010 EPRI Report, Wind Turbine Blade Structural Health Monitoring, Methods and Benefits.
[69]
3.4.5 Controls
The primary objectives of a wind turbine controller are:

• Keep the turbine safe and within its design load envelope
• Maximize the turbine energy production
• Make supervisory decisions such as when to start, stop, or yaw the turbine
• Detect abnormal conditions such as high component temperatures and take the appropriate
action.
The controller achieves these objectives via a combination of sensors, software, hardware logic,
and actuators. The controller takes inputs from sensors and issues commands to actuators based
on a predetermined set of parameters. Controls may be divided into several subsystems as
follows:
• Main turbine controller
• Blade pitch system and controller(s)
• Generator/power electronics controller
• Safety system.
3-25
10581090
Wind Power
The main controller is responsible for most supervisory decisions, fault detection, and providing
input into the closed-loop controls of essential subsystems. The main controller hardware
consists of a central processor, input/output modules, and communication interfaces. These are
typically mounted in one or more racks in the nacelle and/or a cabinet in the tower base and are
linked with a communications cable in the tower. There is often a user interface panel at the
tower base for maintenance personnel to manually control the turbine.
Blade pitch controllers on pitch-regulated turbines typically accept pitch commands from the
main controller and run their own closed-loop control algorithms to rapidly meet the commanded
pitch positions. In addition to blade position commands, the pitch controllers report other data
back to the main controller such as status, motor voltage, current, and temperature. This
information is typically communicated digitally over slip rings. Some turbine manufacturers are
implementing independent blade pitch systems, with controllers for each blade. The primary
advantage of independent blade pitch is to ensure that any single-point failure within the turbine
will only disable one blade pitch system, allowing the other blade pitch systems to remain
functional and capable of stopping the turbine. Independent blade pitch also allows reductions
in the size and cost of mechanical braking systems, the benefits of which have an effect on other
turbine components such as the gearbox.
The generator and power electronics controls for variable-speed turbines are a subsystem similar
to the pitch controller. Their primary function is to accept generator torque commands from the
main controller and rapidly meet them. This controller also communicates status information to
the main controller over a digital communication link.
The turbine safety system is not part of the control system and is actually super-ordinate to
the main controller as a design guideline, which allows the safety system to override the main
controller if necessary to protect the turbine. The safety system is often a series of relays that are
connected to critical sensors such as rotor and generator speed, vibrations, utility voltage, and
the yaw system. When any of these sensors measures an alarm-level signal, the relays open
(or close) and trigger an emergency stop using blade pitch and/or mechanical brakes.
Wind turbine controls have followed the trends for controls in other industries, namely the
increase in performance and decrease in cost of electronics and computing power. In the early
days of wind turbine development, many of the controls used hard-wired logic with sensors and
relays and a minimum of actual computing power. Modern turbines use controllers based on
high-speed CPUs with a considerable amount of memory and high-speed communications from
well known vendors such as Intel and Motorola. Turbine vendors use a variety of hardware
depending on their preference. Many companies have vertically integrated with controller
vendors, others use off-the-shelf control hardware from independent vendors, and still others use
custom-developed controller hardware. Generally, each turbine manufacturer provides some
degree of proprietary software.
As turbines have grown larger in size, the relative per-unit cost of controllers has decreased;
however, their absolute cost has increased because of the greater complexity of most modern
turbines. The development time and costs for control hardware and software have also increased
considerably to support turbine control sophistication, plus the Supervisory Control and Data
Acquisition (SCADA) systems and other remote debugging and operations features that function
over the Internet.
3-26
10581090
Wind Power
The future development of wind turbine controls is primarily focused on the software and
algorithms used to operate, diagnose, and maintain the turbines. In particular, condition
monitoring features are becoming more common but require considerable sensors, algorithms,
bandwidth, and data storage.
3.4.6 SCADA Data Collection & Transmittal
SCADA systems are used on all commercial wind farms. A SCADA system provides
information on turbine operating status, alarms and warnings, and performance parameters such
as rotational speed, power output, and line current and voltage. All systems allow the ability to
remotely acknowledge and reset non-critical faults and to change some parameters in the control
system, such as maximum output power.
Most turbine manufacturers provide a SCADA interface that collects the information from each
turbine controller and displays the data in tabular and graphical format. Available displays
normally include an overall layout of the farm with indicators for the status of each turbine,
meteorological conditions, indicators for aggregate power output, total delivered megawatt-
hours, and event and alarm logs. Often the user can access information at the individual turbine
level such as gearbox temperature or pitch position.
Although wind turbine controllers update at a rate of around 20 Hz, SCADA systems usually poll
the data at a much slower rate, on the order of seconds. Historical data, such as wind speed and
power output, are stored as the average, minimum, maximum, and standard deviation over a 10-
minute period. Some systems allow storage of selected parameters at a higher rate. In order to
increase visibility of project performance, some owners are directly importing higher-resolution
data from the turbine controllers into a time-based enterprise system that allows more options for
data analysis. Several third-party SCADA are also available that can interface directly with the
turbine controller.
3.4.7 Drive Train
The wind turbine drive train performs two functions. First, it supports the weight of the rotor and
the thrust and yawing loads introduced by the blades. Second, it transmits the power-producing
torque loads from the rotor to the generator.
3.4.7.1 Drive Trains with Gearboxes
The vast majority of wind turbines are designed around a common drive train configuration,
which consists of a long main shaft and main bearing at the front of the nacelle, and a generator
at the rear of the nacelle. A gearbox is situated in the middle to increase the low rotation speed of
the rotor to match that of the generator. All of these components are laid out on the nacelle
frame, which is made from cast iron or fabricated steel or, often, a combination of both. In this
configuration, the high-speed shaft of the gearbox is connected to the generator with an
articulating coupling to compensate for inevitable misalignment. The low-speed section of the
gearbox is connected to the main shaft with a fixed compression-style shaft coupling.
3-27
10581090
Wind Power
There are two methods of mounting the gearbox to the nacelle frame. The most common is the
three-point design, where the main shaft is supported by one spherical rolling-element bearing at
the very front of the machine to carry the thrust and weight of the rotor, and the gearbox itself is
pinned to the frame on both sides to counteract the rotor torque and the overturning load from
the rotor. The second method for supporting the gearbox is referred to as the four-point
configuration. The main shaft is supported by two main bearings, one at the front and one at the
other end of the shaft, in front of the gearbox. The gearbox is still pinned to the nacelle frame on
both sides, but only to prevent the gearbox case from rotating; the overturning loads from the
overhang of the rotor are carried by the second main shaft bearing.
Because wind turbine generators are traditionally asynchronous (induction) machines connected
to the grid, they must operate at high speeds (e.g., 1200 or 1800 rpm) compared to the rotor.
This is true for variable-speed turbines also, although the speed range may vary up to 50% from
nominal. Since the rotor speed is usually between 10 and 20 rpm for a ,megawatt-class turbine,
the gearbox must then increase the rotor speed by anywhere from 60 to 110 times. Almost all
gearboxes are based on a three-stage design in order to achieve this high ratio and carry the
torque loads efficiently. Figure 3-13 shows an example of a typical three-stage gearbox with a
planetary low-speed stage and two parallel-shaft stages that all reside within a common gear
housing.
Figure 3-13
Gearbox Internal Schematic Showing One Planetary and Two Parallel-Shaft Stages
Wind turbine gearboxes have in general been a weak point of the turbine drive train. Much
research has gone into understanding why gearboxes fail and how gearbox life can be extended.
Improvements have been realized in design tools to analyze how the gears, bearings, shafts and
3-28
10581090
Wind Power
cases respond under load, and best practices for design manufacture and maintenance have
recently been standardized. New techniques for improving the carrying capacity of gearboxes
include flexible planet gear supports, integrated planet gear bearings, improved lubrication
delivery and super-finishing of gear teeth.
3.4.7.2 Hydrodynamic Fluid Coupling
One exception to the common gearbox configuration that has recently been introduced at a large
scale is the hydrodynamic fluid coupling, placed in the drive train between the rotor and the
generator. Figure 3-14 is a schematic of the “WinDrive” hydrodynamic coupling developed and
commercialized by Voith and in service on some DeWind turbine models.
Figure 3-14
Cross-Section of Hydrodynamic Drive System for a Wind Turbine (Source: Voith Brochure,
“The WinDrive – An Innovative Drive Train Concept for Wind Turbines,” Retrieved
September 2010 via www.voithturbo.com/windrive_publications.php3)
The hydrodynamic coupling combines the characteristics of a mechanically geared system with a
fluid/hydraulic system. It converts the continuously varying speed and torque of the rotor to a
constant-speed output to drive the generator. The fluid subsystem is continuously variable and
control is maintained through an adjustable guide vane within the hydrodynamic unit. In effect,
this system operates with a continuously variable gearbox ratio. While the addition of a
hydrodynamic coupling increases the complexity of the drive train relative to a conventional
drive train, this system allows the generator to operate at synchronous speed while the rotor turns
at variable-speed without sophisticated power electronics. In addition, in the United States, the
technology enables variable-speed operation without the need to pay licensing fees associated
with patents on variable-speed technology.
3-29
10581090
Wind Power
3.4.7.3 Direct Drive Train without Gearbox
Some turbine manufacturers have chosen to avoid gearboxes entirely and connect the rotor shaft
directly to a low-speed generator. Figure 3-15 is an electrical drawing of a direct-drive turbine.
As discussed in an earlier section, the generator speed is inversely proportional to pole count,
and direct-drive generators have multiple poles and are of large diameter. For a megawatt-class
turbine, manufacture and shipping of these direct-drive generators is expensive and a logistical
challenge. Several manufacturers have made a compromise with the “medium-speed” design,
using a single-stage gearbox unit with a generator configuration that is somewhere between
conventional and the direct-drive design.
SG
Figure 3-15
Electrical Diagram of a Typical Direct Drive Turbine
3.4.8 Foundation
The turbine foundation must support the weight of the turbine, and also prevent the turbine from
overturning due to the high thrust loads from the rotor. Foundation design is always specific to
the geological conditions at the site. The turbine manufacturer will provide the design peak and
fatigue loads to the foundation designer who will consider the appropriate soil conditions,
climate, and earthquake potential. The approval of a licensed civil engineer is normally needed.
The common types of foundations for wind turbines with tubular steel towers are reinforced
concrete structures. The turbine tower is usually bolted to the foundation with a circular pattern
of fasteners on the inside and outside of the tower. The spread-foot design is shaped like an
inverted “T” and uses a large-diameter pad that distributes the turbine over-turning loads over
a large area.
Another design that is common in North America consists of a hollow cylinder of pre-stressed
concrete that is sunk into an excavation and then back-filled with a slurry of soil, sand, and
cement. With this design, the foundation walls resist the over-turning loads by bearing against
the sides of the excavation. A much smaller quantity of concrete is required than for the spread-
foot design, but the excavation can be problematic in some geological conditions.
3-30
10581090
Wind Power
A foundation design used in Australia is based on multiple pre-stressed concrete pilings that
are arranged in a circle and joined to a foundation ring at the surface. Similar designs are used in
arctic regions with permafrost ground conditions. The steel pilings are installed by drilling a hole
deep into the ground, placing the piling, then backfilling the hole with a slurry of soil and water,
which then freezes and holds the piling in place. The pilings include thermal siphons that
circulate refrigerant solution to ensure the permafrost temperature is maintained. The base of the
tower is also elevated approximately 1.5 m above ground level to facilitate the dissipation of
heat.
3.4.9 Tower
3.4.9.1 Conventional Steel Towers
All wind turbine towers installed in North America today are fabricated towers. The tower is
fabricated from rolled steel plate formed into welded tubular sections with flanged ends that are
bolted together on site. The towers are painted inside and out. Tubular towers are preferred over
steel lattice towers for aesthetic and practical reasons – access to the nacelle is safer and more
convenient with a ladder mounted inside an enclosed tower, and the bottom of a tubular tower
serves as an enclosed location for switchgear and controls. However, the maximum height of
a tubular tower is limited by the diameter of the base section. Most modern wind turbines are
mounted on 60- to 80-m towers.
Currently, all tubular towers are fabricated in a manufacturing facility and shipped to the site by
rail or truck. This limits the base diameter to approximately 4.1 m. To overcome this limitation,
some very tall prototype turbines have been installed on lattice towers that can be assembled on
site. One company is introducing a unique version of a lattice tower that is wrapped in a fabric
skin, offering some of the protection and visual appeal of a tubular tower. Another company is
re-introducing a technology that was used for kilowatt-class turbines in years past, in which the
tubular steel tower is fabricated in longitudinal sections and assembled into tube sections on-site,
reducing the shipping size of the parts.
In Europe, another approach has been developed to deal with the need for taller towers and
the high cost of manufacturing and transporting large-diameter steel tower sections. Several
manufacturers are making at least the base section of the tower out of reinforced concrete. The
concrete section itself is made of short cylindrical sections that are cast in moulds and stacked
on-site. Given that concrete is also required for the foundation, this approach may prove to be
economically feasible in North America.
3.4.9.2 Concrete and Hybrid Towers
Some of the most innovative developments in wind turbine tower design over the past 10 years
feature the use of reinforced concrete. The main drivers for these innovations are the limitations
on base diameter of the traditional tubular steel towers (due to bridge clearances and other
transportation restrictions) and dramatically increasing steel prices worldwide in recent years.
3-31
10581090
Wind Power
Figure 3-16 shows several concepts of hybrid and concrete towers. There are many advantages to
using concrete for wind turbine towers. The cost of the cement and rebar needed to construct a
concrete tower is considerably less than the cost of steel for an equivalent structure. In addition,
unlike steel towers that must be shipped in tubular segments, concrete tower sections can be
either cast in situ (e.g., concepts H2 or H3) or pre-cast in arc segments that are manageably sized
to meet transportation requirements (e.g., concept H1).
Steel
tower
Pre-str.
concrete
section
H1 H2 H3 Pre-stressed
concrete
Hybrid tower concepts
Figure 3-16
Example Hybrid Steel/Concrete and Concrete Towers
The arc segments can then be combined to form circular sections that are stacked on each other
at the site. Concrete towers constructed in this way have no restriction on base diameter imposed
by external circumstances, so design optimization is possible. This has implications both for the
foundation design as well as tower height. Concrete towers can be not only wider at the base but
also taller than steel towers, potentially increasing the power output of a turbine. Also, the
maintenance on concrete towers is less than that for steel, which may need to be repainted
periodically to prevent corrosion. Finally, designing a concrete tower for longer life adds little to
the cost of construction.
3.4.9.3 Integrated Towers and Foundations
Because concrete towers or hybrids tend to have a wider base than steel towers, foundation
pressure is significantly lower than for a steel tower. This simplifies the foundation technology
and reduces the amount of required concrete by 70%. For example, a basic annular ring 1-m
thick may be sufficient for the foundation, depending on the soil conditions. The foundation for a
comparable steel tubular tower would need to be several meters thick, requiring a significantly
more complicated pour during construction.
3-32
10581090
Wind Power
3.4.9.4 Tall Towers
Tall towers (100 m and taller) reach potentially higher speed winds and increase annual wind
energy generation relative to shorter towers. They are primarily geared for the onshore wind
market, where the influence of the boundary layer is generally more pronounced than offshore.
The boundary layer is the layer of air between ground level, where the wind speed is near zero,
to the elevation above which the wind speed is no longer influenced by surface friction or
obstacles such as trees, uneven terrain and buildings. The height of the boundary layer varies but,
generally speaking, wind turbines operate within this layer. The variation in wind speed with
height above ground is often referred to as wind shear. When wind shear is positive, the taller a
wind turbine is, the higher the average wind speed. This effect is diminished over water, where
the surface friction is considerably lower than on land, and where there are few or no obstacles to
disturb the air flow.
The ability to erect taller towers for land-based installations can lead to significant increases in
annual energy production (AEP), given the appropriate wind conditions. A study by NREL
suggests that increasing tower height by 20 m can increase AEP by 20% for sites with positive
wind shear. Because of this, taller towers can potentially open up vast new areas to development
that were previously uneconomic. Also, as turbine blade length increases, taller towers are
needed to keep the entire rotor sufficiently elevated for good performance.
3.4.10 Offshore Foundations
One of the major differences between an onshore and an offshore wind energy installation is the
more complex foundation structures required offshore. Offshore, the foundations (or “sub-
structures”) are much larger since they have to extend up above the highest wave crest in order to
support the tower and the access platform at a safe distance from the impact of the waves.
Further, in addition to the wind loading, the offshore foundation/sub-structure is subject to
hydrodynamic loads and a harsher environment than onshore structures (with respect to
corrosion, for example).
It is therefore not surprising that the foundation/sub-structure is one of the elements of offshore
wind development where a lot of innovation is taking place. The two issues related to offshore
wind discussed in this section are innovation involving 1) installation of bottom-fixed offshore
foundations in water depths up to 60 m, and 2) floating foundations for deepwater sites with
depths of 60 to 600 m.
3.4.10.1 Bottom-Fixed Offshore Foundation Designs
Currently, nearly all commercial offshore wind projects are installed in water depths of up to
30 m with foundations fixed to the seabed. The most common foundation type for shallow depths
is the steel monopile foundation, consisting of a large diameter pile (about 5 m) which is
typically drilled or driven 25–35 m into the seabed. A transition piece is installed on the
monopile, which supports the wind turbine tower and the main working platform at a safe level
above the highest waves. The transition piece is typically fitted with accessories such as boat
3-33
10581090
Wind Power
landings, ladders, access platforms, and J-tubes for cables. Monopile foundations can be installed
with or without scour protection.
Other types of fixed foundations include concrete or steel gravity bases, which rest on top of the
seabed and rely on the weight of the structure to provide stability. The main platform at the
entrance to the wind turbine tower is typically an integrated part of the foundation structure. The
competitiveness of these foundations is governed by the soil conditions at the site and the distance
to the material and manufacturing locations. Compared to the monopile foundation, the gravity
foundation in general requires higher strength of the upper soil layers and a shorter distance to
material and manufacturing locations. Gravity foundations also usually require scour protection.
Bucket foundations are large-diameter hollow steel structures that are partially driven into the
subsea structure by suction and filled with soil and rock to stabilize the foundation. The bucket
foundation design is currently only in use by a Vestas V90 3.0-MW offshore turbine in the test
field in Denmark, erected in December 2002.
Multi-pile jacket foundations are the standard for offshore oil and gas platforms in waters deeper
than 30 m and in softer soils, and this technology is being transferred to the offshore wind
industry. REpower installed a 5-MW demonstration project on a four-pile jacket foundation at a
44-m water depth in 2007. BARD Engineering is planning to install a prototype 5-MW turbine
using the company’s patented Tripile foundation structure. Figure 3-17 shows examples of
foundations that can be use in transitional water depths between 30 and 60 m.
The construction of offshore foundations is a major part of the cost of an offshore wind
installation. The construction includes fabrication (normally done onshore), transportation to the
site, and installation at the site. Since the installation vessels are typically very expensive and the
installation process depends on the weather conditions, a lot can be gained from optimizing the
installation process. Such optimization can be achieved by innovative foundation design that
accommodates simple installations or by innovative installation processes and vessels.
One example of an installation method that optimizes both the design and the installation of
offshore foundations is the concept proposed by Seatower. [78] Seatower proposes a foundation
that can be towed to a site and lowered to the seabed without the use of a crane. The foundation is
reportedly designed so that it can be installed directly on the seabed without requiring either seabed
preparation or pile driving. It would be lowered to the seabed by pumping sand into a hollow
structure until it was stabilized by the combined weight of its components and ballast (i.e., it
becomes a gravity-based foundation). The concept is only briefly described in the literature, so it is
difficult to evaluate how viable the presented solution is. However, it shows that some
development work is taking place in the industry that has the potential to significantly reduce the
cost of installing offshore foundations.
3-34
10581090
Wind Power
Figure 3-17
Transitional Depth Foundations (30-m to 60-m Depths) (Source: “Offshore Wind
Technology” presented by Walt Musial, National Renewable Energy Laboratory, at AWEA
Offshore Wind Workshop, September 9, 2008)
3.4.10.2 Floating Foundation Concepts
Floating foundations offer another way to support offshore wind turbines and enable wind
installations in much greater water depths than the bottom-fixed structures. Different concepts
are being developed, but in general it is assumed that floating foundation solutions will mainly
be relevant for water depths greater than 50 m.
To date, there are three different design concepts (or a combination thereof) being pursued for
floating foundations, which are shown in Figure 3-18:
1) The ballast-stabilized or spar buoy concept is a long and slender structure with ballast at the
lower end which makes it float vertically. The structure is typically anchored to the seabed
with un-tensioned cables.
2) The mooring-line stabilized or tension leg concept is a floating structure that is pulled into
the water by pre-tensioned cables anchored into the seabed. The structure thereby floats
below the water surface and at a lower elevation than it would without the tension cables
(tension legs).
3) The buoyancy-stabilized or barge concept is a wide structure floating on the water and is
typically anchored to the seabed with un-tensioned cables.
3-35
10581090
Wind Power
Figure 3-18
Floating Foundations (greater than 60-m Depth) (Source: “Offshore Wind Technology”
presented by Walt Musial, National Renewable Energy Laboratory, at AWEA Offshore Wind
Workshop, September 9, 2008)
An NREL study [79] of the tension leg platform, spar buoy, and barge floating foundation
concepts compared the loading on a wind turbine installed on the different floating foundations
to that on a similar turbine installed onshore. In general, the wind turbines installed on floating
foundations experience higher loads, which are highest for a turbine installed on a barge due to
the relatively low bending stiffness of the barge/turbine system. The movements of the wind
turbine are therefore increased, resulting in higher loading of the components. No economic
comparison is included in the NREL study.
Floating foundations have not yet been used for large-scale wind projects but there are two
known demonstration projects (prototypes), one of which is still in operation. In 2007, the Blue
H tension leg concept was installed off the coast of Italy at a water depth of 108 m. An 80-kW
turbine installed on the prototype Submerged Deepwater Platform (SDP) was operated for only
one year in order to gather data on the offshore floating foundation concept. In 2009, the
Norwegian company Statoil completed the installation of the demonstration project Hywind. The
project consists of one 2.3-MW Siemens wind turbine with a ballast-stabilized floating
foundation installed off the west coast of Norway at a water depth of about 220 m. The
installation was completed in June 2009 and will run as a test case for at least two years [80].
3-36
10581090
Wind Power
3.5 Trends in the Turbine Supply Market
This section summarizes trends in the current turbine supply market. The primary source of
information is the wind energy World Market Update 2009 published by BTM Consult ApS in
March 2010.
A handful of manufacturers supply the majority of the world’s wind turbine demand. In terms of
market share, the leading wind turbine supplier is Vestas, based in Denmark, which captured
12.5% of the global market share in 2009. General Electric (GE) was the second largest supplier
(12.4%), followed by the Chinese manufacturer Sinovel in third position (9.2%) and Germany-
based Enercon close behind in fourth (8.5%)2.
All of the top suppliers are expanding their manufacturing facilities in emerging markets such as
India, China, and the United States. GE, Vestas, Suzlon, and Enercon are some of the major
suppliers that have established facilities in India. Suzlon Energy opened a major component
manufacturing facility in China in 2007, which manufactures blades, generators, nacelles, and
control panels. In blade manufacturing alone, new facilities all over the world have been
established to reduce shipping costs and logistics for these large components. In the United
States, blade factories in Colorado (Vestas), Iowa (TPI Composites) and Arkansas (LM
Glasfiber) were all built in 2008. Moreover, Germany-based Fuhrländer plans to expand its
presence in the United States with a manufacturing facility in Butte, Montana. All told,
manufacturers announced, added, or expanded approximately 50 U.S. facilities in 2008.
Consequently, 50% of turbine components installed in United States are now made domestically,
up from 30% in 2005. However, the global market downturn has slowed the speed at which
planned manufacturing expansion is ensuing. Figure 3-19 shows the locations of manufacturing
facilities for major turbine components in the United States.
In addition to expanded manufacturing facilities, several major wind turbine companies are
expanding their research and development capabilities in the United States. Siemens has
announced plans to develop a wind technology research and development center in Boulder,
Colorado, in cooperation with the National Renewable Energy Laboratory (NREL). Vestas has
announced plans to develop a wind research center in Houston, Texas.
Several of the smaller turbine manufacturers are gaining market share with a number of
noteworthy developments, though the recent economic downturn has somewhat dampened
expectations. Suzlon purchased a majority stake (85%) in the German turbine manufacturer
REpower in 2007, which increased Suzlon’s global market share from 10.5% to 13.8% (though
its market share shrunk to 9% in 2008 due to poor order inflow). The company purchased a
further stake in REpower in June 2009. Suzlon also purchased the world’s second largest
gearbox supplier, Hansen Transmissions, in 2006, giving the company greater control over its
sub-component supply chain. Ecotécnia, which operates five production facilities in Spain, was
acquired by the major power generation system supplier Alstom in October 2007, representing
Alstom’s first move into the wind energy sector.
2
BTM Consult ApS. March 2010. “International Wind Energy Development World Market Update 2009” (pg 28)
3-37
10581090
Wind Power
T Bailey Inc.
LM Glassfiber
DMI Industries
Tower Tech TPI

Knight & Carver
Suzlon Clipper
Acciona
TPI Siemens
Gamesa
NEW!
T Bailey Inc. Hendricks
Trinity
Vestas Structural Tower
NEW!
General Electric
DMI Industries
Ameron International
Trinity
Knight & Carver Structural Tower Aerism
LEGEND:
General Electric
LM Glassfiber
MFG
Berger NEW! Blade Manufacturer
Southwest Steel
Trinity
Structural Tower Beaird Co.
Tower Fabrication
DeWind Turbine Assembly
Figure 3-19
Manufacturing Plants for Turbine Blades and Other Components (Source: Presentation
from Tom Ashwill at 2008 Sandia Blade Workshop, Sandia National Laboratories,
Albuquerque, NM)
In 2007, U.S.-based manufacturer Clipper Windpower began serial production of the 2.5-MW
Liberty turbine, shown in Figure 3-20. The turbine drive train consists of a forged steel main
shaft integrated with the gear housing and a bull gear that drives four permanent-magnet
generators arranged around the housing. This configuration eliminates the conventional three-
stage gearbox and spreads the torque loads across the four pinion gears. The drive train is
matched with advanced power electronics allowing for variable-speed operation and providing
full power quality management capability.
China hosts a number of new entrants into the turbine supply market, and turbine manufacturing
capacity in that country is expected to exceed 10 GW per year by 20103. The top manufacturers
with the greatest market share in China are Goldwind, Sinovel, and Dongfang Wind Co. Ltd. In
2007, Goldwind became the first alternative energy company in China to be listed on the
Shenzhen Stock Exchange. Sinovel is now the largest Chinese turbine manufacturer, delivering
3,510 MW in 2009. Goldwind installed 2,721 MW in 2009. Although Goldwind and Sinovel
have not yet exported their turbines to the United States, developer GreenHunter placed the first
3
Milford, Edward. July-August 2008. “Record Growth for Wind”. Renewable Energy Magazine Volume 11, Issue
4, Pg. 42.
3-38
10581090
Wind Power
order of Chinese-made turbines for delivery outside of China, ordering 108 1.5-MW turbines
from Mingyang Wind Power Technology4.
Figure 3-20
Clipper Windpower’s 2.5 MW Liberty Series Wind Turbine Nacelle, with Four Modular
Permanent-Magnet Generators (Source: Clipper Windpower Inc.)
The industry has seen a trend in the globalization of wind turbine technology via licensing of
manufacturing rights. Notably, American Superconductor Corp. (ASC), known for its high-
temperature superconducting products, entered the wind turbine supply market with the
acquisition in 2007 of Austria-based Windtec, a designer and licenser of complete wind turbine
systems or subcomponents. ASC sells Windtec’s proprietary designs to third parties for an initial
fee plus royalty payments for each installation. Purchase of the ASC design allows new
companies to quickly enter the turbine supply market with a tested and certified product. ASC
has sold turbine design licenses to several manufacturers including Sinovel (China), AAER
(Canada), Wikov (Czech Republic), and Fuhrländer (Germany).
3.6 Trends in Wind Turbine and Plant Sizes
The average turbine size has steadily increased with technological advances such as improved
blade manufacturing technology, more sophisticated controls, and power electronics. Globally,
the average size of individual wind turbines installed in 2008 was 1,560 kW, an increase of 66
kW since 2007. There was almost no growth in the average turbine size in 20095, when 82% of
individual turbines installed were in the 1,500 to 2,500 kW range, and only 12.0% were in the
750-1,499 kW range. Less than 7% of installations used turbines outside of these capacity
4
Wise and Bolinger. May 2008. Annual Report on U.S. Wind Power Installation, Cost and Performance Trends:
2007. U.S. Department of Energy Lawrence Berkeley National Laboratory.
5
BTM Consult ApS. March 2010. “International Wind Energy Development World Market Update 2009” (pg 23).
3-39
10581090
Wind Power
ranges. In California, many of the projects that utilize kilowatt-scale turbines installed in the
1970s and 1980s are being repowered with modern megawatt-scale turbines, with each modern
turbine replacing up to 10 of the original turbines in some cases.
The average wind turbine size is expected to continue to increase in the short term. However, due
to the logistical constraints of transporting and lifting large turbine components, it is expected
that land-based turbine designs will be limited to rotor diameters of about 100 m to 110 m, which
corresponds to generating capacities of approximately 3 to 5 MW.
Table 3-3 summarizes selected turbine models over 2.3 MW in capacity in various stages of
development. This is not a comprehensive list of commercially available turbines in the market
but rather an indication of the next generation of large wind turbines expected to be widely
deployed for onshore projects in North America.
Table 3-3
Selected Onshore Wind Turbine Suppliers and Turbine Models over 2.3 MW
Rotor
Rating Control
Country Manufacturer Model Diameter Status
(kW) Scheme
(Meters)
100/109/1
Spain Acciona AW-3000 3000 Prototype 2008
16
Alstom
Spain ECO100 3000 100.8 PH, DFIG Commercial
Ecotecnia
Liberty C89 2500 89 Commercial
Clipper Liberty C93 2500 93 Commercial
USA HSP, PMG
Windpower Liberty C96 2500 96 Commercial
Liberty C99 2500 99 Commercial
Germany Enercon E-82 3000 82 DD, SG Commercial
General Commercial 2008 (not
USA GE 2.5xl 2500 100 PH, PMG
Electric (GE) available in US)
MWT92/2.4 2400 92 Commercial
Japan Mitsubishi MWT92/2/4 2400 95 PH, DFIG Commercial
MWT-/2.8 2400 N.A. N.A.
N90/2500 2500 90 PH, DFIG Commercial
Germany Nordex
N100/2500 2500 100 Prototype 2008
Germany Siemens SWT-2.3-101 2300 101 PH, AG Commercial
SWT-2.3-Mk II 2300 93 Commercial
Germany Vensys Vensys 90 2500 90 DD, PMG Prototype 2008
Vensys 100 2500 100 Prototype 2008
Denmark Vestas V90-3.0MW 3000 90 PH, DFIG Commercial
WWD-3/90 3000 90 Commercial
Finland WinWind PH, PMG*
WWD-3/100 3000 100 Commercial
PH = Planetary/Helical HSP = Helical Split Path TSP = 3-Stage Planetary DD = Direct Drive .
AG = Asynchronous Generator PMG = Permanent Magnet Generator PMG* = Permanent Magnet Medium-Speed Generator.
SG = Synchronous Generator DFIG = Doubly-fed Induction Generator.
3-40
10581090
Wind Power
Most of the listed models are designed for onshore applications; however, some models include a
marinized version for offshore deployment. Notable developments in the land-based turbine
market include the commercial production of the GE 2.5xl turbine, now GE’s largest onshore
model. Additionally, Enercon has developed a 3-MW direct-drive turbine based on its original 2-
MW E-82 model. Enercon has tested the prototype and plans to install 67 units at the Dutch
Eemshaven project.
In addition to an increase in individual turbine rated capacities, the land-based turbine market has
experienced an increase in the overall size of a wind power plant. The average size of wind
projects installed in 2007 was 120 MW, roughly double that during the 2004 to 2005 period and
6
nearly quadruple that during 1998 to 1999 . Currently, the largest wind plant in operation is the
735 MW Horse Hollow plant in Texas; however, a number of gigawatt-scale plants are under
development with expected operation beginning in 2010 and beyond.
3.7 Onshore Capital and O&M Cost Trends
The U.S. wind industry has seen a dramatic increase of turbine and construction pricing. The
average cost of a 2-MW machine, including turbine, blades, towers and transportation, has
increased by approximately 87% since 2005. Primary drivers for the cost increases include steel
and copper material prices, the weakened U.S. dollar, strong market demand for turbines,
constrained component supply chain, and a shift toward profits over market share from some
manufacturers. To a lesser extent, increased fuel and labor costs and limited high-capacity crane
availability have placed upward pressure on transportation and construction costs. Component
supply will continue to prove a major bottleneck in the next few years. With some components,
such as precision bearings, the global wind industry has to compete with equally strong demand
from other booming industrial and related sectors7. Looking forward, supply and demand will
remain a key driver in the rising costs of turbines and balance-of-plant construction8.
Figure 3-21 presents a breakdown of project costs by category, based on documentation of costs
from actual projects constructed in 2006 and 2007. The turbine, tower, and shipping costs are by
far the highest-cost components, contributing over 70% of the total project costs.
6
Bolinger, Wiser. “Surpassing Expectations: State of the U.S. Wind Power Market”. Renewable Energy World.
September 4, 2008.
7
De Vries, Eize. Renewable Energy World. July 2008.
8
Andrew Fowler, June 2, 2008. AWEA Windpower 2008 Presentation, “Trends in Wind Power Prices B.O.P. and
Turbine Costs” Renewable Energy Systems Americas (RES).
3-41
10581090
Wind Power
Balance of Plant
(interconnection, Financing/ Legal
substation) Fees
5% 4%
Construction/
Installation
14%
Development
Activity (design/
engineering/
permitting)
Wind turbine
4%
(nacelle, tower,
rotor, shipping)
73%
Figure 3-21
Breakdown of Estimated Capital Costs for Land-Based Wind Plants
The technology advances of turbine technology discussed in the previous section are expected to
make an impact on the cost of wind turbines as the new technologies are implemented. Table 3-4
summarizes the expected technological advances in turbine technology and the range of
associated increases in capital cost and energy production. This information is based on an
analysis completed by the DOE in order to determine the feasibility of achieving 20% of the U.S.
electric supply mix from wind power by the year 2030 and the technological advances that
would be required. The analysis includes certain assumptions about future market conditions,
such as the long-term extension of the federal production tax credit, which would create a stable
market environment in which companies would be willing to invest in the required research and
development activities. It is expected that most of the listed improvements would not be attained
in the short term. Enlarged rotors are likely to provide the most improvements between now and
2020, whereas the advanced tower concepts may be developed later than the 2030 time frame.
Operations and maintenance (O&M) costs are highly variable and depend on a number of factors
such as the O&M strategy employed, the reliability of the equipment, the operating environment,
and the roles and responsibilities of the equipment manufacturer in providing service and
warranty repairs. In the traditional O&M model, the turbine supplier provides warranted service
and repairs for a fixed fee during the first few years of the project. Five to eight years ago, five-
year warranties were available in the market; however, the trend in recent years among turbine
suppliers is to reduce the term to two years and to eliminate operations tasks from the contract.
Options are still available to extend the warranty period beyond the initial two-year term, but
owners of large fleets tend to take over all turbine service, repairs and operations at the close of
the warranty period. Owners that do not have large fleets have also been turning to third-party
operators to service and repair their projects.
3-42
10581090
Wind Power
Table 3-4
Areas of Potential Technology Improvement
Cost Increments
(Best/Expected/Least, Percent)
Technical Area Potential Advances
Annual Energy Turbine
Production Capital Cost
* Taller towers in difficult locations

Advanced Tower * New materials and/or processes
+11/+11/+11 +8/+12/+20
Concepts * Advanced structures/foundations
* Self-erecting, initial, or for service
* Advanced materials
* Improved structural-aero design
Advanced (Enlarged)
* Active controls +35/+25/+10 -6/-3/+3
Rotors
* Passive controls
* Higher tip speed/lower acoustics
* Reduced blade soiling losses

Reduced Energy Losses * Damage-tolerant sensors
+7/+5/0 0/0/0
and Improved Availability * Robust control systems
* Prognostic maintenance
* Fewer gear stages or direct drive

* Medium/low speed generators
* Distributed gearbox topologies
Drive train
* Permanent-magnet generators
(Gearboxes and
* Medium voltage equipment +8/+4/0 -11/-6/+1
Generators and Power
* Advanced gear tooth profiles
Electronics)
* New circuit topologies
* New semiconductor devices
* New materials
* Sustained, incremental design

Manufacturing and and process improvements
0/0/0 -27/-13/-3
Learning Curve * Large-scale manufacturing
* Reduced design loads
Totals +61/+45/+21 -36/-10/+21

Source: 20% Wind Energy by 2030: Increasing Wind Energy’s Contribution to U.S. Electricity Supply, DOE, May 2008.
DOE/GO-102008-2578
3-43
10581090
Wind Power
3.8 Developments in Offshore Wind Technology
As of the end of 2009, the worldwide installed offshore wind power capacity was 2,112 MW,
located primarily in waters off the coast of Denmark, the United Kingdom, Ireland, Sweden and
9
the Netherlands . An additional 1,500 MW is currently under construction, primarily in Germany
and the UK, and BTM Consult ApS estimates there will be just under 13,500 MW of offshore
wind capacity installed between 2010-201410.
There are currently no offshore wind farms installed in the United States. The vast onshore wind
capacity that has yet to be developed and the high cost of offshore development have prevented
significant interest and investment in offshore wind. Other challenges to offshore wind
development in the United States include the lack of specialized equipment for transporting and
erecting offshore turbines and a traditional lack of a defined permitting process.
Some of the major challenges to offshore wind development are beginning to be addressed,
however, and commercial interest appears to be on the rise. For example, in April 2009, the
Department of the Interior finalized a long-awaited framework for renewable energy production
on the U.S. Outer Continental Shelf (OCS). The framework establishes a program to grant leases,
easements, and rights-of-way for orderly, safe, and environmentally responsible renewable
energy development activities, such as the siting and construction of offshore wind farms, on the
OCS. In addition to establishing a process for granting leases, easements, and rights-of-way for
offshore renewable energy development, the new program also establishes methods for sharing
revenues generated from OCS renewable energy projects with adjacent coastal states.
The Interior Department’s Minerals Management Service (MMS) and the Federal Energy
Regulatory Commission (FERC) cleared the way for the publication of these final rules by
signing an agreement in April 2009 that clarifies their agencies’ jurisdictional responsibilities for
leasing and licensing renewable energy projects on the OCS. The MMS has exclusive
jurisdiction with regard to the production, transportation, or transmission of energy from non-
hydrokinetic renewable energy projects, including wind and solar. FERC will have exclusive
jurisdiction to issue licenses for the construction and operation of hydrokinetic projects,
including wave and current, but companies will be required to first obtain a lease through MMS.
Regulatory activity and other developments are helping to fuel project work. There were more
than 2,000 MW of wind plants under consideration in the United States as of September 2009,
and DOE estimates 54,000 MW of electricity could come from offshore wind farms in the
11
United Sates by 2030 . Proposed offshore projects in the United States are primarily
concentrated in the northeast and mid-Atlantic regions, where land-based wind development is
9
“Offshore Wind Technology” presented by Walt Musial, National Renewable Energy Laboratory, at AWEA
Offshore Wind Workshop, September 9, 2008.
10
BTM Consult ApS. March 2010. “International Wind Energy Development World Market Update 2009” (pg 64)
11
“Offshore Wind Technology” presented by Walt Musial, National Renewable Energy Laboratory, at AWEA
Offshore Wind Workshop, September 9, 2008.
3-44
10581090
Wind Power
limited due to the high population density and limited wind resource. Fewer constraints on noise
and visual impact, in addition to the higher quality wind resource and relatively close proximity
to significant electric demand along the coast, contribute to the appeal of offshore wind in the
northeast.
3.8.1 Trends in the Offshore Turbine Supply Market
The primary differences between offshore and onshore wind turbines are size and foundation
requirements. Due to the high cost of offshore wind turbine foundations and undersea electric
cables, offshore wind turbines are typically larger than their onshore counterparts in order to take
advantage of economies of scale. In addition to the difference in size, offshore wind turbines
have been modified in a number of ways to withstand the corrosive marine environment, such as
implementation of a fully-sealed or positive-pressure nacelle to prevent corrosive saline air from
coming in contact with critical electrical components, structural upgrades to the tower to
withstand wave loading, and enhanced condition monitoring and controls to minimize service
trips.
Vestas and Siemens have historically supplied the bulk of the offshore turbine market, with
Vestas holding a cumulative operational offshore market share of about 60%. Vestas introduced
the V90 3.0-MW to the commercial market in 2005, and well over 500 units are now deployed
both onshore and offshore. Siemens installed the first 25 of its 3.6 flagship machines at the
Burbo Bank wind farm in the UK in 2007. These turbines, for both onshore and offshore
application, were also installed at two offshore projects in the UK: one at Inner Dowsing and one
at Lynn. General Electric also entered the commercial offshore market with the installation of
several 3.6-MW turbines off the coast of Ireland in 2004, but has since discontinued production
of offshore turbines12.
There are several new entrants into the commercial offshore market. German-based
manufacturers REpower, Bard Engineering, and Multibrid have installed pilot projects using 5-
MW offshore turbines, and each company expects to manufacture up to 100 turbines per year by
the end of the decade13. In 2008, Enercon reports installation of two prototype 6-MW turbines
with 126-m rotor diameters in Emden, Germany14. WinWinD of Finland, a near-shore wind
entrant, installed five of its 3-MW WWD-3 Multibrid-type wind turbines at the Kemi Ajos
Harbour wind farm in Finland15.
Other companies report research and development activities of multi-megawatt class offshore
turbines. In 2007, Clipper Windpower announced plans to develop the world’s largest wind
turbine at 7.5 MW to be tested in Blyth, UK. Gamesa has announced a 4.5-MW turbine that will
feature the world’s largest rotor diameter at 128 m. Sinovel Wind Co. of China is undertaking
research and development on 2-, 3-,and 5-MW wind turbines, for both onshore and offshore
application. Additionally, Germany-based Nordex has commenced a concept study for a new
12
“40,000 MW by 2020: Building offshore wind in Europe”. Renewable Energy World. January 3, 2008.
13
Renewable Energy World. July 2008, Pg 104.
14
Enercon website: http://www.enercon.de/www/en/nachrichten.nsf.
15
http://www.abb.com/cawp/seitp202/261428146168c203c12573bf002f3466.aspx.
3-45
10581090
Wind Power
3- to 5-MW offshore wind turbine, with a prototype envisaged for 2010 and commercial
production planned to begin by 2011 or 2012.
Table 3-5 summarizes some of the commercially available and prototype wind turbines in the
offshore market.
Table 3-5
Selected Offshore Wind Turbine Suppliers16
Rotor
Rating Control
Country Manufacturer Model Diameter Status
(kW) Scheme
(Meters)
Onshore
BARD Prototype 2007,
Germany VM – 5 MW 5000 122 PH, DFIG
Engineering Offshore
Prototype 2008
USA Clipper Brittania 7500 140 to 160 N.A. Design Stage
Netherlands DarWinD DD115 4700 115 DD, PMG Protoype 2009
Spain Gamesa G-128 4500 N.A. N.A. N.A.
General Shelved
USA GE 3.6sl 3600 111 TSP, DFIG
Electric (GE)
Onshore
Germany Multibrid M5000 5000 116 PH, PMG*
Prototype 2005
3000 to Prototype
Germany Nordex N- N.A. N.A.
5000 2010/2011
REpower Commercial
3300 104
Germany REpower 3.3M TSP, DFIG
REpower 5M 5000 126 Prototype 2008
Germany Siemens SWT-3.6-107 3600 107 VS, VP Commercial
SW-90-3500 N.A.
3500 90
DL
Norwegian Scanwind DD, PMG
SW-100-3500 N.A.
3500 100
DL
China Sinovel SL 3000 3000 N.A. N.A. N.A.
Denmark Vestas V120-4.5MW 4500 120 PH, DFIG N.A.

PH = Planetary/Helical HSP = Helical Split Path TSP = 3-Stage Planetary DD = Direct Drive.
AG = Asynchronous Generator PMG = Permanent Magnet Generator PMG* = Permanent Magnet Medium-Speed Generator.
SG = Synchronous Generator DFIG = Doubly-fed Induction Generator.
16
Source: “Offshore Wind Technology” presented by Walt Musial, National Renewable Energy Laboratory, at
AWEA Offshore Wind Workshop, Sept 9, 2008.
3-46
10581090
Wind Power
3.8.2 Offshore Transmission Technology Status
Currently, offshore wind farms are installed at distances from shore ranging from 0.8 km to
20 km. Undersea cables connect the wind turbines within a project to an offshore substation
and from the substation to the mainland. Most offshore wind farms utilize high-voltage ac
transmission lines to transmit power from the offshore substation to the mainland. High-voltage
dc (HVDC) transmission is a new technology that experiences lower electrical line losses than
high-voltage ac; however, rectifier and inverter losses are introduced when converting from ac to
dc at the offshore substation and from dc back to ac at the onshore grid-connection point. The
lower line losses are expected to outweigh the additional electrical conversion losses and cost
differential only for projects located a significant distance from shore. In Germany, one HVDC
cable connection to an offshore wind project is near completion and contracts have been awarded
for three more HVDC-wind projects.
3.8.3 Offshore Capital and O&M Cost Trends
Costs for offshore wind projects are not well established in the United States, and depend
significantly on the water depth, distance from shore, and type of support structure. Available
estimates based on experience in Europe indicate a 50% to 150% increase in installed cost
17
compared to land-based projects . Factors contributing to the higher cost of offshore projects
include the higher cost and complexity of foundations, underwater power cables and
connections, marinization of the wind turbines, the difficult conditions for construction, and
specialty equipment required for construction.
One of the primary contributors to the high cost of generating electricity offshore is the cost of
collecting and transmitting the power to shore. Figure 3-22 shows the capital cost distribution
for selected UK wind plants. It reveals that the power collection system and transmission line
to shore account for 21% of the total project cost. As discussed previously, the distribution lines
and interconnection for land-based projects constitute only about 5% of the total project cost.
The cost of the support structure, which is a significant portion of the total project cost, depends
on the water depth, with deep-water floating technology more expensive than shallow-water
fixed technology.
Operations and maintenance costs are also expected to be higher for offshore wind projects than
for onshore projects due to limited access to the turbines during inclement weather conditions,
the specialized equipment needed to access the turbines, and specially trained maintenance
personnel. However, there is no track record of long-term operation of offshore wind turbines,
even in Europe, and long-term O&M estimates are highly uncertain.
17
(“Offshore Wind Energy: Economic Opportunities” presented at AWEA Offshore Workshop 2008.)
3-47
10581090
Wind Power
Figure 3-22
Breakdown of Estimated Capital Costs for Selected Installed Offshore UK Wind Plants
(Source: Study of the Costs of Offshore Wind Generation-A Report to the Renewables
Advisory Board and DTI, Offshore Design Engineering (ODE) Limited, Sept 14, 2006.)
3.9 Technology Performance and Cost Tables
This section addresses the performance and capital, O&M, and levelized cost of electricity
estimates for five representative utility-scale onshore and one offshore wind power plants. The
estimates are based on data presented in the 2009 EPRI report, Engineering and Economic
Evaluation of Utility-Scale Wind Power Plants. [36]
3.9.1 Site Assumptions
Tables 3-6 and 3-7 summarize the site and wind plant assumptions for the six sites. The site
conditions are typical of those for wind projects located in California, Texas, Michigan, and New
York State and an offshore site near New York. There is also a base-case onshore plant located at
a site with a very high-quality wind resource. The performance and cost estimates assume the
onshore wind projects are built on moderate to easy terrain and the offshore plant is built on flat
ocean bottom suitable for a monopile foundation and with water depth less than 30 meters. The
meteorological and wind resource data in Tables 3-8 and 3-9 are typical of wind projects located
in each region. The onshore wind projects use 2.0-MW horizontal-axis variable-speed wind
turbines with doubly-fed induction generators and mounted on 80-m towers. The offshore wind
project uses 3.6-MW turbines mounted on 100-m towers (above water), and the project is located
15 km offshore. The wind plants are developed and operated by a knowledgeable, experienced
utility or development/operating company.
3-48
10581090
Wind Power
Table 3-6
Wind Plant Site Assumptions
Geotechnical Conditions
Rated
Rated
Thermal Foundation Capacity
Site Site Terrain Soil Type Capacity
Resistivity Type Density
(MW)
(MW/km2)
Perfectly Rippable
Base Low Mat 8.5 100
Flat but strong
CA Rough Rocky Medium Mat 6.3 100
Rippable Rock
TX Flat Medium 6.5 200
but strong Anchor
Aggregate
MI Flat Farmland High 2.6 150
Pier
Rippable
NY Rolling Hills Medium Micropile 3.6 50
and weak
Flat Ocean
Bottom Suitable
NY Offshore — Monopile 3.6 201.6
for Monopile
Foundation
Table 3-7
Wind Plant Design Assumptions
Turbine WTG
Rated Hub Rotor Crosswind
Capacity Height Diameter Separation WTG Downwind
Site (MW) (m) (m) (rotor dia.) Separation (rotor dia.)
Base 2.0 80 90 3 10
CA 2.0 80 90 3 10
TX 2.0 80 90 2.5 8
MI 2.0 80 90 4 11
NY 2.0 80 90 3 10
NY 3.6 100 110 2 7
3-49
10581090
Wind Power
Table 3-8
Assumed Project Site Weather Conditions
Air Seasonal Average Temps. (˚F) Max. Min.

Elev. Dens. High Low Rel.
Sum- Win- Temp Temp Humid.
Site (ft) (km/m3) Spring mer Fall ter (˚F) (˚F) (%)
1 150 1.2108 59.3 71.1 62.7 47.7 112 17 56.3

2 2,500 1.1306 63.0 80.7 63.3 44.3 113 -11 44.4
3 750 1.1900 31.7 62.7 58.7 28.7 101 -23 78.2

4 1,500 1.1643 42.3 65.6 47.0 20.3 100 -40 69.2
5 0 1.2160 46.2 66.2 55.7 42.3 90 7 84.3
Table 3-9
Assumed Project Site Wind Regimes
Annual Average Weibull C Weibull K

Site Wind Speed (m/s) Std. Deviation Parameter (m/s) Parameter
1 7.45 3.68 8.09 2.15
2 8.15 2.50 9.04 3.61

3 7.27 2.54 8.13 3.14
4 7.37 2.86 8.28 2.79
5 9.19 3.41 10.3 2.94
3.9.2 Plant Performance
Plant performance is measured by the annual capacity factor, which is the annual kilowatt-hour
energy production as a fraction of the energy that would be produced if plant operated at its rated
capacity 100% of the time. For wind plants, it is a function of average on wind speed, wind
speed frequency distribution, energy losses due to planned and unplanned outages and electrical
losses, soiling and icing of wind turbine blades, wake turbulence, and other losses. As described
in Reference 36, the annual capacity factors presented in this report were developed by analyzing
performance data from modern wind farms that were installed no earlier than 2004. The
assessment also excluded wind farms that had less than six months of available operating data.
Ventyx Velocity Suite was used to gather the data.
The resulting annual capacity factor estimates for the five conceptual onshore and one offshore
wind plants are:
• Base Case: 35%
• California: 33%
• Texas: 31%
3-50
10581090
Wind Power
• Michigan: 28%
• New York: 29%
• New York Offshore: 45%
The base-case capacity factor was chosen to be representative of a realistic but reasonably
productive wind farm.
3.9.3 Total Capital Requirement
Table 3-10 summarizes the Total Capital Requirement estimates for the six conceptual wind
plants. The reference date of the cost estimates is the fourth quarter of 2009.
The Total Capital Requirement consists of the Total Plant Cost, escalation and interest during
construction, and owner’s costs. The Total Plant Cost includes wind turbine procurement and
transportation, balance-of-plant direct labor and material, and indirect and EPC construction
costs. The Total Plant Investment is the sum of Total Plant Cost and escalation and interest
during construction (also known as allowance for funds used during construction. or AFUDC).
Total Capital Requirement is the sum of Total Plant Investment and owners costs (due diligence,
permitting, legal, development, taxes and fees). Larger projects generally benefit from
economies of scale and this is reflected in the cost estimates.
The Total Capital Requirement estimates for the onshore wind projects range from about
$2,000/kW for the 200-MW Texas wind plant to about $3,200/kW for the 50-MW New York
plant. The New York project is the most expensive of the onshore wind projects, mainly due to
diseconomies of scale and higher development cost. Wind turbine procurement and shipping cost
contributes 56% of Total Plant Cost in New York and 76% in Texas.
For the 201.6-MW offshore wind plant, the Total Capital Requirement is about $4,200/kW and
wind turbine procurement and shipping contribute 48% of Total Plant Cost. The higher cost
reflects the high complexity and cost of erecting offshore foundations, towers, nacelles,
underwater cables and other major components relative to onshore wind projects.
3.9.4 Operation and Maintenance Cost
Table 3-11 presents the Operation and Maintenance cost estimates for the six conceptual wind
plants. The components of the O&M costs include operating and maintenance labor and material,
insurance, and land usage (royalty/land lease fees). The O&M costs are presented for two
situations: one in which the O&M services are provided by the plant owner, and one in which the
services are provided by a contract services company.
3-51
10581090
Wind Power
Table 3-10
Total Capital Requirement Estimates (4th Quarter 2009 $)
Site #1 Site #2 Site #3 Site #4 Site #5 Site #6

Base NY
Site Location Case California Texas Michigan New York Offshore
Rated Capacity (MW) 100 MW 100 MW 200 MW 150 MW 50 MW 201.6 MW
Wind Turbine Generator Cost

Wind Turbine Procurement and Shipping $1,570 $1,570 $1,560 $1,590 $1,620 $1,997
Direct Construction Costs

Site Work $25 $153 $22 $38 $49 $0
Concrete $109 $161 $95 $151 $459 $587
Wind Turbine Generator Erection $73 $114 $72 $94 $99 $94
Electrical $68 $90 $87 $117 $97 $470
Electrical Collection Substation $36 $40 $29 $32 $55 $94
Miscellaneous Construction Indirects $7 $11 $0 $0 $0 $0
O&M Building $6 $6 $3 $4 $12 $16
Met Tower $4 $4 $4 $4 $4 $0
Total Direct Construction Costs $328 $579 $318 $446 $793 $1,260
Indirect Construction Costs

Engineering $10 $11 $6 $7 $18 $47
Construction Management Staff/Expenses $13 $15 $9 $10 $22 $117
EPC Contractor Profit (15%) $53 $77 $52 $63 $121 $214
EPC Contractor Contingency (10%) $35 $52 $35 $42 $80 $142
Total Indirect Construction Costs $110 $155 $102 $122 $241 $521
3-52
10581090
Wind Power

Total Capital Requirement Estimates (4th Quarter 2009 $, Continued)
Base NY
Site Location Case California Texas Michigan New York Offshore
Total Plant Cost
Total WTG, Direct, and Indirect Costs $2,008 $2,304 $1,980 $2,158 $2,654 $3,778
Recommended Contingency (10%) $44 $73 $42 $57 $103 $178
Total Plant Cost $2,051 $2,378 $2,022 $2,214 $2,757 $3,956
Total Plant Investment

Escalation and Interest During Construction $0 $0 $0 $0 $0 $0
Total Plant Investment $2,051 $2,378 $2,022 $2,214 $2,757 $3,956
Total Capital Requirement

Development Fee For Purchasing the Project
From a Developer. $50 $200 $25 $67 $400 $164
Recommended Owner Contingency (10%) $5 $20 $3 $7 $40 $16
Total Capital Requirement $2,106 $2,598 $2,049 $2,288 $3,197 $4,137
3-53
10581090
Wind Power
Table 3-11
Operation and Maintenance Cost Estimates (4th Quarter 2009 $)
Site #1 Site #2 Site #3

Site Location Base Case California Texas
O&M Provider Plant Contract Plant Contract Plant Contract

Greenfield Greenfield Greenfield Greenfield Greenfield Greenfield
Site Site Site Site Site Site
Rated Capacity 100 100 100 100 200 200
Number of turbines 50 50 50 50 100 100
Number of personnel or FTEs 14 0 14 0 26 0
Hours Per Year 3,154 3,154 2,803 2,803 3,154 3,154
Reference Date of Cost Estimates 4th Q 2009 4th Q 2009 4th Q 2009 4th Q 2009 4th Q 2009 4th Q 2009
Operation and Maintenance Costs, $k/yr

Labor $841 $0 $1,330 $0 $1,602 $0
Maintenance $986 $1,770 $1,560 $2,800 $2,034 $3,651
Insurance $533 $533 $533 $533 $1,067 $1,067
Land Usage Fees (royalties) $667 $667 $667 $667 $1,333 $1,333
Management/G&A $0 $421 $0 $667 $0 $869
Other Fixed Expenses $76 $0 $100 $0 $138 $0
Total O&M Costs, $k/yr) $3,103 $3,391 $4,190 $4,667 $6,174 $6,921
$/kW-yr) $31.03 $33.91 $41.90 $46.67 $30.87 $34.60
3-54
10581090
Wind Power

Operation and Maintenance Cost Estimates (4th Quarter 2009 $, Continued)
Site #4 Site #5 Site #6

Site Location Michigan New York New York Offshore
O&M Provider Plant Contract Plant Contract

General Plant Information Greenfield Greenfield Greenfield Greenfield Greenfield Site
Site Site Site Site
Rated Capacity 150 MW 150 MW 50 MW 50 MW 201.6
Number of turbines 75 75 25 25 56
Number of personnel or FTEs 20 0 9 0 -
Hours Per Year 2,453 2453 2,628 2,628 -
Reference Date of Cost Estimates 4th Q 2009 4th Q 2009 4th Q 2009 4th Q 2009 4th Q 2009
Operation and Maintenance Costs, $k/yr

Labor $1,590 $0 $683 $0 -
Maintenance $1,966 $3,528 $612 $1,098 -
Insurance $800 $800 $267 $267 -
Land Usage Fees (royalties) $1,000 $1,000 $333 $333 -
Management/G&A $0 $840 $0 $261 -
Other Fixed Expenses $126 $0 $58 $0 -
Total O&M Costs, $k/yr $5,482 $6,168 $1,953 $1,959 $25,828

$/kW-yr) $36.54 $41.12 $39.05 $39.18 $128.12
3-55
10581090
Wind Power
3.9.5 Levelized Cost of Electricity
The levelized cost of electricity (LCOE) is a measure of the life-cycle cost of electricity, and it
provides a consistent basis for comparing the economics of the projects evaluated across all of
the technologies and locations. The LCOE is calculated using a model that sums the present
values of the annual O&M and capital-related costs, and amortizes the result to yield the
levelized cost. The capital component of LCOE is a function of the Total Capital Requirement
and the economic assumptions.
Table 3-12 presents the LCOE estimates for each of the conceptual wind plant cases and both
with and without the 30% Investment Tax Credit. With investment tax credit, the LCOE ranges
between $58/MWh for the 100-MW base-case onshore wind plant to $110/MWh for the 201.6-
MW New York offshore plant. Without the tax credit, the LCOE ranges from $86/MWh to
$152/MWh for the same cases.
All costs are in fourth quarter 2009 U.S. dollars, and the current-dollar levelized costs assume
20-year plant life, 55/45 debt-equity ratio, 6.5%/yr interest on debt, 11%/yr return on equity,
3%/yr, inflation and escalation of operation and maintenance costs, five-year MACRS tax
depreciation, 40% income tax rate, 30% investment tax credit, and no Production Tax Credit.
Table 3-12
Levelized Cost of Electricity With and Without 30% Investment Tax Credit (4th Quarter 2009)
LCOE With 30% ITC LCOE Without 30% ITC

($/MWh) ($/MWh)
Wind Plant 1 (Base Case) $57 $86

Wind Plant 2 (CA) $76 $114
Wind Plant 3 (TX) $63 $95
Wind Plant 4 (MI) $79 $118

Wind Plant 5 (NY) $99 $152
Wind Plant 6 (NY Offshore) $107 $152
A probabilistic analysis of the levelized cost of energy is performed to generate a cumulative

probability distribution of the LCOE for each of the wind plant cases. The first step in the
process is to perform a sensitivity analysis to identify the parameters that make the most impact
on the LCOE. For the wind plant cases, the five highest sensitivity parameters are the capacity
factor, capital cost, return on equity, interest on debt, and O&M cost. Then, a probability
distribution is developed for each of the selected parameters based on the range of uncertainty of
the parameter. Finally, a 10,000-iteration Monte Carlo simulation is performed using the Crystal
Ball risk analysis software and the LCOE model to generate the cumulative probability as a
function of the LCOE.
3-56
10581090
Wind Power
Figure 3-23 is an example cumulative probability distribution for the 100-MW Base-Case wind
plant. The chart shows the minimum, maximum and 25th, 50th, and 75th percentile values of the
LCOE.
Figure 3-23
LCOE Probabilistic Analysis Results – Wind Plant 1 (Base Case, 4th Q 2009 $)
3.9.6 Performance and Cost Summary
Table 3-13 summarizes the performance, capital, and O&M cost estimates for the five
conceptual onshore wind plants and the offshore wind plant.
3-57
10581090
Wind Power
Table 3-13
Performance and Cost Estimate Summary for Horizontal-Axis Wind Turbine Power Plants (4th Quarter 2009 $)

Plant Size (number of units x unit size, MW) 50 x 2.0 MW 100 x 2.0 MW 50 x 2.0 MW
Nameplate Rating (MW) 100 MW 200 MW 100 MW
Technology Description Upwind, Horizontal Axis, Upwind, Horizontal Axis, Upwind, Horizontal Axis,
Doubly-Fed Induction Gen. Doubly-Fed Induction Gen. Doubly-Fed Induction Gen.
Physical Plant
Land under lease (flat, even terrain), km2 12 16 31
2
Disturbed area (3% of leased area), km 0.4 0.5 0.9
Unit Life, Years 20 20 20
Scheduling
Development time, Months 28-44 28-44 36-60
Construction time, Months 5-11 5-11 8-16
Total Capital Requirement ($/kW)
Reference Date of Costs 4th Quarter 2009 4th Quarter 2009 4th Quarter 2009
Wind Turbine Procurement and Shipping $1,570 $1,570 $1,560
Direct Construction Cost $328 $579 $318
EPC Indirect Construction Cost $110 $155 $102
Total WTG, Direct, and Indirect Costs $2,008 $2,304 $1,980
Recommended Contingency (10%) $44 $73 $42
Total Plant Cost $2,051 $2,378 $2,022
AFUDC (interest during construction) $0 $0 $0
Total Plant Investment $2,051 $2,378 $2,022
Development Fee For Purchasing the Project $50 $200 $25
From a Developer.
Recommended Owner Contingency (10%) $5 $20 $3
Total Capital Requirement $2,106 $2,598 $2,049
3-58
10581090
Wind Power


O&M Costs ($/kW-yr) By Owner By Contract By Owner By Contract By Owner By Contract
Labor $8.41 $0.00 $13.30 $0.00 $6.65 $0.00
Maintenance $9.86 $17.70 $15.60 $28.00 $7.80 $14.00
Insurance $5.33 $5.33 $5.33 $5.33 $2.67 $2.67
Land Usage Fees (royalties) $6.67 $6.67 $6.67 $6.67 $3.33 $3.33
Management/G&A $0.00 $4.21 $0.00 $6.67 $0.00 $3.33
Other Fixed Expenses $0.76 $0.00 $1.00 $0.00 $0.50 $0.00
Total O&M Cost $31.03 $33.91 $41.90 $46.67 $20.95 $23.33
Unit Availability
Equivalent Planned Outage Rate, % — — —
Equivalent Unplanned Outage Rate,% — — —
Equivalent Availability, % 94-97 94-97 94-97
Confidence and Accuracy Rating
Technology Development Rating Commercial Commercial Commercial
Design & Cost Estimate Rating Simplified Simplified Simplified
3-59
10581090
Wind Power

Site Location Michigan New York New York Offshore

Plant Size (number of units x unit size, MW) 75 x 2.0 MW 25 x 2.0 MW 56 x 3.6 MW
Nameplate Rating (MW) 150 MW 50 MW 201.6 MW
Technology Description Upwind, Horizontal Axis, Upwind, Horizontal Axis, Upwind, Horizontal Axis,
Doubly-Fed Induction Gen. Doubly-Fed Induction Gen. Doubly-Fed Induction Gen.
Physical Plant
Land under lease (flat, even terrain), km2 58 14 56.0
2
Disturbed area (3% of leased area), km 1.7 0.4 1.7
Unit Life, Years 20 20 20
Scheduling
Development time, Months 32-52 24-36 36-60
Construction time, Months 7-13 4-8 8-16
Reference Date of Costs 4th Quarter 2009 4th Quarter 2009 4th Quarter 2009
Wind Turbine Procurement and Shipping $1,590 $1,620 $1,997
Direct Construction Cost $446 $793 $1,260
EPC Indirect Construction Cost $122 $241 $521
Total WTG, Direct, and Indirect Costs $2,158 $2,654 $3,778
Recommended Contingency (10%) $57 $103 $178
Total Plant Cost $2,214 $2,757 $3,956
AFUDC (interest during construction) $0 $0 $0
Development Fee For Purchasing the Project $67 $400 $164
From a Developer.
Recommended Owner Contingency (10%) $7 $40 $16
3-60
10581090
Wind Power


O&M Costs ($/kW-yr) By Owner By Contract By Owner By Contract By Owner
Labor $10.60 $0.00 $13.66 $0.00
Maintenance $13.11 $23.52 $12.23 $21.95
Insurance $5.33 $5.33 $5.33 $5.33
Land Usage Fees (royalties) $6.67 $6.67 $6.67 $6.67
Management/G&A $0.00 $5.60 $0.00 $5.23
Other Fixed Expenses $0.84 $0.00 $1.16 $0.00
Total O&M Cost $36.54 $41.12 $39.05 $39.18 $128.12
Unit Availability
Equivalent Planned Outage Rate, % — — —
Equivalent Unplanned Outage Rate,% — — —
Equivalent Availability, % 94-97 94-97 94-97
3-61
10581090
Wind Power
3.10 Grid Integration
3.10.1 Production Variability
While wind is an intermittent generation resource, generation output from wind farms is more
stable than the term “intermittent” may suggest. Grid-connected wind farms typically have
capacity factors ranging from 25% to 40% or more over the course of a year. A common
misconception about wind power is that turbines are either on or off and sit dormant for much of
the year. In actuality, wind turbines are capable of partial output, and most wind farms generate
at some level during 70% to 90% of the year. Further, there are a number of factors that smooth
the output of a wind farm relative to variations in actual wind speed.
First, wind turbines do not begin to generate power until winds reach cut-in speed, usually 4 to
5 m/s, and will generate at the rated capacity at or above the rated wind speed—typically 12 to
14 m/s. This creates a “floor” and a “ceiling” over the range of wind speeds at which wind
turbine output will vary. In addition, the turbine rotor of variable-speed machines acts like a
flywheel, capturing the energy of sudden gusts and converting it to power more smoothly.
Second, the wind at modern turbine tower heights is steadier than the winds closer to the ground.
Third, geographic dispersion of wind turbines has a powerful mitigating effect on wind plant
output variation. Several studies have demonstrated this fact with wind plants spread among
multiple sites, but the smoothing effect of geographic dispersion also occurs within a single wind
farm. Individual wind gusts are typically very local phenomena, with a breadth of 30 m or less.
Since even a medium-sized project would be spread over several miles of land, individual
turbines do not experience the same wind speed simultaneously and ramp up and down at
differing rates.
Although variability over time due to wind speed changes can pose integration challenges, wind
power generally varies less rapidly than customer demand. Experience to date has shown that at
penetration levels of up to 10% to 20% of delivered total system load measured on an energy
basis and for large interconnected electricity grids with no significant transmission bottlenecks,
wind variability has little impact relative to the larger, shifting variability of utility system
demand. In addition to systems with transmission constraints, island and other systems that are
not interconnected may be affected at relatively low levels of wind penetration—especially those
with low nighttime loads and combined heat and power generation. That being said, the more
accurately these output variations can be anticipated and planned for, the more reliably and
economically wind can be integrated into a large utility system.
Alternately, energy storage and backup power generation resources at strategic locations in the
electricity grid can help alleviate some of the integration issues related to wind power. In fact,
some turbine manufacturers, such as GE, are reportedly working to embed inertial storage
capability directly into their turbine models to alleviate grid ramping issues. Cost of production
issues are stymieing commercial roll out. Application of energy storage and other integration
technologies to wind and other renewable technologies is more comprehensively addressed in
Chapter 10.
3-62
10581090
Wind Power
3.10.2 Ancillary Service Costs
In the last few years, several empirical studies have been performed in the United States
analyzing the impacts and ancillary services costs of wind power plant integration into specific
utility systems. Work is ongoing to establish methodologies for identifying integration costs
more accurately. “Good utility practice” for integrating wind will evolve for some time.
However, some initial conclusions can be drawn from work to date.
Recent studies—carried out in cooperation among NREL, other researchers, and utility operators
and planners—analyze the impacts and costs to integrate wind power into several different utility
systems, including Xcel Energy, PacifiCorp, and BPA. The results of the Xcel Energy evaluation
are presented in the Utility Wind Interest Group report, Characterizing the Impacts of Significant
Wind Generation Facilities on Bulk Power System Operations Planning, Excel Energy-North
Case Study (May 2003), also available from EPRI (EPRI 1004807, December 2003). The studies
estimate ancillary service costs for a range of utility operations time frames including unit
commitment and reserves, load following, and regulation. The methodologies vary to some
extent between the studies’ given differences in the wind regimes modeled, the degree of wind
penetration, the mix of utility generation assets, and the cost structures. However, a common
finding among all the studies is that, overall, impacts and costs are relatively low at low wind
penetration levels. Total ancillary cost estimates ranged from negligible costs in the case of
BPA’s system, to $1/MWh for Xcel’s Minnesota system, to a high of $5–$6/MWh for
Pacificorp’s system to integrate 2,000 MW of wind generation—equal to nearly 20% of system
capacity. A paper released in May 2003 summarizes these studies and provides references to the
specific analyses [4].
Building on these studies, the California Energy Commission (CEC) sponsored a similar detailed
study of the cost to integrate renewables into the state’s larger power system in support of the
California Public Utility Commission’s (CPUC) RPS implementation efforts. The study is
motivated by the RPS’s “least-cost, best-fit” bid selection criterion which requires that indirect
costs be considered in addition to the energy bid price when selecting eligible renewable
projects. Accordingly, the three-phase study quantifies the costs of integrating various renewable
technologies, including wind power.
Phase I analyzes the costs to integrate existing renewables [5]. Phase II of the study identifies
and analyzes the key attributes of renewable generators that affect integration cost. Phase III
[http://www.energy.ca.gov/reports/500-04-054.PDF] provides a final methodology to model the
characteristics and costs of renewables in California. Given the transparent and open approach of
the study, broad participation, extensive data availability from the Cal-ISO, and depth of the
analysis, the California study will likely serve as a benchmark for future efforts that attempt to
accurately assess the integration costs for renewables, including wind power, into larger utility
systems.
During 2004, two additional wind integration studies were completed in the United States. One
sponsored by the Minnesota Department of Commerce addressed the impacts of expected higher
wind penetration in the future. The other, sponsored by the New York State Energy Research and
Development Authority (NYSERDA) addressed the impacts of wind development on the New
York State electricity system. Recent wind integration studies are summarized in the 2006 EPRI
report, Survey of Wind Power Integration Studies (EPRI 1011883, 2006).
3-63
10581090
Wind Power
3.10.3 Wind Energy Forecasting and Scheduling
In recent years, the art of forecasting wind plant generation over the next zero to 48 or
72 hours has advanced rapidly. Current forecasting systems can provide system operators with
information necessary to integrate intermittent wind generation in the most reliable and
economic manner. Wind production forecasts can allow a system operator to schedule a wind
plant’s capacity and energy with some accuracy, effectively avoiding some capacity and fuel
costs while maintaining reliability standards. These systems also help wind plant operators
schedule turbine maintenance. Several companies and organizations in the United States, Europe,
and elsewhere in the world offer wind energy forecasting services and are providing next-hour,
next-day, and extended forecasts to wind plant operators, utility system operators, and electric
utilities.
Wind forecast systems integrate regional weather and project production models, utilizing
real-time inputs from on-site and off-site weather stations and wind plant SCADA systems. They
produce rolling output forecasts for periods of up to a month or more in advance, which can be
updated as frequently as every hour. The accuracy of these forecasts increases dramatically with
the proximity of the hour forecast. The week-ahead average forecast error is typically in the
range of 20% of nameplate. The average error for a forecast provided 60 minutes ahead for the
next hour is in the 8% range. Expected error for a forecast provided 30 to 40 minutes ahead of
the next hour should be closer to 4%. These error ranges will vary depending on the specific
wind regime and project layout.
During 2003 and 2004, EPRI published several reports on wind energy forecasting system
development and testing at two wind plants in California and one in Texas [27, 28, 29, 30, 31].
These reports assessed the capability of current forecasting technology to predict hourly wind
speed and energy generation over the next 48 hours. The mean absolute errors of the wind speed
and wind energy generation were 30% to 40% and 50% to 60% of the annual average wind
speed and energy generation, respectively.
During 2005, the California Energy Commission and EPRI completed development and testing
of a regional wind energy forecasting system, designed to generate five-minute forecasts over a
three-hour period, updated every five minutes. In addition, the project evaluated improved
technology for intermediate-term forecasting of hourly wind generation over a 48-hour period.
The results are described in a four-volume final report published during 2006 [32, 33, 34, 35].
To date, most wind projects simply deliver energy into a utility’s system on an unscheduled basis
without the use of forecasting. However, wind production forecasting is increasingly being used
for power scheduling purposes. For example, wind forecasting is now being used by the
California ISO to forecast approximately 500 MW of wind generation across the state.
Forecasting is also being used extensively in the Pacific Northwest to schedule the output of
individual projects, including the 300-MW Stateline project.
For utilities that are required to schedule generation with a control area operator, electronically
delivered next-hour wind production forecasts can be automatically integrated into next-hour
schedule modifications to minimize imbalances. As noted above, forecast error in these next-
hour time frames is expected to be small and may be smaller than a utility’s average control area
load forecast error. Since these two errors are not correlated, the relatively small wind forecast
error should make little net contribution to a utility’s existing level of scheduling imbalances.
3-64
10581090
Wind Power
3.10.4 Regulatory Update
Although the rules governing scheduling and imbalance penalties for wind are not consistent
throughout the United States, FERC is promoting a new approach that should ease the addition
of more wind capacity throughout the country. See Chapter 11 for discussion of the current status
of these and other related topics.
3.10.5 Power Quality
In recent years, wind turbine manufacturers have placed increasing emphasis on improving the
grid integration and power quality of wind turbine generators. This push is being driven largely
by two factors. First, as wind economics have improved and wind power has reached higher
penetration levels, expectations have increased for state-of-the-art performance and better
power quality. Second, many growth opportunities for wind around the world are on weaker
transmission systems where power quality is key for expansion. Slowing growth in the European
onshore wind market is focusing European manufacturers into the large North American market.
Unlike the European power grid, which is characterized by greater density of population, power
generation, and transmission lines, the U.S. market, particularly in the West, is characterized by
expansive geography and long distances between population centers and generation. Much of the
wind resource is located in remote regions with access to only smaller, less robust transmission
lines.
General Electric, the only major U.S. turbine manufacturer, has led the push for power quality
in the U.S. market with its patented dynamic VAR system. This system regulates voltage in real
time, supplying reactive power to the grid through each turbine’s power electronics instead
of through slower-acting capacitors or static VAR compensators. More recently, European
manufacturers have developed similar systems to compete effectively in North America.
In 2004, Danish manufacturer Vestas developed the V90, a 3.0-MW full variable-speed pitch-
regulated turbine that provides similar VAR support and ride-through benefits as the GE turbine.
In 2005, the company received an order for 36 units for an offshore project near the Dutch coast.
In addition to OEM-provided power quality systems, several after-market dynamic VAR support
systems are available, specifically designed for use with wind generation.
3.11 Project Development Process and Market
3.11.1 Buy or Build Considerations
As wind energy costs continue to drop and utility experience with purchasing wind energy
increases, more utilities are likely to consider developing, owning, and operating their own wind
projects. Two important initial considerations are project scale and capital intensity. Wind
project costs decline rapidly with project scale. In Class 5 and 6 wind regimes, projects of 100
MW and larger can typically deliver energy at costs below competing new thermal generation.
However, at smaller scales, this advantage can quickly disappear. Smaller project costs are
driven up by pre-development, financial transaction and construction mobilization expenses that
typically remain relatively flat regardless of project size. Also, wind projects are capital
3-65
10581090
Wind Power
intensive, typically coming in near $1,150/kW before 2005 (now $1,700 to $2,200/kW) installed
for a 100-MW project. For utilities that must carefully allocate capital across many competing
uses, entering a power purchase agreement with a wind developer may provide more strategic
benefit than committing to wind plant ownership.
For utilities that are interested in investing in a wind project, the question of balancing
cost management and risk must be addressed at each project development stage. The pre-
development stage is high risk. Pre-development companies typically invest up to a half million
dollars for site pre-screening, resource assessment, land and easement acquisition, environmental
review, and permitting on a specific site before viability can be determined. Generally, about one
in four sites clears all required hurdles and demonstrates resource quality sufficient to produce
energy in the target cost range. The high levels of in-house expertise maintained by experienced
pre-development companies and consultants helps to reduce pre-development risk and improve
the probability of identifying and successfully developing sites with the optimal resource in any
given area. Utilities choosing to take on the pre-development task can avoid the significant fees
charged by pre-development specialists for “packaged” sites, but may expose themselves to a
high probability of project failure or disappointing results. Utilities may also hire experienced
consultants to guide the pre-development process and manage sub-tasks in disciplines unfamiliar
to utility personnel, thereby reducing risk.
3.11.2 Project Development Process
The vast majority of wind capacity on-line in the United States has been developed, financed,
built, and operated by wind independent power producer companies, with the energy being
sold under long-term power purchase agreements with utilities. This arrangement stems from
the complexity of wind project development and project operation. Specialized activities require
expertise across a range of disciplines, a number of which do not fall within the areas of
expertise of typical electric utilities. In addition, equipment procurement and operations are
subject to substantial economies of scale. A small but growing portion of U.S. wind capacity is
owned and operated by utilities. Much of the utility-owned facilities have traditionally comprised
small wind installations with less than five turbines and been operated by municipal and public
utilities. However, large-scale, multi-megawatt utility wind ownership is on the upswing.
MidAmerican Energy developed a 310-MW project in its Iowa territory, intends to embark on
future projects totaling 545 MW, and is proposing to add another 1,000 MW of owned wind
generation, thereafter. Other utilities—including Alliant Energy, Kansas City Power & Light,
Minnesota Power, Nebraska Public Power District Oklahoma Gas & Electric, Puget Sound
Energy, Sacramento Municipal Utility District, and WE Energies, among others—have either
already invested in or have announced plans to own and develop wind projects.
3.11.2.1 Pre-Development Activities
The project development process falls into three distinct stages: 1) pre-development;
2) engineering, procurement, and construction (EPC); and 3) operations. Primary pre-
development tasks include resource assessment, transmission access evaluation, land acquisition,
permitting, pre-engineering/costing, marketing, and sometimes, finance.
3-66
10581090
Wind Power
Wind resource quality can vary significantly from site to site. Obviously some locations are
windier than others, but even within a known wind resource area, the wind resource can vary
substantially between adjoining land parcels, depending on topography. This is further
complicated by the fact that, for a given site, wind resources generally exhibit seasonal, diurnal,
and hourly variations. Accordingly, optimizing site selection for resource quality is perhaps the
most important and difficult pre-development task. Wind resource quality is characterized by
wind speed and direction—typically expressed as frequency distributions, the wind shear or
variation of wind speed with elevation, and the turbulence intensity.
Prior to final site selection, the wind resource is measured for an extended period of time—at
least one year, and usually two to three years—to statistically quantify the resource. If the data
collected show a high correlation to a nearby meteorological tower with a longer operating
history—at an airport, for example—a shorter period of data collection may be acceptable.
A meteorological tower or mast is erected at one or more locations to continuously measure wind
speed, direction, temperature, and sometimes other weather parameters. The measurements are
made at multiple elevations above the ground (typically 10, 30, and 50 meters) to allow the wind
shear to be estimated. As a rule of thumb, meteorological tower height should be at least
two-thirds the planned turbine hub height.
The recent trend toward turbine towers 80 m and taller will likely engender broader use of
new 60-m meteorological towers and SODAR, sonic detection, and ranging technology that uses
sound to passively profile the vertical shear from the ground up to several hundred meters. The
resulting data are stored on site by a data logger and periodically downloaded on site or remotely
by modem. Data are analyzed to resolve erroneous values and calculate average wind speeds,
directions, and temperatures over annual, seasonal, monthly, and hourly time intervals. The
information is often expressed in wind speed frequency distributions and wind roses, which
graphically show the relative frequency of wind speed and direction, and wind energy.
Most regions of the United States have specific locations where significant wind resources exist,
especially on mountain passes, ridges, and coastal areas near oceans and large lakes. Figure 3-24
shows the distribution of wind power in the United States.
3-67
10581090
Wind Power
Figure 3-24
Distribution of Average Wind Power in the United States. Source: Renewable Resource Data Center
3-68
10581090
Wind Power
Wind speed class characterizes the wind resource, with the darkest areas representing high-speed
Class 7 winds and the lightest representing low-speed Class 1 winds. In the last few years, more
detailed wind maps have been generated for most of the country that show the wind resource
distribution in finer detail than was available in previous maps. A listing and web links to these
updated wind maps is available from the DOE at
www.eere.energy.gov/windandhydro/windpoweringamerica/.
For cost-effective wind power generation, a wind power plant must be located at a site with high
wind speeds and must use reliable, efficient wind turbines. Generally, Class 3 is representative of
a low-to-moderate wind resource and Classes 4 through 6 are considered to be the most desirable
for commercial wind projects.
Site selection is typically an iterative screening process in which the pros and cons for a large
number of potential sites are evaluated with increasing levels of scrutiny until the highest
viability site or sites are selected for further pursuit. Pre-engineering, costing, and output
projection also include many variables and components that are unique to wind generation
and require specialized knowledge.
In today’s market, the wind project pre-development stage is often conducted by small,
regionally focused, independent wind pre-development companies that actively seek out sites
and, in turn, sell a completed pre-development package to large energy companies that self-
finance, build, own, and operate multiple wind projects. A completed pre-development package
typically includes: fully-documented wind resource studies, land leases, land use entitlements
(permits), an interconnection agreement, and a power purchase agreement. The build, own,
operate (BOO) companies typically also maintain in-house pre-development groups that actively
seek and pre-develop sites and market output to utilities or other buyers.
3.11.2.2 Engineering Procurement and Construction Activities (EPC)
At the EPC stage, there is also a range of cost and risk considerations that should be evaluated.
Wind generation technology and associated power management equipment is fully commercial
but still evolving rapidly. Advances are made annually in improving project engineering, layout,
equipment, and material choices. Accordingly, to achieve optimal project performance, engineers
with substantial and current wind generation experience should be utilized.
During the 1980s and 1990s, most wind projects were developed by independent pre-
development companies that also managed EPC tasks and project finance, tapping financial
markets for debt and equity. However, over the past several years, large utility or oil and gas
company affiliates have consolidated dominant wind development market share using a balance
sheet finance approach. The independent project finance model may re-emerge in the later half
of this decade as the broader financial markets recover their confidence in electricity generation
investments.
At the procurement level, economies of scale in purchasing can be important. Large wind BOO
companies often maintain multi-year bulk purchasing agreements with major equipment vendors,
which can allow them to gain the pricing benefit of very large purchase orders and spread them
3-69
10581090
Wind Power
across numerous projects, both large and small. Utilities and developers purchasing equipment
and services for single projects do not have this advantage. Nevertheless, the wind turbine
market in the United States is extremely competitive and has the benefit of a diversity of high-
quality manufacturers all competing to increase market share. Turbine equipment, which
typically accounts for approximately 75% of total project cost, can often be purchased in smaller
volumes at competitive prices, particularly if the purchaser can allow for flexible lead times.
Purchase order timing is important because the supply/demand balance in the U.S. turbine
market can be subject to discontinuities related to tax incentive deadlines. Also, long lead-time
orders allow turbine manufacturers to order major turbine components for just-in-time delivery,
saving inventory costs.
Regarding construction, wind projects are highly modularized, and the major tasks of pouring
foundations and erecting turbines can be accomplished very efficiently in a production line
manner as long as the work is properly staged, equipment deliveries are timely, and weather is
cooperative. Typically, crews can erect one turbine per day. Large projects may employ multiple
crews. This allows even very large wind projects in the 200-MW range to be completed within
six months from first groundbreaking to commercial start-up. However, because high crane work
and regular crane movement are major components of the job, excessive wind, rain, snow, or
other force majeure events can cause substantial delays and cost overruns. Accordingly,
allocation of construction risk should be a key consideration when evaluating a “buy or build”
decision.
3.11.2.3 Operation and Maintenance Activities
Wind project operations and maintenance are following the trends in other generating
technologies and are becoming increasingly systematized. More sensors are being used to gather
statistical performance data on turbines and subcomponents, and around-the-clock remote
monitoring is becoming the norm. Wind turbine manufactures typically offer full-wrap warranty
service contracts for up to five years and on-going support and service relationships are available
from both turbine manufacturers and wind O&M companies. Generally, current wind turbine
technology is highly reliable, with the historical fleet availability for leading manufacturers in
excess of 97%. Well-established turbine models typically show availability in excess of 98%.
However, because wind technology is still evolving rapidly and new turbine models are being
introduced annually, technology risk can be a factor.
With advanced instrumentation and SCADA systems, modern wind power plants are capable of
operating unattended, and on-site labor is required only for routine scheduled maintenance and
periodic troubleshooting. Figure 3-25 includes two graphs: one of O&M personnel vs. number of
turbines, and one of O&M personnel vs. size of facility. Both graphs include data from the TVP
projects as well as a number of other commercial projects. Some of the commercial projects are
based on proposals rather than actual experience; however, because the data for these planned
projects did not appear to distort the results, they are included in the graphs.
The graphs show a strong relationship between the number of O&M personnel at a wind power
2
plant and the number of turbines (r = 0.92). Although there also appears to be a relationship
between the number of people and the size of the project (r2=0.70), the size of the maintenance
crew is more dependent on the number of turbines than on the rated capacity of the project. This
3-70
10581090
Wind Power
is due to the fact that, although some of the maintenance activities on larger wind turbines may
require more time or different equipment to complete repairs, many maintenance activities
require the same level of effort regardless of turbine size.
O&M Personnel versus Number of Turbines

14
12 y = 0.0533x + 0.6185
R2 = 0.9196
10
O&M Personnel
0
0 25 50 75 100 125 150 175
Number of Turbines
O&M Personnel versus Project Rating

14
y = 0.0477x + 1.2231
12
R 2 = 0.7002
O&M Personnel
10
8
6
4
2
0
0 50 100 150 200
Project Rating (MW)
Figure 3-25
The Number of O&M Personnel at a Wind Power Plant Versus the Number of Turbines and
Project Rating
3.12 Environmental Issues

Environmental and social issues associated with wind energy include public acceptance,
permitting, land use, soil erosion, visual and noise impacts, and impacts on resident and
migratory bird and animal populations. These issues are addressed in the 1999 National Wind
Coordinating Committee report, Permitting of Wind Generation Facilities [17].
3-71
10581090
Wind Power
Experience in the DOE-EPRI Wind Turbine Verification Program indicates that, through
vigorous public outreach and involvement programs and a proactive approach to assessing
environmental impacts, it is possible to address community concerns and, ultimately, receive
support and foster a sense of ownership by the community.
3.12.1 Avian and Bat Issues
Avian interaction with wind facilities became a central issue for the wind industry in the late
1980s, when bird carcasses were first reported in the Altamont Pass wind resource area in
northern California. More recently, concerns over endangered bat species have been raised by
observations of increased bat mortality at the Mountaineer wind project in West Virginia
and other wind projects.
Since numerous birds are killed each year by man-made structures such as buildings,
communication towers, and bridges, not to mention cars, trucks, and airplanes, the idea that local
bird populations might be affected was not new to the wind industry. However, the presence of
raptor carcasses drew attention to the issue in the Altamont Pass, and the fact that the state’s
golden eagle population could potentially be affected raised the stakes.
In 1989 the California Energy Commission, along with the counties of Solano, Alameda, and
Contra Costa, contracted with BioSystems to conduct the first extensive study of bird fatalities
in the Altamont Pass. The results from that initial research were published in 1992. While it left
many questions unanswered, the BioSystems study did indicate that wind turbines were indeed
killing birds, particularly raptors, and in numbers significant enough to cause concern. It also
marked the beginning of the debate over what has become the most contentious environmental
issue to face the wind industry.
Since 1992, much more research has been done at the Altamont Pass, as well as numerous other
wind resource areas, to identify the species found in each area and define their use and mortality
rates; understand why raptors seem particularly susceptible to collisions with wind turbines;
and identify the factors that lead to bird deaths (i.e., turbine designs, placement, geography,
vegetation, and prey availability at the site; habitat encroachment in surrounding areas; and
interaction behaviors such as flying, perching, hunting, etc.). To date, this body of research
suggests that biological resource and risk factors vary significantly from site to site, and that
results from the Altamont Pass are not necessarily representative of other sites. In fact, all
research to date on sites in the United States indicates that significant bird mortality rates are
only correlated with one commercial wind energy facility—Altamont Pass. However, this does
not mean that the issue can be dismissed.
In 1999, the National Wind Coordinating Committee published a definitive report, Studying
Wind Energy/Bird Interactions: A Guidance Document [16]. The report provides extensive
information on designing and conducting field avian studies to better understand the risk that
wind facilities create to local avian populations. If widely adopted and implemented, this report
will lead to a body of research that will produce “credible and comparable results.” The report
stresses that, while a general protocol for conducting research should be adopted, each site must
be evaluated on the basis of its unique set of parameters.
The current wind industry trend toward larger wind turbines is also considered to be beneficial
to reducing the avian mortality risk at wind facilities. Compared to the small 50- to 200-kW wind
turbines installed during the 1980s at Altamont Pass and other wind areas, the large wind
3-72
10581090
Wind Power
turbines being installed today typically use taller towers and operate at much lower rotor speeds,
typically 20 to 30 rpm. This may elevate the turbine rotor above the elevation of much of the
avian habitat, while slower rotor speed should reduce the probability of striking birds. In addition
to new wind projects, repowering projects in California are replacing large numbers of small
turbines with fewer large machines, which should reduce the avian risk at existing wind projects.
The industry must continue to take this issue seriously and strive to minimize avian and bat
risk at all wind facilities through ongoing research and proper siting. The National Wind
Coordinating Committee continues to take a lead role in coordinating and disseminating research
on biological impacts of wind farms and is an excellent resource for information on this subject
[9].
3.12.2 Noise
Although wind generation facilities generate noise during both construction and operation, it is
the long-term operating noise that receives the most attention during permitting. In August 2002,
the National Wind Coordinating Committee issued an update to their report, Permitting of Wind
Generation Facilities, which addresses noise characteristics, impacts on receptors, prediction and
measurement, and mitigation strategies [17].
The noise produced by operating wind generation facilities is much different in both level and
character from the noise generated by large power plants and other industrial facilities. It is
generally considered to be low-level noise and consists of both mechanical and aerodynamic
components.
Mechanical noise is typically only tonal noise at discrete frequencies, e.g., noise caused by wind
turbine mechanical components, vortex shedding from blades, and unstable flows over holes or
slits. Mechanical components that generate noise include gearboxes, yaw drives, brakes, and
cooling fans for generators. Mechanical noise can increase as components move out of alignment
and gears and bearings wear over time.
Aerodynamic noise is caused by airflow over and past the turbine blades and tends to increase
with rotor speed and wind speed. It consists of three components: broadband, impulsive, and
low-frequency noise. Broadband noise is characterized by a continuous distribution of sound
pressure with frequencies greater than 100 Hertz (Hz) and is often caused by the interaction of
turbine blades and atmospheric turbulence, producing the familiar “swishing” or “whooshing”
sound. Impulsive noise typically appears as a thumping noise as the wind turbine blades of
downwind turbines pass through the wind shadow of the tower. Low-frequency noise, in the
range 20 to 100 Hz, is also associated with older downwind turbines and results from interactions
between the blades and wake turbulence caused by towers and nearby wind turbines.
In general, the more the noise from a new source exceeds the background level or generates a
different tonal characteristic than background noise, the more it will be unacceptable to the local
community. The perception of wind generator noise by a receptor depends on the noise intensity,
frequency, frequency distribution, and pattern; background noise level; proximity of the receptor
to the wind turbine; terrain and vegetation between the wind generator and the receptor; and the
nature of the receptor. Background noise tends to increase with wind speed and thus mask the
3-73
10581090
Wind Power
wind generator noise. Therefore wind generator noise is generally most noticeable at lower wind
speeds. If the receptor is at a location where the wind speed is lower than at the wind turbine,
e.g., in a hollow or depression in the terrain, then the background noise will also be lower there.
In general, a wind turbine is likely to be heard at twice the distance in hilly terrain as in flat
terrain.
Although no federal and few state noise standards exist, the U.S. Environmental Protection
Agency has promulgated noise guidelines. Many local governments have enacted local noise
ordinances that must be considered when siting wind facilities.
The NWCC Permitting Handbook cites two proposed noise measurement techniques for wind
energy systems, one prepared by the Solar Energy Research Institute (now NREL) in 1987,
A Proposed Metric for Assessing the Potential of Community Annoyance; the other prepared
by the American Wind Energy Association (AWEA) in 1989, Procedure for Measurement of
Acoustic Emissions from Wind Turbine Generator Systems, Tier 1 – 2.1. In addition, the
International Electrotechnical Commission (IEC) proposed a standard in 1988 that was
rejected in the balloting process.
Turbine noise studies should include separate measurements of low-frequency and A-weighted
noise levels across a range of wind speeds, turbulence conditions, distances from the turbine,
and locations of the receptor relative to wind direction. Preferably before the wind project is
installed, background noise measurements should be conducted at representative dwellings
up to one-quarter mile from the site for flat terrain, and up to one-half mile for uneven terrain.
Receptors at less windy locations should be emphasized.
Several software packages are available that predict noise level profiles for the area around
a wind facility. However, prediction of wind generator noise levels at a given site is difficult
because conditions vary between sites. For example, the variations of wind turbulence and
noise with wind speed depend on wind direction, terrain, vegetation, and other site variables
and are rarely the same at two sites. Thus, the measured noise levels at one site are not
necessarily representative of the noise levels that would occur at another site. Noise mitigation
measures include requirements for noise setbacks of 400 to 1,000 meters from the edge of the
property line; installation of sound insulation and baffles in the turbine nacelle; use of low-speed
cooling fans, special finishing of gear teeth, vibration isolators and soft mounts for major
components; and design of wind turbine rotors for low noise. If local residences are shielded
from the wind, a larger setback may be required than in an exposed location.
3.12.3 Visual Impact
Wind projects are usually located in rural and remote areas, and their visual impacts are
somewhat different from those of conventional power plants. Visual impact considerations
during permitting typically focus on the impact of the wind project on the “viewshed,” or visual
appearance of the project setting from different locations. They include the impacts on the
natural terrain and landscape; the form, line, color, and texture of the viewshed; and the visibility
and appearance of landmarks in the area. In addition, the compliance of the project with public
preferences, local goals, policies, and guidelines regarding visual quality is crucial toward
gaining acceptance.
3-74
10581090
Wind Power
The elements of wind projects that affect visual impact include the relative elevations of the site
and the surrounding area; the presence of dense woodland or other vegetation cover at the site;
the number of wind turbines; wind turbine spacing and uniformity; height and color of the wind
turbine nacelle, tower, and rotor; the type of tower used (lattice or tubular steel); markings and
lighting required for FAA approval; roads built on slopes; and the locations and sizes of service
buildings, substations, electrical switchgear, transmission lines, and other on-site facilities. In
general, deploying fewer and wider-spaced turbines of uniform type, color, tower design, and
rotational direction enhances a project’s visual appearance and makes it more likely to gain
public acceptance.
In a March 2003 study, the Renewable Energy Policy Project (REPP) found that property values
within the viewshed of recently constructed commercial-scale wind farms have not been
negatively impacted [10]. The study assembled a database covering every wind development that
had come online in the United States after 1998 with 10-MW installed capacity or greater. REPP
then compared actual property sales records for all land within the viewshed and for a
comparable community not within the viewshed. The report found that for the great majority of
projects, property values actually rose more quickly in the viewshed than they did in the
comparable community. Moreover, values increased faster in the viewshed after the projects
came on-line than they did before. In all, 30 cases were analyzed; in 26 of them, property values
in the affected viewshed increased more than those in other areas without wind turbines.
3.13 Equipment Markets and Key Participants

The world market for wind turbine equipment and services is growing at a high rate, and the
rapid growth is expected to continue into the 21st century. According to the annual report,
International Wind Energy Development, World Market Update 2008, Forecast 2009-2013
(BTM Consult ApS, Denmark, March 2009), installed world wind capacity grew at an average
annual rate of 27.6% per year from 2004 through 2008, and total installed capacity is expected to
almost triple from 122,158 MW to about 343,150 MW between 2009 and 2013. At the same
time, the world turbine market is expected to grow. The cumulative wind turbine market is
valued at more than $180 billion over the coming five years.
The drivers for this rapid growth include growing electricity demand throughout the world, the
declining cost of wind energy, and government mandates for clean energy and CO2 emissions
reduction, especially in the European Community. In the United States, the Production Tax
Credit, ITC election and cash grant stipulations within ARRA, and the establishment of
renewable portfolio standards in a growing number of states will continue to stimulate the U.S.
market. The governments in Germany and Spain, among other EU nations, have enacted
renewable energy feed-in tariffs (REFITs) and other incentives which are expected to further
stimulate the European market—the leading market for wind energy in the world. Provincial-
and state-level governments within Canada (e.g., Prince Edward Island, Ontario) and various
U.S. states (e.g. California and Vermont, among others) have also more recently enacted REFITs
to stimulate growth.
3-75
10581090
Wind Power
3.13.1 World Markets
Table 3-14 summarizes the 10 largest wind turbine markets by country in 2007, 2008, and 2009.
During 2009, the top 10 installed 33,982 MW of new capacity represented about 89% of the total
world market for that year. The largest markets were China (13,750 MW), the United States
(9,922 MW), Spain (2,331 MW), Germany (1,917 MW), and India (1,172 MW).
Table 3-15 presents the growth rates of installed wind capacity for the top 10 wind turbine
markets in 2008. The highest 2008-2009 growth rates occurred in China (113.3%), the United
States (39.3%) and the UK (33.0%), while the highest three-year average growth rates occurred
in China (115.4%), the United States (44.6%), and France (44.4%). The 2007–2008 and three-
year growth rates for the top 10 markets were 31.9% and 29.5%, respectively.
Two ongoing trends that will continue to shape the market are the increasing size of new wind
turbine generators and the shift of new installations in Europe to offshore applications. The trend
toward larger wind turbines is driven both by the economies of scale of larger machines and the
need for larger machines in offshore applications.
Table 3-14
Ten Largest Wind Turbine Markets: 2007–2009
Country 2007 2008 2009

China 3,287 6,246 13,750
United States 5,244 8,358 9,922
Spain 3,100 1,739 2,331
Germany 1,667 1,665 1,917
India 1,617 1,810 1,172
Italy 603 1,010 1,114
France 888 1,200 1,104
United Kingdom 427 869 1,077
Canada 386 526 950
Portugal 434 679 645
Total 17,653 24,102 33,982
Percent of World 89.2% 85.5% 89.2%
Source: BTM Consult ApS – March 2010.
3-76
10581090
Wind Power
Table 3-15
Growth Rates of Installed Wind Capacity in the Top 10 Markets During 2008
Installed Capacity (MW) Growth Rate (%/yr)

Country End 2006 End 2007 End 2008 End 2009 2008-2008 3-Year Avg.
USA 11,635 16,879 25,237 35,159 39.3% 44.6%
China 2,588 5,875 12,121 25,583 113.3% 115.4%
Germany 20,652 22,277 23,933 25,813 7.9% 7.7%
Spain 11,614 14,714 16,453 18,784 14.2% 17.4%
India 6,228 7,845 9,655 10,287 12.1% 20.2%
Italy 2,118 2,721 3,731 4,845 29.9% 31.8%
France 1,585 2,471 3,671 4,775 30.1% 44.4%
UK 1,967 2,394 3,263 4,340 33.0% 30.2%
Portugal 1,716 2,150 2,829 3,474 22.8% 26.5%
Denmark 3,101 3,088 3,159 3,408 7.9% 3.2%
Total “Ten” 63,203 80,415 104,051 137,277 31.9% 29.5%
Sources: Wind Power Monthly, BTM Consult ApS 2009.
3.13.1.1 Trend to Larger Turbines
Table 3-16 shows the average nameplate ratings of wind turbine generators installed between
2004 and 2009 in eight countries. The average turbine rating was about 974 kW in 2004 and
increased to about 2,251 kW in 2009. Megawatt-class wind turbines continued to gain an
increasing share of the overall market in 2009. The average turbine rating surpassed 1 MW
in the UK (2,251 kW), Germany (1,977 kW), Sweden (1,974 kW), the United States (1,731 kW),
Spain (1,897 kW), and China (1,360 kW).
The noticeable increase in ever-larger turbines as a major proportion of the market prompted
BTM Consult Aps to redefine the ranges associated with the various market segments in 2004
and again in 2008. The ranges are: 0-749 kW for small turbines, 750-1,499 kW for megawatt-
class turbines, 1,500-2,500 kW for mainstream turbines, and >2,500 kW for multi-megawatt
class turbines. Using these definitions, the 2009 market shares were 1.1% for small wind
turbines, 7.6% for megawatt class turbines, 81.8% for mainstream turbines, and 5.1% for multi-
megawatt class turbines. It is expected that the small turbine market share will decrease, the
megawatt class turbine market will decrease slightly, and the mainstream turbine market share
will continue to increase.
3-77
10581090
Wind Power
Table 3-16
Annual Average Nameplate Ratings (kW) of Wind Turbines Installed by Country
Year China Denmark Germany India Spain Sweden UK USA
2004 771 2,225 1,715 767 1,123 1,336 1,695 1,466
2005 897 1,381 1,634 780 1,105 1,126 2,172 1,499
2006 931 1,875 1,848 926 1,469 1,138 1,953 1,667

2007 1,079 850 1,879 986 1,648 1,670 2,049 1,669
2008 1,220 2,277 1,916 999 1,837 1,738 2,256 1,677

2009 1,360 2,368 1,977 1,117 1,897 1,974 2,251 1,731
The installed capacity share of small wind turbines (<750 kW) increased to 1.1% (398 MW) of
total capacity between 2008 and 2009. The installed capacity shares of megawatt-class turbines
(750-1,499 kW) decreased to 7.6% while the installed capacity share of multi-megawatt-class
turbines (> 2,500 kW) reached 5.1% (1,882 MW) of total capacity during 2008 and 2009.
During 2009, the leading suppliers of the mainstream machines (1500 to 2500 kW) included GE
Wind (14.7%), Vestas Wind Systems (13.2%),Sinovel PRC (10.9%), Enercon (8.8%), and
Dongfang PRC (7.6%).
Table 3-17 presents data for selected commercial megawatt-class wind turbine generators offered
by the major suppliers. Of the manufacturers producing multi-megawatt turbines, Enercon and
Repower lead the way with development and testing of 6-MW machines, followed by Areva and
Multibrid with 5.0-MW machines, and Vestas and WinWind with 3.0-MW machines.
Table 3-17
Commercial Megawatt-Class Wind Turbine Generators
Manufacturer/Turbine Rating Rotor Control

Comments
Model (kW) Dia. (m) Features
Acciona AW 3000 3000 116 Pitch (V)
Alstom, DEC, 1.5 MW 3000 100 Pitch (V)
Areva Multibrid 5000 116 Pitch (V)
Bard 5000 122 Pitch (V)
Clipper 2500 89-100 Pitch (V) Rotors avail. 89/93/96/100m
Dongfang, DEC, 1.5 MW 1500 77 Pitch (V)
Enercon E 66, 1.8 MW 1800 70 Pitch (V) Multipole Generator
3-78
10581090
Wind Power


Comments
Enercon E112, 4.5 MW 4500 112/114 Pitch (V) Multipole Generator
Enercon E112, 6.0 6000 112 Pitch (V) Multipole Generator
Enercon E126, 6.0 6000 126 Pitch (V) Multipole Generator
Fuhrlander FL MD77 1500 77 Pitch (V)
Fuhrlander FL 2500 2500 100 Pitch (V)
GAMESA G80 2000 80 Pitch (V) IGBT Inverter
GE Wind 1.5s,sl,se 1500 70.5-77 Pitch (V) IGBT Inverter
GE Wind 2.3 2300 82 Pitch (V) IGBT Inverter
GE Wind 2.5 2500 104 Pitch (V) IGBT Inverter
Goldwind 1.5 MW 1500 70 Pitch (V) Direct Drive-PMG
Hunan XEMC XE72 2000 72
MHI, MWT 62 1000 62
MHI, MWT 92 2400 92
Mingyang MY1.5se/see 1500 77/82
Nordex N60/N62 1300 60/62 Stall (C) 2-Speed Switchable Poles
Nordex S70/77 1500 70/77 Pitch (V)
Nordex N90, 2.3 MW 2300 90 Pitch (V)
REpower MD70/MM77 1500 70/77 Pitch (V)
REpower MM82 2000 82 Pitch (V) IGBT Inverter
REpower MD70/MM77 1500 70/77 Pitch (V)
REpower MM82 2000 82 Pitch (V) IGBT Inverter
REpower MM92 2000 92
REpower M5, 5 MW 5000 126
3-79
10581090
Wind Power


Comments
REpower M6, 6 MW 6000 126
Sewind W1250 1250 64
Siemens SWT-1.3-62 1300 62 Active Stall (C) 2-Speed CombiStall
Siemens SWT-2.0-76 2000 76 Active Stall (C) 2-Speed CombiStall
Siemens SWT-2.3-82 2300 82.4 Active Stall (C) CombiStall
Siemens SWT-2.3-93 2300 93 Pitch (V) Variable-speed
Siemens SWT-3.6-107 3600 107 Pitch (V) Variable-speed
Sinovel 70-1500 1500 70
Sinovel 77-1500 1500 77
Sinovel 82-1500 1500 82
Sinovel SL-3000 3000 90-110 Pitch (V) Rotors avail: 90/100/105/110
Suzlon 1.25 MW 1250 64/66
Suzlon 1.5 MW 1500 82
Suzlon 2.1 MW 2100 88
United Power UP77/1500 1500 77 Pitch (V)
United Power UP82/1500 1500 82 Pitch (V)
VESTAS V82 1650 82 Active Stall (C) 1 or 2 Speed
VESTAS V80-1.8MW 1800 80 Pitch (V) N/A
VESTAS V80 2000 80 Pitch (V) Variable Frequency Control
1800/
VESTAS V90-1.8/2MW 90 Pitch (V) Variable Frequency Control
2000
VESTAS V90 3000 90 Pitch (V) Variable Frequency Control
WinWind 1 MW 1000 62 Pitch (V) Multibrid
WinWind 3 MW 3000 90/100 Pitch (V) Multibrid
Windey WD77 1500 77
(C) = Constant-Speed Turbine; (V) = Variable-Speed Turbine.
3.13.1.2 Offshore Market in Europe
The growing offshore market in Europe is driven by the declining availability of good sites
on land, especially in Denmark. Installation of large megawatt-class machines on offshore
foundations and pilings is economically attractive due to both the economies of scale of larger
machines and the availability of wind resources with more uniform wind speeds and less wind
turbulence. Assessments of potential resources and feasibility of European offshore wind show
that the market can develop very fast. The exploitable offshore wind resource in Europe is
3-80
10581090
Wind Power
approximately 300 TWh/year. Together with an identified onshore potential of approximately

600 TWh/year, the total wind power potential in Europe is 900 TWh/year, equaling almost
one-third of the total electricity consumption in Europe today.
3.13.1.3 Market Forecast
Figure 3-26 presents the BTM Consult forecasts of wind power development for 2009 to 2013
and 2014 to 2018. The forecast is based on an analysis of a number of technical, economic, and
political factors in each country. The factors include national energy plans, government support
for renewable energy, present market for wind turbines, new project announcements, wind
resource assessment, and expected technology development and cost reduction.
Cumulative Global Wind Power Development

1,000,000
Prediction 2015-2019 Forecast 2010-2014 Actual 1990-2009
900,000
800,000
700,000
600,000
MW
500,000
400,000
300,000
200,000
100,000
0
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Source: BTM Consult ApS - March 2010 Year
Figure 3-26
Wind Capacity and Forecast: 1990-2018 (Source: BTM Consult ApS, March 2010)
3.13.1.3.1 2009 Through 2013
Installed world wind capacity reached 122,158 MW during 2008 and is expected to grow to
343,153 MW in 2013. Over the same period, U.S. wind capacity reached 25,237 MW during
2008 and is expected to grow to 77,237 MW in 2013.
3-81
10581090
Wind Power
Table 3-18 presents the global forecast of installed wind capacity for 2009-2013. The installed
capacity is forecast to grow to 93,968 MW in the Americas, 145,186 MW in Europe, 83,224
MW in South and East Asia, 12,456 MW in the OECD-Pacific region, and 8,320 MW in the rest
of the world.
Table 3-18
Global Wind Power Development Forecast: 2009-2013
Cum Inst’d Cum

Region Forecast Installed MW: 2009-2013
MW MW MW
2009 2009 2010 2011 2012 2013 2014 Total 2014
Total
40,351 9,527 7,650 10,450 12,450 16,200 18,300 65,050 93,968
Americas
Total
65,971 9,179 11,580 13,505 15,900 18,080 20,150 79,215 145,186
Europe
Total
South and 22,174 8,201 9,650 10,300 12,400 13,400 15,300 61,050 83,224
East Asia
Total
OECD- 4,256 1,051 1,100 1,350 1,600 1,900 2,250 8,200 12,456
Pacific
Total
Other 840 232 645 1,035 1,470 1,810 2,520 7,480 8,320
Areas
Total New
28,190 30,625 36,640 43,820 51,390 58,520 220,995
MW
Cum MW 122,158 152,783 189,423 233,243 284,633 343,153 343,153

3.13.1.3.2 2013 Through 2017
As shown in Figure 3-26, the installed world wind capacity is predicted to grow from about
343,153 MW in 2013 to about 833,546 MW in 2018. The forecast is based on an assessment of
global trends in energy consumption, economic growth, restructuring of the utility markets, and
the eventual impacts of the Kyoto Protocol and future agreements concerning global climate
change.
It is anticipated that the wind power growth will no longer be dependent on environmental
regulations and will instead be driven by traditional market drivers, including energy demand
growth, declining cost of wind energy, and the expansion of offshore wind generation in Europe.
3-82
10581090
Wind Power
3.13.2 Key Players in the Wind Power Industry
Table 3-19 lists the major global participants in the wind power industry. They include wind
turbine manufacturers, component suppliers, and project developers.
Table 3-19
Global Participants in the Wind Power Industry
Wind Turbine Manufacturers Component Suppliers Wind Project Developers

Vestas Wind System A/S Blades WKN Windkraft Nord AG
Enercon GmbH LM Glasfiber A/S FPL Energy LLC
NEG Micon A/S Polymarin National Wind Power Ltd.
GE Wind Systems Aerpac BV Renewable Energy System
Gamesa Eolica Abeking & Rasmussen Elsam A/S
Bonus Energy NOI Rotortechnik EnXco
Nordex-Borsig Energy TPI Composite SeaWest WindPower, Inc.
MADE Energias Renovables Huiteng Winkra GmbH
Mitsubishi Heavy Industries MHI Zhongfu Lianzhong Umweltkontor GmbH
REpower Systems AG REG Entwicklungs KG
Ecotecnia Gearboxes Das Grune Emissionshaus
Suzlon Energy Jahnel Kestermann Getriebewerke Italian Vento Power Corp.
DEWIND GmbH Winergy AG EHN
Desarrollos Plambeck Neue Energien AG
Fuhrlander GmbH Hansen Transmissions Gesellschaft fur Energie un
Jeumont Industrie Bosch Rexroth Oekologie
Lagerway the Windmaster Moventas (formerly Metso Drives EnergieKontor
Nordic Windpower AB Tehnology) Jysk Vindkraft A/S
Turbowinds NV China High Speed SEAS Wind Energy Centre
Wincon West Wind TechWise
Siemens Wind Power Generators Viking Wind farms A/S
Sinovel Wind Company ABB AIRICOLE AB
Acciona Windpower Elin Espace Eolien Developpement
Goldwind Siemens (EED)
Dongfang Steam Turbine Co Weier Elektromotrenwerk GmbH Greentech Energy Systems A/S
Alstom Ecotecnia Winergy AG DS SM A/S
Clipper Windpower Ingeteam DMI
WinWinD Leroy-Somer Hendricks Industries
Yongji Welcon
CS Wind
Towers Dajin
KGW Schweriner Maschinenbau
GmbH
Chemieanlagenbau Stassfurt AG
Omnical Borsig Energy GmbH
Pfleiderer AG
Bladt Industry A/S
Valmont Wind Energy
Erik Roug A/S
3-83
10581090
Wind Power
Table 3-20 presents the global market shares of the leading manufacturers of wind turbines in
2006, 2007, and 2008. During 2008, the five market share leaders were Vestas (19.8%), GE
Wind (18.6%), Gamesa (12.0%), Enercon (10.0%), and Suzlon (9.0%). In most years, the total
market shares do not add to 100% due to an imbalance between turbine sales and installations.
Table 3-20
Global Market Shares of Wind Turbine Manufacturers in 2006, 2007 and 2008
MW Sales % Market MW Sales % Market MW Sales % Market

Manufacturer
2006 Share 2007 Share 2008 Share
Vestas (DK) 4,239 28.2% 4,503 22.8% 5,581 19.8%
GE WIND (U.S.) 2,326 15.5% 3,283 16.6% 5,239 18.6%
Gamesa (ES) 2,236 15.6% 3,047 15.4% 3,373 12.0%
Enercon (GE) 2,316 15.4% 2,769 14.0% 2,806 10.0%
Suzlon (Ind) 1,157 7.7% 2,082 10.5% 2,526 9.0%
Siemens (DK) 1,103 7.3% 1,397 7.1% 1,947 6.9%
Sinovel (PRC) 75 0.5% 671 3.4% 1,403 5.0%
Acciona (ES) 426 2.8% 873 4.4% 1,290 4.6%
Goldwind (PRC) 416 2.8% 830 4.2% 1,132 4.0%
Nordex (GE) 505 3.4% 676 3.4% 1,075 3.8%
Others 689 4.6% 2,076 10.5% 4,955 17.6%
Total 16,003 22,207 31,326
3.14 References
1. American Wind Energy Association (AWEA) news release, August 20, 2003, available at
http://www.awea.org/news/news030820nmw.html.
2. Wind Turbine Verification Project Experience: 1999. EPRI, Palo Alto, CA: November 2000.
1000961.
3. Long Island’s Offshore Wind Energy Development Potential: A Preliminary Assessment,
Long Island Power Authority and the New York State Energy Research and Development
Authority, April 2002.
4. B. Parsons & M. Milligan, B. Zavadil & D. Brooks, B. Kirby, K. Dragoon, and J. Caldwell,
“Grid Impacts of Wind Power: A Summary of Recent Studies in the United States,” June
2003 presented at the EWEA Conference in Madrid, Spain.
5. California Renewables Portfolio Standard Renewable Generation Integration Cost Analysis,
Phase I: One-Year Analysis of Existing Resources, Results and Recommendations, Final
Report, Oct. 9, 2003. See
http://cwec.ucdavis.edu/rpsintegration/RPS_Int_Cost_PhaseI_pubrev1.pdf.
6. M. Milligan. Modeling Utility-Scale Wind Power Plants, Part 2: Capacity Credit.
NREL/TP-500-29701. Golden, Colorado: National Renewable Energy Laboratory, 2002.
3-84
10581090
Wind Power
7. MAPP Reliability Handbook, Section 3, Sub-section 4.7.2.7, “Procedures for setting Monthly
Net Capability for Variable Capacity Generation.”
8. NEPOOL Manual for Installed Capacity, Manual M-20, Supplement D.
9. Updated research and information on current activities is available at NWCC’s website at
http://www.nationalwind.org/.
10. The Effect of Wind Development on Local Property Values, REPP, Washington, D.C.,
May 2003. See www.repp.org/articles/static/1/binaries/wind_online_final.pdf.
11. Planning Your First Wind Power Project: A Primer for Utilities. EPRI, Palo Alto, CA:
December 1994. TR-104398.
12. Building Community Support for Local Renewable and Green-Pricing Projects. EPRI,
Palo Alto, CA: December 1999. TR-114203.
13. Renewable Energy Technology Characterizations. U.S. Department of Energy and
EPRI, Palo Alto, CA: December 1997. TR-109496.
14. Power-Electronic, Variable-Speed Wind Turbine Development: 1988-1993. EPRI,
Palo Alto, CA: September 1995. TR-104738.
15. Gipe, P., Wind Energy Comes of Age, John Wiley & Sons, Inc., New York, NY, 1995,
pp. 216-219.
16. Studying Wind Energy/Bird Interactions: A Guidance Document, National Wind
Coordinating Committee, Washington, D.C.: December 1999.
17. Permitting of Wind Facilities, National Wind Coordinating Committee, Washington,
D.C.: August 1999.
18. Windicator, Wind Power Monthly, Knebel, Denmark: October 2005.
19. International Wind Energy Development, World Market Update 2004, BTM Consult ApS,
Ringkobing, Denmark, March 2005.
20. “Stretching the Boundaries – Wind Energy Technology Review 2004-2005,” Renewable
Energy World Magazine, February, 2005.
21. www.re-focus.net, March 2002.
22. Offshore Wind Could Quadruple UK Wind Energy, Renewable Energy World Magazine,
May-June 2001.
23. Wind Turbine Productivity Improvement and Procurement Guidelines. EPRI, Palo Alto,
CA: March 2002. TR-104738.
24. Wind Power For Pennies, Technology Review, July/August 2002.
25. Characterizing the Impacts of Significant Wind Generation Facilities on Bulk Power System
Operations Planning, Xcel Energy-North Case Study, Final Report, Utility Wind Interest
Group, December 2003: 1004807.
26. California Wind Energy Forecasting System Development and Testing Phase 1: 12-Month
Testing. EPRI, Palo Alto, CA, California Energy Commission, Sacramento, CA: March
2003. 1007338.
3-85
10581090
Wind Power
Testing. EPRI, Palo Alto, CA, California Energy Commission, Sacramento, CA: July 2003.
1007339.
28. Texas Wind Energy Forecasting System Development and Testing, Phase 1: Initial Testing.
EPRI, Palo Alto, CA, U.S. Department of Energy, Washington, DC: December 2003.
1008032.
29. Wind Energy Forecasting Applications in Texas and California: EPRI-California Energy
Commission-U.S. Department of Energy Wind Energy Forecasting Program: EPRI-
California Energy Commission-U.S. Department of Energy Wind Energy Forecasting
Program. EPRI, Palo Alto, CA: December 2004. 1004038.
30. Texas Wind Energy Forecasting System Development and Testing, Phase 1: 12-Month
Testing. EPRI, Palo Alto, CA, U.S. Department of Energy, Washington, DC: September
2004. 1008033.
31. Wind Energy Forecasting Technology Update: 2004, Palo Alto, CA: April 2005. 1008389.
32. California Regional Wind Energy Forecasting System Development, Volume 1: Executive
Summary. EPRI, Palo Alto, CA, California Energy Commission, Sacramento, CA: 2006.
1013262.
33. California Regional Wind Energy Forecasting System Development, Volume 2: Wind Energy
Forecasting System Development and Testing and Numerical Modeling of Wind Flow over
Complex Terrain. EPRI, Palo Alto, CA, California Energy Commission, Sacramento, CA:
2006. 1013263.
34. California Regional Wind Energy Forecasting System Development, Volume 3: Wind Tunnel
Modeling of Wind Flow over Complex Terrain. EPRI, Palo Alto, CA, California Energy
Commission, Sacramento, CA: 2006. 1013264.
35. California Regional Wind Energy Forecasting System Development, Volume 4: California
Wind Generation Research Dataset (CARD). EPRI, Palo Alto, CA, California Energy
36. Engineering and Economic Evaluation of Utility-Scale Wind Power Plants. EPRI, Palo Alto,
CA: 2009. 1017599.
3.14.1 DOE-EPRI Wind Turbine Verification Program Reports

37. Tennessee Valley Authority Buffalo Mountain Wind Power Project First-and Second-Year
Operating Experience: 2001-2003. EPRI, Palo Alto, CA: December 2003. 1008031.
38. Tennessee Valley Authority Buffalo Mountain Wind Power Project Development. EPRI,
Palo Alto, CA: January 2003. 1004207.
39. Big Spring Wind Power Project Third-Through Fifth-Year Operating Experience:
2001-2004. EPRI, Palo Alto, CA: October 2004. 1008384.
40. Big Spring Wind Power Plant Second-Year Operating Experience: 2000-2001. EPRI,
Palo Alto, CA: December 2001. 1004042.
41. Big Spring Wind Power Plant First-Year Operating Experience: 1999-2000. EPRI,
Palo Alto, CA: December 2000. 1000958.
3-86
10581090
Wind Power
42. Project Development Experience at the Big Spring Wind Power Plant. EPRI, Palo Alto,
CA: December 1999. TR-113919.
43. Lessons Learned at the Iowa and Nebraska Public Power Wind Projects. EPRI, Palo Alto,
CA: October 2000. 1000962.
44. Iowa/Nebraska Distributed Wind Generation Projects First-and Second-Year Operating
Experience: 1999-2001. EPRI, Palo Alto, CA: December 2001. 1004039.
45. Project Development Experience at the Iowa and Nebraska Distributed Wind Generation
Projects. EPRI, Palo Alto, CA: December 1999. TR-112835.
46. Kotzebue Electric Association Wind Power Project Third-Year Operating Experience:
2001-2002. EPRI, Palo Alto, CA: December 2002. 1004206.
47. Kotzebue Electric Association Wind Power Project Second-Year Operating Experience:
2000-2001. EPRI, Palo Alto, CA: December 2001. 1004040.
48. Kotzebue Wind Power Project Initial Operating Experience: 1998-2000. EPRI, Palo Alto,
CA: December 2000. 1000957.
49. Project Development Experience at the Kotzebue Wind Power Project. EPRI, Palo Alto,
CA: December 1999. TR-113918.
50. Wisconsin Low Wind Speed Turbine Project Third-Year Operating Experience: 2000-2001.
EPRI, Palo Alto, CA: December 2001. 1004041.
51. Wisconsin Low Wind Speed Turbine First and Second Year Operating Experience:
1998-2000. EPRI, Palo Alto, CA: November 2000. 1000959.
52. Wisconsin Low Wind Speed Turbine Project Development. EPRI, Palo Alto, CA: December
1998. TR-111438.
53. Green Mountain Power Wind Power Project Third Year Operating Experience: 1999-2000.
EPRI, Palo Alto, CA: December 2000. 1000960.
54. Green Mountain Power Wind Power Project Second Year Operating Experience: 1998-1999.
55. Green Mountain Power Wind Power Project First Year Operating Experience: 1997-1998.
56. Green Mountain Power Wind Power Project Development. EPRI, Palo Alto, CA: December
1997. TR-109061.
57. Central & South West Wind Power Project Third Year Operating Experience: 1998-1999.
58. Central & South West Wind Power Project Second Year Operating Experience: 1997-1998.
59. Central & South West Wind Power Project First Year Operating Experience: 1996-1997.
60. Central and South West Wind Power Project Development. EPRI, Palo Alto, CA: December
1996. TR-107300.
61. DOE-EPRI Wind Turbine Verification Program TVP MI-112231 Status Report, 1998.
3-87
10581090
Wind Power
62. DOE-EPRI Wind Turbine Verification Program TVP MI-112231 Status Report, 2001.
63. Review of Operation and Maintenance Experience in the TVP Program, Report No.
LP-500-28620, 2000.
3.14.2 Other Reports

64. Musial, W. and S. Butterfield, “Future for Offshore Wind Energy in the United States,”
National Renewable Energy Laboratory NREL/CP-500-36313, presented at EnergyOcean
2004, Palm Beach Florida, June 28-30, 2004.
65. Frandsen, Stan, “Design of Offshore Wind Turbines,” Professional Course, Developing
Offshore Wind Energy, European Renewable Energy Centers Agency (EUREC), June 2003.
66. Engineering and Economic Evaluation of Renewable Energy Technology. EPRI, Palo Alto,
CA: 2007. 1012726.
67. Advanced Wind Turbine Technology Assessment - 2010. EPRI, Palo Alto, CA: 2010.
1019772.
68. Wind Power Technology Status and Performance and Cost Estimates – 2009. EPRI, Palo
Alto, CA: 2010. 1020362.
69. Wind Turbine Blade Structural Health Monitoring: Methods and Benefits. EPRI, Palo Alto,
CA: 2010. Product ID 1021655.
70. Qu, Ronghai. Workshop on next generation Wind Power, Rensselaer Polytechnic Institute.
May 2010. “Development and challenges of permanent Magnet Wind Generators.” Published
May 12, 2010. Retrieved September 2010 via http://www.rpi.edu/cfes/news-and-
events/Wind%20Workshop/Development%20Challenges%20of%20PM_Generator_RPI_Qu
_v8.pdf
71. de Vries, Eize. Renewable Energy World. “Permanent Solution?” Published April 8, 2010.
Retrieved September 2010 via
http://www.renewableenergyworld.com/rea/news/article/2010/04/permanent-solution
72. Tan, Andrew. Windsystems Magazine. “A direct drive to sustainable wind energy.”
Published March 2010. Retrieved September, 2010 via
http://windsystemsmag.com/media/pdfs/Articles/2010_March/Zenergy_0310.pdf
73. Abrahamsen, A B, Mijatovic, N, Seiler, E, Zirngibl, T, Traeholt, C, Norgard, P B, Pedersen,
N F, Andersen, N H, Ostergard, J, “Superconducting Wind Turbine Generators.”
Superconducting Science and Technology Vol. 23, 2010. Retrieved September, 2010 via
http://dx.doi.org/doi:10.1088/0953-2048/23/3/034019.
74. Tyler, Dave. Clean Technica. “American Superconductor, DOE Study Large Wind Turbine
Design”. Published February 11 2009. Retrieved August 2010 via
http://cleantechnica.com/2009/02/11/american-superconductor-doe-to-study-large-wind-
turbine-design/
75. Chow, Raymond, and van Dam, C. P. “A Focus on the Flow in the Inboard Part of the
Blade.” Presented at the Sandia National Laboratories Blade Workshop, July 20-22, 2010.
Retrieved August 2010 via http://windpower.sandia.gov/2010BladeWorkshop/PDFs/2-2-A-
2%20VanDam.pdf
3-88
10581090
Wind Power
76. Fairley, Peter. Technology Review. “Stealth-Mode Wind Turbines”. Published November 2,
2009. Retrieved August 2010 via http://www.technologyreview.com/energy/23837/
77. Renewable Energy Focus. “Updated: QinetiQ and Vestas Test ‘Stealth Technology’ for Wind
Turbines”. Published October 26, 2009. Retrieved September 9, 2010 via
http://www.renewableenergyfocus.com/view/4715/updated-qinetiq-and-vestas-test-stealth-
technology-for-wind-turbines/
78. Karal, Petter. Stancich, Rikki. Wind Energy Update. “What’s New about Gravity Base
Foundations for Offshore Wind?”. Published Interview. Published September 3, 2010.
Retrived September 7, 2010 via http://social.windenergyupdate.com/qa/what%E2%80%99s-
new-about-gravity-base-foundations-offshore-wind
79. Jonkman, J., Matha, D. National Renewable Energy Laboratory. “A Quantitative Comparison
of the Responses of Three Floating Platforms”. March 2010. Report NREL/CP-500-46726.
Retrieved August 2010 via http://permanent.access.gpo.gov/lps123972/46726.pdf
80. Lombardi, Candace. CNET News, Green Tech. “Norway Oil Giant Floats Idea for Bobbing
Windmills”. Published August 19, 2010. Retrieved September 8, 2010 via
http://news.cnet.com/8301-11128_3-20014140-54.html
3-89
10581090
10581090
4
BIOMASS ENERGY
Installed Capacity (est.) • 55,000+ MW worldwide (2008)

• 11,940 MW in United States (2008)
Technology Readiness • Combustion systems are commercial, widespread; gasification/pyrolysis
in initial deployments.
• High confidence in cost estimates and projections for combustion
systems.
• Moderate to low confidence in cost estimates and projections for
gasification and pyrolysis systems.
Environmental Impact • Atmospheric and solid emissions from combustion and gasification
technologies controllable using commercial technology.
• When fuel is grown sustainably, no net increase in CO2 emissions.
Can help meet local solid waste disposal challenges.
Economic Status • Competitive or near-competitive in many markets, particularly if CO2
fossil emissions costs are externalized.
• Particularly attractive in cogeneration applications.
• Cost of securing and transporting fuel remains a challenge on case-by-
case basis.
Policy Status • Not consistently addressed by state RPS or energy policies
(except digester gas and ethanol production).
• Cap-and-trade/tax legislation for GHG gases in flux.
• Combustion technologies can encounter resistance from the public,
policymakers, and environmental groups; education is needed.
• Policies and government R&D investment tend to favor biomass to
liquid fuels over other biomass energy.
• European Union policies and regulations have produced a robust and
growing biomass-to-power industry.
Trends to Watch • Development of cellulosic ethanol and related impact on biomass
availability.
• Repowering older coal units for biomass firing.
• Progress toward more efficient and economical gasification.
• Dedicated high-yield fuel crops for bio-fuels and power.
• Resolution on carbon neutrality of biopower
• Biomass pre-treatments and international biomass trade.
• Biomass certification standards.
4-1
10581090
Biomass Energy
4.1 Introduction
Biomass fuels produced by living plant and animal matter provide electric utilities and other
electricity generators with dispatchable renewable power. For the most part, biomass fuels can
be produced, concentrated, and stored for use when it is economic to do so. The U.S. Energy
15
Information Agency (EIA) estimates that biomass supplied 3.8 Exajoules (EJ) or 3.615x10 Btu
(Quads) of energy to the U.S. economy in 2007. Altogether, biomass supplies some 21 EJ to the
world’s economy—largely in North America, Scandinavia, China, and numerous developing
economies. Over the past decade, annual biomass consumption in the United States has ranged
from 2.77 EJ to 3.8 EJ (2.627 to 3.615 Quads), depending upon economic activity among the
wood products industries, agribusinesses, electric utilities, and other industries. Biomass
consumption in the U.S. economy has been generally quite stable, though it did climb more than
7% in 2007 due to increased demand for biofuels in the transportation sector.
Biomass fuels are largely industrial fuels. Table 4-1 shows the distribution of biomass fuels
among the economic sectors, based on EIA data. These data do not include the more than 0.66 EJ
of biomass used in transportation in the form of ethanol added to gasoline and biodiesel added to
diesel fuel.
Table 4-1
Biomass Energy Consumption in the U.S. Economy by Sector and Type, 2007 (Exajoules)
Biomass Type Economic Sector
Residential and Commercial Industrial Electricity Total

Wood and Wood Waste 0.554 1.537 0.194 2.285
MSW/Landfill Gas 0.032 0.1 0.235 0.367
Other Biomass 0.009 0.47 0.022 0.501
Total 0.595 2.107 0.451 3.153
Source: EIA
In the electricity sector, use of biomass for power generation is expected to double every 10
years through 2030. Biopower uses bioenergy systems to generate electricity through the
following methods:
• Direct firing in dedicated 100% biomass fueled boilers;
• Cofiring where biomass substitutes a portion of coal in existing power plants;
• Repowering existing coal boiler to 100% biomass;
• Gasification, which converts bio-fuels or biomass into carbon monoxide and hydrogen by
reacting the raw materials at high temperatures with a controlled amount of oxygen and/or
steam;
• Pyrolysis, which converts biomass into easily stored and transported liquid by subjecting the
feedstock to high temperatures in the absence of oxygen;
• Anaerobic digestion processes, which capture methane from decomposing bio-matter.
4-2
10581090
Biomass Energy
Figure 4-1 illustrates the percentage share of renewable electricity (in billions kWh) that biomass
is expected to contribute through 2030.
Figure 4-1
Renewable Electricity Projections (Billion kWh/yr) (Source: EIA, Energy Outlook 2009)
Historically, biomass consumption for energy use has remained at low levels. However, as
Figure 4-1 illustrates, biomass is expected to be one of the most important sources of U.S.
renewable electricity generation through 2030, trailing only hydropower, based on projected
kilowatt-hours delivered. In fact, biomass is predicted to grow from 11% of total renewable
electricity generation in 2007 to roughly 32% in 2030. Electricity generation from both dedicated
and cofired biomass is meanwhile estimated to grow from 39 billion kWh in 2007 to 231 billion
kWh in 2030.
Furthermore, in scenarios that reflect the impact of a 20% federal renewable portfolio standard
(RPS) and in scenarios that assume carbon dioxide reduction requirements based on international
agreements, electricity generation from biomass is projected to increase more substantially. In
these scenarios, environmental benefit is the primary reason driving increased biomass
utilization.
Compared with coal, biomass feedstocks have lower levels of sulfur, sulfur compounds and
mercury, and demonstrations have shown that biomass cofiring with coal can also lead to lower
nitrogen oxide emissions. Perhaps the most significant environmental benefit of biomass,
however, is potential reduction in carbon dioxide emissions. The closed-loop process harnessed
to cultivate biomass, in which power is generated using feedstocks that are grown specifically for
the purpose of energy production, virtually offsets the CO2 emissions resulting from biomass
firing. Many varieties of energy crops are being considered—including hybrid willow, switch
grass, and hybrid poplar—and may become commercially available in the United States as early
as 2010-2011.
4-3
10581090
Biomass Energy
Feedstock cost and feedstock availability are the main barriers to widespread use of biomass for
power generation. In the long term, bio-power potential will depend on technology advances as
well as on the level of competition for feedstocks and with food and fiber production for arable
land use. Land-based competition may not be an issue until 2020, however, as enough land that
has been idled by farm programs appears to exist to satisfy biomass crop growing needs. Once
biomass crop acreage requirements exceed 30 million acres, which is likely by 2020, according
to Oak Ridge National Laboratory (ORNL) estimates, biomass crops will begin to compete with
traditional agricultural crops.
In addition, risks associated with widespread use of biomass include intensive farming, over-use
of fertilizers, chemicals use, and bio-diversity conservation. They pose a challenge to increased
biomass usage. Certifications attesting the sustainable production of biomass feedstock are
needed to, for example, improve upon land management techniques. While over-exploitation of
biomass resources in developed and developing countries should be avoided, sustainable biomass
can offer an important, productive use for marginal lands and bring socio-economic benefits to
many rural regions of the world.
Separately, innovative densification pre-treatments, such as pelletization, torrefaction, and

conversion into bio-oil (e.g. by pyrolysis) may help overcome the economically and
environmentally challenging logistics of long-distance transportation. The commercial
emergence of these pre-treatment methods could help facilitate a global international trade by
reducing logistic costs.
4.1.1 Basic Issues Associated with Biomass Fuel Utilization
Biomass fuels exhibit certain fundamental differences from other fossil fuels. Typically,
biomass fuels are either gathered or harvested from diffuse sources and concentrated at
a given location. Consequently, there are practical limits on the quantities that can be obtained at
any location without experiencing significant cost pressures. This phenomenon limited the
capacities of the early iron furnaces fueled by charcoal in Pennsylvania and other states in the
late 18th century and early 19th century [1], and remains an issue today, though
at a significantly expanded scale. This is in distinct contrast to fossil fuels such as coal that are
produced in centralized locations and distributed to users such as power plants.
Biomass fuels currently used as fuels are, almost exclusively, residues from other processes.
They may be wood processing residues such as hog fuel, bark, sawdust, or spent pulping liquor.
They may be agricultural and agribusiness residues such as bagasse. They may be wastewater
treatment gas or landfill gas. These are commodities that are presently outside the commercial
mainstream. In many cases, these commodities have both material and energy value. Wood
waste markets, for example, can include mulch for urban areas, bedding for livestock and
poultry, feedstocks for materials such as particleboard, and feedstocks for niche chemical and
related products. As a result, fuel pricing is highly sensitive to locale and the competitive
pressures of local and regional economies.
Significant efforts have also been made to develop short-rotation biomass crops to be used
exclusively as fuel or energy feedstocks. To date, these have not produced competitively priced
fuels, though development continues.
4-4
10581090
Biomass Energy
4.1.2 Technology Considerations for Using Biomass Fuels
Biomass-to-power technologies can be divided into two basic categories: 1) existing

technologies that are either commercialized or nearly ready for commercial implementation,
and 2) goal technologies that have been proposed and are of long-term interest. It is important
to focus on the first category, for which economics are reasonably well defined. Time frames
for commercial availability and costs associated with goal technologies are more speculative
in nature and can only be discussed in such terms.
4.1.2.1 Commercially Available Technologies
Commercially available technologies include stand-alone condensing power plants, such as

those owned by electric utilities and independent power producers (IPPs). Stand-alone power
plants fired by wood waste and certain agricultural wastes were developed extensively during the
early years of the Public Utilities Regulatory Policies Act (PURPA). Stand-alone power plants
fired with municipal solid waste (MSW) and landfill gas have also been developed. For example,
landfill gas is used to fuel a 50-MWe steam boiler installation at the Puente Hills, Los Angeles
landfill. The energy installation at Puente Hills also includes a combustion turbine and internal
combustion engines.
Equally significantly, commercially available technologies include Rankine-cycle cogeneration

facilities such as those owned and operated by companies in the forest products industry, the
sugar industry, and other manufacturing interests. Such installations couple a medium-pressure
boiler with main steam conditions, typically 600 psig/750°F, 850 psig/825°F, 1250 psig/950°F,
and 1450 psig/1005°F to either backpressure turbines or automatic extraction turbines. Process
steam is typically generated at 50 psig, 150 psig, or 450 psig depending on the application. The
energy released when steam is expanded from throttle conditions to exhaust conditions is used
to generate electricity. In selected cases, such as the Snohomish County Public Utility District in
Washington, utilities team with forest products industry firms to jointly develop such
cogeneration capabilities.
Biomass cofiring technologies have been employed at several locations. For example, wood
waste can be blended with coal on the main belt leading to the Willow Island Generating Station
#2 boiler, a 188-MWe cyclone boiler owned and operated by Allegheny Energy Supply Co.,
LLC. Sawdust and other biofuels have been ground and separately injected into the 105-MWe
boiler at Greenidge Station in a project developed in the mid-1990s by New York State Electric
and Gas (NYSEG) and continued by AES, the current plant owner. Southern Company has
pioneered numerous cofiring approaches at Plant Hammond, Plant Yates, Plant Gadsden, and
other installations. Previously, Northern States Power implemented biomass cofiring at its Allen
S. King Generating Station. The New York Power Authority (NYPA) cofired biodiesel with oil
in its 890-MW Poletti station. Numerous vendors now offer commercial systems.
An example of coupling atmospheric gasification of biomass to electricity-generating Rankine

cycle installations is the Kymijarvi power plant in Lahti, Finland, owned and operated by Lahden
Lampovoima Oy. The Lahti gasification project, an EU Thermie demonstration, involved
numerous partners, including Lahden Lampovoima Oy, Foster Wheeler Energia Oy, VTT (the
Technical Research Center of Finland), Elkraft Power Co., Ltd. of Denmark, and Plibrico Ab
4-5
10581090
Biomass Energy
from Sweden. The project involved installing a Foster Wheeler atmospheric circulating fluidized
bed gasifier capable of converting sawdust, bark, wood chips, plywood trim, particleboard trim,
and recycled fuel (REF) into a producer gas for firing in the coal-fired boiler used for generating
electricity and district heat. Based upon the success of that project and other gasification projects,
atmospheric circulating fluidized bed boilers can now be procured commercially for such
cofiring projects used to generate electricity.
For each of these commercially available technologies, the issues of fuel cost and the availability
of fuel in significant quantities remain driving issues. The process of gathering and concentrating
the fuel in a single location is a primary consideration. All of these commercially available
technologies exhibit technical advantages and disadvantages associated with using biomass fuels.
Such technical considerations, along with economics, are discussed in subsequent sections of this
report.
4.1.2.2 Goal Technologies
The technologies that have long-term promise, but remain beyond current technical and
economic feasibility on a proven basis, are many and varied.
The practice of raising silvicultural or agricultural crops as a fuel supply is one such technology.
This technology was employed, as mentioned previously, by the early iron industry in the
fledgling United States. It has appeared significantly in the literature since the early 1970s (see,
for example, Szego and Kemp [2], Bethel et al. [3] and Henry [4]). Government research in this
area has been significant over a long period of time. To date, however, experience has shown
that biomass cannot be grown and harvested at costs close to those associated with alternative
combustible resources. Further, there are significant economic issues associated with competition
for the resource, once grown, that make this a goal technology rather than a near-term
commercial reality. In the United Kingdom, however, government policies are in place which
may offer a significant role for dedicated biomass. The success of this program will be watched
carefully in the coming years.
A program in the United States is also being watched carefully. In May, 2006, the Biomass
Investment Group announced they had signed a power purchase agreement for 130 MW of
power derived from the gasification of a dedicated herbaceous crop (called E-Grass by the
developer). Florida was identified as the site of this development, but the E-Grass species was
not identified in public documents. After a period of relative inactivity, including a change in
ownership, it is unclear whether development work has been restarted. A separate project is
being pursued by Xcel Energy to convert a coal boiler at the utility's Bay Front power plant with
a 20-MW fluidized bed boiler.
Pressurized gasification with hot or warm gas clean-up, coupled with combustion turbines
in integrated gasification combined-cycle (IGCC) applications, also remains a long-term
possibility. Scale limitations severely constrain the economics of this technology. Further, hot
or warm gas clean-up remains an elusive target. Pyrolysis to produce liquid fuels is yet another
goal technology, but with serious practical problems.
4-6
10581090
Biomass Energy
Atmospheric (or near-atmospheric) gasification with warm gas cleanup in a configuration that
provides fuel to reciprocating engines and district heating is also considered a goal technology,
though is much nearer to commercial readiness than pressurized biomass gasification used in
combined cycle mode. Demonstrations at Harboøre, Denmark, featuring the Vølund gasifier,
provide evidence of approaching commercial deployment. This technology category is addressed
in more detail in 4.4.5.
4.1.2.3 Prospectus
Within this expanding framework, the generation of electricity from biomass can be evaluated.
In order to make such an evaluation, this chapter considers the physical and chemical
characteristics of various biomass fuels, and the technical characteristics and costs of various
biomass technologies.
4.1.2 Characteristics of Representative Biomass Fuels
Biomass is commonly defined as material derived from living organic matter (e.g., trees, grasses,
animal manure). While there are alternate political/regulatory definitions of biomass, from the
physical and chemical perspectives (including the biochemical perspective) this definition
appears to be most accurate. The consequence is that fractions of municipal solid waste—paper,
wood waste, food waste, yard waste—are forms of biomass fuel. This definition also supports
the inclusion of both residues and potential energy crops as biomass fuel. Consequently, biomass
includes wood and wood waste, herbaceous crops and crop wastes, agribusiness wastes (e.g.,
food processing wastes such as bagasse), animal manures and methane-rich gas from anaerobic
digestion of such fecal mater, methane-rich gases generated from wastewater treatment plants
and landfills, and miscellaneous related materials.
4.2 Biomass Fuel Resources
Numerous national studies of biomass availability have been conducted (see, for example,
Walsh et al. [5]). Such studies estimate the availability of biomass as a function of price. Oak
Ridge National Laboratory estimates that over 5 EJ (or 5 Quads) of biomass as residue could
be available at a price of $50/dry ton (approximately $3/GJ or $3/106 Btu) nationally. This study
focused upon forestry and wood residues, crop residues, and energy crops. It did not address
animal manures, landfill gas, wastewater treatment gas, or biomass components of municipal
solid waste. If those wastes are added in, the estimated quantity of residue-based biomass fuel
available at some price could well be on the order of 10 EJ (10 Quads).
A separate study conducted by the EIA [42] meanwhile estimates that the United States will have
an upper limit of 7.1 quadrillion Btu of biomass available at prices of $5 per million Btu or less,
by 2020. The available low-cost feedstock ($1 per million Btu) is almost exclusively composed
of urban wood waste and mill residues. At about $2 per million Btu, agricultural residues
become viable as a second source of biomass. And energy crops and forestry residues begin to
make significant contribution at prices around $2.40 per million Btu or higher.
4-7
10581090
Biomass Energy
As a practical matter, national estimates have policy implications but are otherwise not useful
to utilities and generating stations that require local availability estimates when considering
firing biomass singly or in combination with fossil fuels. Local estimates are based upon forest
industry or agricultural industry activity, plus local markets for forest industry residues.
Residues include bark, sawdust, shavings, chips, hog fuel, logging residues, silvicultural residues
(e.g., pre-commercial thinnings), dead and diseased timber, and fire-damaged timber that cannot
be salvaged. Local residues may also include an array of agricultural materials such as:
• Corn stover
• Corn cobs
• Out-of-date seed corn
• Wheat straw
• Rice straw
• Rice hulls
• Vineyard prunings
• Orchard prunings
• Oat hulls
• Bagasse.
Determination of locally available biomass depends upon local forest industry or agricultural
activity, local competing markets for the material, and transportation systems available. Local
conditions vary substantially. For example, in the Chicago area, bark commands a premium price
as mulch. In some areas with high concentrations of poultry farms, sawdust and rice hulls both
command good prices as bedding material. Chips command prices greater than fuel prices in
areas that are within economical haul distances of pulp mills or manufacturers of oriented strand
board (OSB), medium density fiberboard (MDF), or related products. Other competing uses that
have been encountered include manufacturing charcoal, particleboard, and specialty products.
When considering agricultural residues, seasonality becomes a concern. For wastewater

treatment gas, the population served by the wastewater treatment plant is the primary factor and
can include industries as well as residential populations. Similarly, landfill gas utilization
depends upon landfills serving large populations. As a consequence, California has more than
200 MWe of generation from landfill gas [6]. In all cases, however, local resource availability is
the governing factor.
Local resource availability is critical because biomass fuels have limited transportability. Green
wood wastes typically have bulk densities of 0.21 to 0.26 kg/m3 (16 to 20 lb/ft3). This compares
to coals with bulk densities of 0.68 to 0.78 kg/m3 (52 to 60 lb/ft3). Crop wastes can have bulk
densities of 0.11 to 0.13 kg/m3 (5 to 8 lb/ft3). Landfill and wastewater treatment gases are
typically saturated with water, and have heating values that are about half of that associated with
natural gas (on a dry gas basis). Further, landfill and wastewater treatment gases are not
compressed, while natural gas is transported under significant pressure. The typical estimated
4-8
10581090
Biomass Energy
maximum haul distances for woody biomass are on the order of 50 to 75 miles (80 to 120 km);
economical transport distances for other biomass fuels are significantly lower.
Energy crops are just beginning to be commercially grown in the United States, and RD&D is
underway to explore optimal energy crop species. The main potential energy crop species being
considered are hybrid poplar, hybrid willow, and switchgrass. POLYSYS, an agricultural policy
simulation model of the U.S. agricultural sector, assumes that energy crops production will be
limited to areas that are climatically suited for their production, thus excluding all states in the
Rocky Mountain and Western Plains regions. Future genetic improvements in energy crops may
extend this range. POLYSYS excludes irrigation as a viable management practice for energy
crops production.
Crop, forest, and primary mills provide roughly 70% of the total biomass resource. The single
largest source of biomass is crop residue. Perennial crops have the largest growing potential for
energy production. Dedicated energy crops (switchgrass, willow, hybrid poplar) can often be
economically grown on land that is not suitable for conventional crops and can provide erosion
protection for agricultural set-aside or Conservation Reserve Program (CRP) lands. Figure 4-2
breaks out the various sources used for biomass generation.
Switchgras s on
CRP Lands
20%
Methane from
Dom es tic
Was tewater
0%
Crops Res idues
Methane from
37%
Landfills
3%
Urban Wood Methane from

7% Manure
Secondary Mill Managem ent
1% 1%
Fores t Res idues
Prim ary Mill
13%
18%
Figure 4-2
Breakdown of Biomass Feedstock
4-9
10581090
Biomass Energy
The key to producing low-cost plantation feedstocks is land availability and quality, which to a
significant extent determine the degree of site preparation necessary; the choice of species and
rotations (cutting cycles); required cultural management and soil amendments (fertilization,
weed control, etc); and fuel transport and logistics. Land and site quality also defines biomass
productivity, the major determinant of total feedstock cost. In addition, productivity is a key
factor in determining the total size of the plantation, annual feedstock production, and the size of
the conversion facility that can be supported. Figure 4-3 shows the effect of biomass plant scale
on feedstock requirements and transport distances.
Figure 4-3
The Effect of Biomass Plant Scale on Feedstock Requirements and Plantation Area
Required [43]
Greater plantation productivity directly translates into reduced land requirements. Higher
conversion efficiencies translates into more installed plant capacity for a given feedstock
requirement or less required plantation area for a given installed capacity.
4.2.1 Biofuels vs. Biomass Electricity
A 2009 study performed by researchers at Stanford University, the University of California

Merced and others [44] concludes that, on average, using biomass to produce electricity is 80%
more efficient than transforming the biomass into bio-fuel. In addition, the electricity option was
deemed to be twice as effective at reducing greenhouse gas emissions. The analysis covered a
range of harvested crops, including corn and switchgrass, and a number of different energy
4-10
10581090
Biomass Energy
conversion technologies. Data collected were applied to electric and combustion-engine versions
of four vehicle types, and their operating efficiencies were estimated during city and highway
driving. The study accounted for the energy required to convert the biomass into ethanol and
electricity, as well as for the energy intensiveness of manufacturing and disposing of each
vehicle type. Bioelectricity far outperformed ethanol in most scenarios. (It some cases it was
found to be 80% more efficient.)
4.2.1.1 Refuse Derived Fuels (RDF)
Refuse Derived Fuels (RDF) refers to the segregated high calorific fraction of processed
Municipal Solid Waste (MSW). Other terms related to MSW-derived fuels are Recovered Fuel
(REF), Packaging Derived Fuel (PDF), Paper and Plastic Fraction (PPF), and Processed
Engineered Fuel (PEF). RDF fuel typically consist of pelletized or fluff MSW that is the by-
product of a resource recovery operation. Processing removes ferrous materials, glass, grit, and
other materials that are not combustible. The remaining material is then sold as RDF.
The RDF can be used in one of several configurations, including:

• Dedicated RDF boilers designed with traveling grate spreader-stockers;
• Cofiring with coal or oil in multi-fuel boilers;
• Dedicated RDF fluidized boiler.
When RDF is fired as a supplemental fuel in coal-fired boilers, experience has shown that very
high quality RDF (such as PEF) with heating values in the range of 8,500 to 11,500 Btu/lb (as-
received basis) can successfully contribute up to 30% of the input energy.
Various qualities of RDF can be produced, depending on the needs of the user or market. A high-
quality RDF would possess a higher heating value and lower moisture and ash contents. PEF, for
example, is high-quality RDF in which toxic and high-ash components have been removed, as
well as metals, rock, glass, electronics, sheet rock, plaster, and other high ash non-combustible
components. PEF exists as a fluff, while densified PEF (dPEF) is the same product packed as
cubes or pellets for easy storage and transportation.
PEF composite fuels can be formulated to substitute for other solid fuels without modification to
the combustion system. E-Fuel is an example of a commercial PEF fuel made with a mixture of
70% paper-making waste sludge, 25% waste from a low-density polyethylene plastic (LDPE)
used to line food cartons, and 5% coal fines. This mixture is blended and then pelletized. Other
PEF fuels come in the form of briquettes made from coal fines and other waste materials such as
wood.
Table 4-2 below shows typical fuel properties of RDF recovered from municipal solid waste.
4-11
10581090
Biomass Energy
Table 4-2
Refuse Derived Fuel Properties [45]
Fuel HHV as Received Moisture % Ash % S, dry basis Cl %wt

(Btu/lb) %wt
RDF 5,500-6,500 3-35 8-25-22 0.1-0.50 0.1-1
PEF 6,500-16,000 3-20 2-15 0.02-0.20 0.03-0.5
PEF-Paper Pellets 8,500-11,500 2-4 2-4 0.02-0.2 <0.2
Bituminous Coal 11,000-14,500 2-20 3-16 0.5-4.7 0.01-0.90
MSW 4,500-5,500 15-50 18-30 0.1-0.5 0.1-1
RDF has relatively high concentrations of paper and plastics, both of which have a high calorific
value (paper about 17,500J/g; plastic about 37,000J/g). In comparison with most coals, it also
may contain materials that have a relatively high percentage of ash, can be damaging to burners
and boilers, and can impact the quality of exhaust gases. For example, RDF typically contains
materials with substantial concentrations of chlorides that may induce a corrosive effect on boiler
tubes. The presence of small particles of metal and of glass fines in RDF can present problems in
the combustion system. Moreover, excluding these small particles from RDF is difficult.
RDF combustion can produce four to six times as much ash as coal combustion. Consequently,
when using RDF in cofiring with coal, some provisions must be made for handling the additional
burden of ash.
An important prerequisite for the successful combustion of RDF/PEF in a combustion system,

whether fired solely or in combination with another fuel, is developing the proper fuel
specification, which should be provided to the RDF supplier by the combustion system supplier.
The need for the RDF to be compatible with all the applicable elements of the combustion
system cannot be over emphasized. For financial reasons, optimum performance of the
combustion process and thermal conversion system is required. Therefore, the properties of the
RDF must be carefully evaluated and selected. These requirements apply to dedicated RDF
plants and to cofiring with coal.
RDF composites fuels hold strong potential for both reducing emissions from coal-fired boilers
and reducing fuel costs. In order to realize both potentials, sources of low-cost feedstock must be
converted into solid fuel that can be used in existing coal boilers without the need for a capital-
intensive boiler retrofit or a high increase in fuel purchase, transportation, handling and storage
costs. Gasification and pyrolysis technologies can convert MSW into energy sources. However,
these technologies are less proven in operation, even at relatively modest scale.
Federal- and state-level regulatory hurdles currently obstruct the expanded use of PEF. These
issues will need to be addressed by the fuel and energy industries if PEF is to achieve its full
potential. However, some progress is being made in this regard, as several states have approved
the use of PEF fuels through legislative or regulatory action.
4-12
10581090
Biomass Energy
4.2.2 Basic Properties of Solid Biomass Fuels
Fuel properties can be characterized in terms of proximate and ultimate analyses, reactivity
measurements, ash elemental analyses, ash reactivity, and trace metal composition.
4.2.2.1 Proximate and Ultimate Analyses of Biomass Fuels
Research at The Energy Institute of Pennsylvania State University (PSU) has led to
characterization of a broad array of biomass fuels, including the proximate and ultimate
analyses shown in Tables 4-3 through 4-5.
Note that the sawdust in Table 4-3 is mixed hardwood-softwood sawdust. As a practical matter
the difference between softwood and hardwood is insignificant from a fuels perspective, unless
a fuel loaded with extractives is used, e.g. slash pine. The urban wood waste is clean and does
not include pentachlorophenol or copper chromium arsenate (CCA) treated wood. The
switchgrass shown is from the Plant Gadsden tests of Southern Company.
The data in the tables document the fact that all biomass fuels have similar heating values on a
moisture and ash-free (MAF) basis; however, the various biomass fuels differ with regard to
their relative percentages of holocellulose (cellulose plus the hemicelluloses), lignin, and
extractives.
Table 4-3
Proximate and Ultimate Analyses for Typical Woody Biomass Fuels
Parameter Fuel
Pine Red Oak Fresh Mixed Urban Wood
Pine Chips
Shavings Shavings Sawdust Waste (*)
Moisture % 45.0 45.0 28.8 40.0 30.8
Proximate Analysis (wt % dry basis)
Volatiles 84.7 84.7 79.5 80.0 76.0
Fixed Carbon 15.2 15.2 19.0 19.0 18.1
Ash 0.1 0.1 1.5 1.0 5.9
Ultimate Analysis (wt % dry basis)
Carbon 49.1 49.1 51.6 49.2 48.0
Hydrogen 6.4 6.4 5.8 6.0 5.5
Nitrogen 0.2 0.2 0.5 0.4 1.4
Sulfur 0.2 0.2 0.0 <0.1 0.1
Oxygen 44.0 44.0 40.6 43.0 39.1
Ash 0.1 0.1 1.5 1.0 5.9
HHV (Btu/lb) 8502 8502 8069 8400 8364
Sources: Miller et al., 2000; Johnson et al., 2001; Tillman, 2001.
Note: (*) Urban Wood Waste shown is from the Bailly Generating Station Tests.
4-13
10581090
Biomass Energy
Table 4-4
Proximate and Ultimate Analyses for Typical Herbaceous Biomass Fuels
Parameter Fuel
Fresh Weathered Reed Canary
Mulch Hay
Switchgrass Switchgrass Grass
Moisture % 15 15 65.2 19.5
Proximate Analysis (wt % dry basis)
Volatile Matter 76.18 81.8 76.1 77.6
Fixed Carbon 16.08 14.8 4.1 17.1
Ash 7.74 3.4 19.8 5.3
Ultimate Analysis (wt % dry basis)
Carbon 46.73 49.4 45.8 46.5
Hydrogen 5.88 5.9 6.1 5.7
Nitrogen 0.54 0.4 1.0 1.7
Sulfur 0.13 0.3 0.1 0.2
Oxygen 38.99 40.6 42.9 40.6
Ash 7.74 3.4 4.1 5.3
HHV (Btu/lb) 7750 8150 7103 8058
Sources: Miller et al., 2000; Johnson et al., 2001.
Table 4-5
Proximate and Ultimate Analyses for Typical Manures
Parameter Fuel
Dairy Tie- Dairy Free- Sheep
Poultry Litter Swine Waste
Stall Manure Stall Manure Manure
Moisture % 64.7 69.8 20.0 47.8 97.8(*)
Proximate Analysis (wt %, dry basis)
Volatiles 76.0 30.1 55.3 65.2 59.6
Fixed Carbon 18.1 7.4 7.7 14.0 7.3
Ash 6.0 62.5 17.0 20.9 33.1
Ultimate Analysis (wt %, dry basis)
Carbon 48.6 22.6 38.1 40.6 38.0
Hydrogen 5.8 2.9 5.6 5.1 5.5
Nitrogen 1.4 1.1 3.5 2.1 3.2
Sulfur 0.1 0.1 0.6 0.6 0.6
Oxygen 38.1 10.8 30.9 30.7 19.6
Ash 6.0 62.5 17.0 20.9 33.0
HHV (Btu/lb) 8203 3644 6399 6895 7328
Source: Miller et al., 2000; Miller et al., 2002.
Note: (*) Swine waste is liquefied for transportation and disposal.
4-14
10581090
Biomass Energy
4.2.2.2 Structural Relationships and Heteroatoms
The proximate and ultimate analyses relate directly to summative analysis—the distribution of
cellulose, hemicelluloses (e.g., galactoglucomannans, arabinogalactan), lignin, and extractives
(e.g., pinoresinol). Typical distributions of these compounds in softwoods on an extractive-free
basis are cellulose, 45% to 50%; hemicelluloses, 25% to 35%; lignin, 25% to 35%. Extractives
comprise 1% to 5% of softwoods. Hardwoods are characterized by somewhat lower
concentrations of lignin. Herbaceous materials also are characterized by lower lignin
concentrations. The resulting fuels are highly reactive, with little aromaticity and a significant
concentration of functional groups such as hydroxyl and methoxyl functionalities. Aromatic
structures, where they occur, exist as single aromatic rings rather than fused aromatic clusters.
All of this contributes to the high volatile/fixed carbon ratio of the biomass fuels.
Structurally, carbon 13 nuclear magnetic resonance (13C NMR) research has shown that woody
and herbaceous biomass has an aromaticity of about 8% in sawdust and 10% in weathered
switchgrass (Johnson et al. [7], Tillman [8]). This contrasts with 57% aromaticity in Black
Thunder PRB coal and 73% aromaticity in Pittsburgh #8 coal. Biofuels typically contain six
aromatic carbons per aromatic cluster while the PRB coals contain approximately 10 aromatic
carbons per cluster and the eastern bituminous coals contain 13 to 15 aromatic carbons per
cluster [7]. Further, structural research shows that the concentration of methoxy functionalities
(-OCH3) represents 8.67% of the carbon atoms in sawdust and 7.87% of the carbon atoms in
weathered switchgrass. Methoxy functional groups provide a useful measure of fuel reactivity.
As a comparison, 4.21% of the carbon atoms in Black Thunder PRB coal are contained in
methoxy functional groups and 2.27% of the carbon atoms in Beulah lignite are contained in the
– OCH3 groups; this percentage drops to 0.79% for western bituminous coals and to 0% for
eastern and Midwestern bituminous coals [7].
The structural analysis is consistent with the coalification results reported by Jenkins et al. [9]
and numerous other authors. Biomass is at the extreme upper right-hand corner on the typical
coalification diagrams plotting hydrogen/carbon atomic ratios on the X-axis and oxygen/carbon
ratios on the Y-axis.
Heteroatom analysis shows significant variability. Nitrogen concentrations, for example, can
vary dramatically. Generally, the woody biofuels have lower fuel nitrogen contents; however,
urban wood waste with high concentrations of glues is a notable exception. Herbaceous materials
tend to have higher nitrogen concentrations, while manures have extremely high nitrogen
concentrations, commonly in the form of free ammonia. Concentrations of inorganic matter
(ash) are typically lowest in woody biomass and highest in manures as well. Sulfur contents can
vary significantly, with very low concentrations in woody and herbaceous materials and higher
concentrations in some manures, though even these materials contain less than 1% sulfur.
The high concentrations of nitrogen in urban wood waste are a function of glues such as urea-
formaldehyde. Such glues are used in the manufacture of various grades of plywood, OSB, and
MDF, as well as particleboard. The higher concentrations of nitrogen in herbaceous crops result
more from fertilization practices. Common fertilizers such as 30-30-30 have high concentrations
of nitrogen, along with potassium and phosphorus.
4-15
10581090
Biomass Energy
Note the extremely high concentrations of inorganic ash in some of the dairy manure and swine
wastes. Table 4-6 shows the concentrations of fuel nitrogen and ash for selected biomass
materials, expressed in lb/106 Btu.
Table 4-6
Nitrogen and Ash Concentrations in Biomass Fuels (Values in lb/106 Btu)
Fuel Nitrogen (lb/106 Btu) Ash (lb/106 Btu)

Pine Chips 0.24 0.12
Pine Shavings 0.24 0.12
Red Oak Shavings 0.61 1.86
Fresh Mixed Hardwood-Softwood Sawdust 0.47 1.19
Urban Wood Waste 1.67 7.05
Fresh Switchgrass 0.70 9.99
Weathered Switchgrass 0.49 4.17
Reed Canary Grass 1.41 5.77
Mulch Hay 2.11 6.58
Dairy Tie-stall Manure 1.71 7.31
Dairy Free-stall Manure 3.02 171
Poultry Litter 5.47 26.6
Sheep Manure 3.04 30.31
Swine Manure 4.37 45.0
Source: Miller et al., 2000.
4.2.2.3 Inorganic Constituents in Biomass
Ash composition is of considerable significance in addressing concerns associated with slagging

and fouling. Traditional ash analyses are presented in Tables 4-7 and 4-8 for representative
biomass fuels and Pittsburgh #8 coal.
Note that the potassium oxide content is always approximately 10%—or greater—for herbaceous
biomass and manures. Woody biomass, on the other hand, typically has lower potassium content
unless it is a short-rotation woody crop grown with an aggressive regime for fuel or fiber
purposes. The higher concentrations of potassium and phosphorus in the herbaceous crop
residues and in the animal manures again result from fertilization practices.
The consequence of the high alkali metal concentrations in the herbaceous materials and
manures is the potential for formation of slagging and fouling deposits. Miles et al. [10] propose
a useful measure of slagging and fouling potential for biomass fuels as follows:
• <0.4 lb alkali (K2O + Na2O)/106 Btu, low slagging and fouling potential
• 0.4 to 0.8 lb alkali/106 Btu, probable slagging and fouling
•
6
>0.8 lb alkali/10 Btu, certain slagging and fouling
4-16
10581090
Biomass Energy
Table 4-7
Ash Analyses of Various Biomass Fuels Compared to Pittsburgh #8 Coal
Parameter Fuel
Pittsburgh Pine Reed Canary Dairy Tie-
Poultry Litter
#8 Coal Shavings Grass Stall Manure
Al2O3 25.34 13.4 1.66 2.26 9.14
BaO – 0.15 0.05 0.02 0.05
CaO 2.28 8.75 9.57 23.3 12.7
Fe2O3 18.34 5.94 1.47 1.37 4.04
K2O 2.22 4.94 18.1 10.7 9.94
MgO 0.82 3.35 5.29 8.91 4.01
MnO – 0.49 0.11 0.14 0.36
Na2O 0.25 1.38 2.34 7.04 3.60
P2O5 0.4 1.44 13.8 14.7 14.0
SiO2 48.2 57.2 43.0 26.0 39.4
SO3 0.67 0.05 0.02 0.14 2.58
SrO – 0.80 0.11 0.11 0.03
TiO2 – 1.16 4.99 5.08 0.51
Base/Acid Ratio 0.325 0.339 0.741 1.539 0.699
Source: Miller et al., 2002; Miller and Miller, 2002.
Table 4-8
Ash Analyses of Switchgrass and Other Herbaceous Crops
Parameter Fuel
Wheat
Switchgrass Switchgrass Alfalfa Stems Rice Straw
Straw
SiO2 65.18 65.42 1.44 55.70 73.00
Al2O3 4.51 6.98 0.60 1.80 1.40
TiO2 0.24 0.34 0.05 0.00 0.00
Fe2O3 2.03 3.56 0.25 0.70 0.60
CaO 5.60 7.14 12.90 2.60 1.90
MgO 3.00 3.17 4.24 2.40 1.80
Na2O 0.58 1.03 0.61 0.90 0.40
K2O 11.60 7.00 40.53 22.80 13.50
P2O5 4.50 2.80 7.67 1.20 1.40
SO3 0.44 2.00 1.60 1.70 0.70
CO2 0.00 0.00 17.44 0.00 0.00
Base/Acid Ratio 0.33 0.30 28.00 0.51 0.24
Source: Prinzing, 1996; Baxter et al., 1996a; Bryers, 1993.
4-17
10581090
Biomass Energy
Table 4-9 presents the Miles et al. [10] slagging and fouling index for selected biomass fuels.
Note the comparison of the manures—tie stall dairy manure and chicken litter—to woody
biomass and herbaceous biomass.
Table 4-9
Slagging and Fouling Index for Selected Biomass Fuels
Fuel Index (lb/106 Btu of K2O and Na2O)
Pine Shavings and Chips 0.01 – 0.02
Fresh Mixed Hardwood/Softwood Sawdust 0.08 – 0.12

Clean Urban Wood Waste 0.45
Alfalfa Stems 3.38
Fresh Switchgrass 0.80 – 1.22
Weathered Switchgrass 0.33 – 0.51

Reed Canary Grass 1.18
Tie Stall Manure 1.30
Chicken Litter 3.60

Sources: Miller et al. (2002); Prinzing (1996); Baxter et al. (1996a); Miles et al. (1993).
Trace metal concentrations in biomass are typically lower than in coal. This has been shown
with repeated testing (see Tillman [8] [11], Payette et al. [12], and Hus and Tillman [13]). In
particular, mercury concentrations, expressed as lb/106 Btu or mg/kJ—are significantly lower for
biomass than for coal [12].
Trace metals in wood waste, agricultural materials, and other biomass fuels are a reflection
of where they came from. Trees grown in areas where there are high concentrations of trace
metals will exhibit somewhat elevated concentrations of same. Herbaceous crops will exhibit
metal concentrations associated with fertilization practices. Urban wood waste will exhibit
metal concentrations associated with glues, laminates, paints, coatings, and related materials
(see Tillman [14]). Recent studies have shown that mercury concentrations in sawdust are
typically 0.001 to 0.002 mg/kg in the as-received wood, or 0.002 to 0.004 mg/kg in the wood
waste on an oven dry basis [15]. Studies of urban wood waste have shown the following trace
metal concentrations in mg/kg of fresh material [13]:
• Arsenic: 2.143 mg/kg
• Chromium: 6.570 mg/kg
• Lead: 2.923 mg/kg
• Mercury: 0.012 mg/kg
• Nickel: 2.643 mg/kg
• Vanadium: 3.06 mg/kg
4-18
10581090
Biomass Energy
Previous studies [16] have shown wide ranges in trace metal concentrations, as shown in
Table 4-10. These ranges reflect differences in the originating environment of the wood and
wood waste. Table 4-11 shows ranges measured for wood waste burned at a pulp mill in the
Pacific Northwest.
Table 4-10
Ranges in Concentrations of Trace Metals in Woody Biomass (mg/kg in Dry Wood)
Metal Mean Range
Arsenic 6.5 3 – 10
Barium — 0.5 – 910

Cadmium 10.4 3 – 26
Chromium 31.4 9.1 – 92
Lead 72 38 – 127
Nickel 36.6 11.6 – 50

Vanadium 53 27 – 79
Zinc 500 200 – 794

Source: Campbell, 1990.
Table 4-11
Ranges in Concentrations of Trace Metals in Woody Biomass Burned at a Pulp Mill in the
Pacific Northwest (mg/kg in Dry Wood)
Metal Mean Range
Arsenic 0.475 BDL – 1.8
Barium 51.5 43 – 66
Beryllium BDL BDL

Cadmium BDL BDL
Chromium 128.4 16 – 524

Chromium +6 0.063 BDL – 0.3
Lead 2.71 1.1 – 6.1
Nickel 137.3 BDL – 556
Zinc 99 35 – 207
Source: Tillman, 1994.
Trace metals in herbaceous crops and animal manures exhibit the same wide variability. Again,
the originating and processing environments govern trace metal concentrations.
4-19
10581090
Biomass Energy
4.2.3 Performance Characteristics of Biomass Fuels
Performance characteristics of biomass fuels are critical to understanding their behavior in

energy production systems. Fuel volatility, nitrogen evolution patterns, and ash characteristics
are among the parameters of most significance.
4.2.3.1 Biomass Fuel Volatility
Biomass fuels exhibit significant volatility measured through conventional analyses by the
volatile/fixed carbon (V/FC) ratio of the proximate analysis and the hydrogen/carbon (H/C) and
oxygen/carbon (O/C) atomic ratios derived from the ultimate analysis. Representative biomass
fuels exhibit a highly consistent, very volatile pattern. Typical volatility measures for the fuels
shown in Tables 4-3 through 4-5 are presented in Table 4-12.
Table 4-12
Volatility Measures for Representative Biomass Fuels
Volatility Measure Statistical Measure
95% Confidence
Average Standard Deviation
Interval
Volatile/Fixed Carbon Ratio 5.14 1.27 2.65 – 7.63
Hydrogen/Carbon Atomic Ratio 1.51 0.12 1.27 – 1.75
Oxygen/Carbon Atomic Ratio 0.59 0.10 0.39 – 0.79
There is a reasonably consistent pattern of volatility measured through conventional techniques.

Further, these measures are significantly higher than the volatility measures for the range of
coals burned in the United States and throughout the world. Eastern bituminous coals from the
Pittsburgh Seam typically have V/FC ratios of 0.40, H/C ratios of 0.69, and O/C ratios of 0.03.
Illinois basin coals have V/FC ratios of 0.70, H/C ratios of 0.81, and O/C ratios of 0.06. PRB
coals from Wyoming can have V/FC ratios of 0.84, H/C ratios of 0.85, and O/C ratios of 0.19
[17].
Alternative fuel analyses provide significantly more information concerning the volatility
of biomass fuels in relation to coal. These analyses are directly related either to the chemical
structure of the fuel or to their performance in combustion environments. Such alternative
measures include:
• Maximum volatile yield in drop tube reactor (DTR) experiments, in an inert environment,
including both the percentage yield and the temperature at which that yield was obtained
• Devolatilization kinetic parameters for Arrhenius equations as derived from drop tube
experiments
•
13
Aromaticity as measured in carbon 13 nuclear magnetic resonance ( C NMR) experiments
• Concentrations of methoxy (-OCH3) functional groups measured during 13C NMR
experiments.
4-20
10581090
Biomass Energy
DTR experiments conducted at The Energy Institute of Pennsylvania State University [7] [18]
and reported by Tillman [8] [19] measured the maximum volatile yield of four biomass fuel
samples. Drop tube reactor temperatures varied from 752°F (400°C) to 3092°F (1700°C).
All DTR experiments were conducted in an argon atmosphere. In addition to the biomass fuels,
Pittsburgh Seam #8 coal and Black Thunder PRB coal were tested as representative fossil fuels.
Figure 4-4 shows the results of these tests.
1 0 0 .0 0 %
9 0 .0 0 %
8 0 .0 0 %
7 0 .0 0 %
6 0 .0 0 %
M a x im u m
5 0 .0 0 %
V o la tile Y ie ld (% )
4 0 .0 0 %
3 0 .0 0 %
2 0 .0 0 %
1 0 .0 0 %
0 .0 0 %
U rb a n w o o d F re s h W e a th e re d B la c k P itts b u rg h
Sawdust W a s te S w itc h g ra s s S w itc h g ra s s Thunder #8
F u e l T yp e
Figure 4-4
Maximum Volatile Yield of Biomass Fuels Compared to Reference Coals
(Sources: Johnson et al., 2001; Johnson et al., 2002)
All biomass fuels exhibit maximum volatile yields of about 90+%. This compares with the
lower maximum volatile yields for the two representative coals. It is also significant to note that
the fresh sawdust and fresh switchgrass achieved maximum volatile yields at 1832°F (1000°C)
while the urban wood waste and weathered switchgrass required temperatures of 2732°F
(1500°C) to achieve maximum volatile yield.
The DTR experiments were conducted to obtain devolatilization kinetics, using the following
equations:
R = (V/V∞)/tr Equation 4-1
where R is reactivity, V is percentage weight loss of the particle at a given temperature, V∞ is the
maximum percentage weight loss, and tr is residence time in the DTR. V is further defined by
Equation 4-2:
V = [(wf –wc)/wf] x 100 Equation 4-2
where wf is weight of fuel being fed and wc is weight of char produced by the DTR experiment.
4-21
10581090
Biomass Energy
The DTR kinetics are summarized in Table 4-13, and illustrated in Figures 4-5 through 4-11.
Note that the fresh sawdust exhibits a two-stage devolatilization mechanism with much of the
pyrolysis occurring below 1112°F (600°C). This is consistent with the evolution of extractives
and holocellulose (cellulose plus the hemicelluloses). Research by Shafizadeh and DeGroot [20]
[21] indicates that significant pyrolysis of woody cellulose occurs at less than 570°F (300°C).
Research by Broido [22] also confirms the low temperature at which woody cellulose pyrolysis
occurs, and more significantly confirms the multi-stage process of woody cellulose
devolatilization. Two-stage devolatilization also occurs for many coals.
Table 4-13
Devolatilization Kinetic Parameters for Biomass Fuels and Reference Coals (400-1700°C)
Pre-Exponential Constant Activation Energy

Fuel
A (1/sec) E (kcal/mol)
Fresh Sawdust (400-600°C) 1.17 0.681
Fresh Sawdust (600-1000°C) 5.74 3.42
Urban Wood Waste 3.53 2.41
Fresh Switchgrass 4.01 2.67
Weathered Switchgrass 1.91 1.35

Black Thunder PRB Coal 59.1 9.53
Pittsburgh Seam #8 Coal 66.2 10.3
10
E = 0.681 kcal/mol K
A = 1.17 1/sec
Reactivity, R (1/sec)
600 C 500 C
400 C
0.1
1 1.1 1.2 1.3 1.4 1.5 1.6
1/T X 1000, 1/K
Figure 4-5
Low-Temperature Devolatilization Kinetics for Fresh Sawdust
(Source: Johnson et al., 2001)
4-22
10581090
Biomass Energy
10
E = 3.42 kcal/mol K
Reactivity, R (1/sec) A = 5.74 1/sec
1000 C
800 C
1
700 C
600 C
0.1
0.6 0.7 0.8 0.9 1 1.1 1.2
1/T X 1000, 1/K
Figure 4-6
High-Temperature Devolatilization Kinetics for Fresh Sawdust
Urban Wood Waste
10
E = 2.41 kcal/mol
R = 1.986 kcal/kmol K
A = 3.53 1/sec
1300 C 1000 C
800 C
1200 C 600 C
1
0.1
0.40 0.60 0.80 1.00 1.20 1.40 1.60
1/T X 1000 (1/K)
Figure 4-7
Devolatilization Kinetics for Urban Wood Waste
4-23
10581090
Biomass Energy
Switchgrass 2002
10
E = 2.67 kcal/mol
R = 1.986 kcal/kmol K
A = 4.01 1/sec
1300 C
1000 C
800 C
1200 C
1
600 C 400 C
0.1
0.4 0.6 0.8 1.0 1.2 1.4 1.6
1/T x 1000 (1/K)
Figure 4-8
Devolatilization Kinetics for Fresh Switchgrass
10
E = 1.35 kcal/mol
A = 1.91 1/sec
1000 C 800 C
700 C
600 C
1
0.1
0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
1/T X 1000, 1/K
Figure 4-9
Devolatilization Kinetics for Weathered Switchgrass
4-24
10581090
Biomass Energy
10
1700 C E = 9.53 kcal/mol

Reactivity, R (1/sec) 1500 C A = 59.1 1/sec
1000 C
800 C
0.1
0.4 0.5 0.6 0.7 0.8 0.9 1
1/T X 1000, 1/K
Figure 4-10
Devolatilization Kinetics for Black Thunder PRB Coal
10
1700 C
E = 10.3 kcal/mol
A = 66.2 1/sec
1500 C
1000 C
1
0.1
0.4 0.5 0.6 0.7 0.8 0.9 1
1/T X 1000, 1/K
Figure 4-11
Devolatilization Kinetics for Pittsburgh Seam #8 Coal
4-25
10581090
Biomass Energy
13
C NMR experiments also highlight the fundamental differences in reactivity, or volatility,
between biomass fuels and coals. Aromaticity—calculated by Equation 4-3—is the first measure
of volatility.
Ar = (Caromatic/Ctotal) x 100 Equation 4-3
where Ar is aromaticity, expressed as a percentage, Caromatic is the number of aromatic carbon

atoms in the sample, and Ctotal is the total number of carbon atoms in the sample. Aromaticities
measured by the Energy Institute of Pennsylvania State University are as follows:
• Sawdust: 8%
• Weathered switchgrass: 10%
• Black Thunder PRB coal: 57%
• Pittsburgh #8 eastern bituminous coal: 70%
In all cases, the biomass contained an average of six aromatic carbons per cluster, resulting from
single aromatic ring structures. The Black Thunder coal averaged 10 aromatic carbons per
cluster, indicating an average of two fused aromatic rings per cluster. The Pittsburgh seam
eastern bituminous coal had an average of 15 aromatic carbons per cluster, indicating an average
of three to four fused aromatic rings per cluster.
Aromaticity is directly related to devolatilization kinetics, according to Equation 4-4:
Eact = 1.91 + 17.23(Ar) –2.00(AC/Cl) Equation 4-4
where Eact is activation energy in the Arrhenius equation in kcal/mol, Ar is aromaticity, and
AC/Cl is the measured number of aromatic rings/cluster. This equation was developed for a
broad range of fuels including biomass, lignite, subbituminous coals, eastern and Midwestern
bituminous coals, and petroleum coke. It has an r2 of 0.87 and a significance >0.999. It clearly
relates biomass structure to its reactivity, and places the biomass reactivity into the overall
context of solid fuels.
The final measure of reactivity is the presence of methoxy functional groups in a given solid
fuel. Some 8.7% of the carbon atoms in fresh sawdust are in the form of –OCH3 functionalities.
Switchgrass has 7.9% of its carbon atoms in methoxy functional groups. In contrast, Black
Thunder coal has 4.2% of its carbon atoms in the form of –OCH3 groups, and Pittsburgh seam
bituminous coal contains no such functionalities.
4.2.3.2 Fuel Nitrogen Characteristics of Biomass Fuels
The overall reactivity of biomass fuels is clearly established. The reactivity of nitrogen from the
various biomass fuels significantly impacts its ability to influence NOX formation—or NOX
control. Key measures include fuel nitrogen in lb/106 Btu as well as the following:
• Maximum fuel nitrogen volatility, paralleling the maximum fuel volatility
• Distribution of volatile fuel nitrogen and char-bound fuel nitrogen
• Pattern of volatile fuel nitrogen evolution.
4-26
10581090
Biomass Energy
These parameters are important in that the volatile fuel nitrogen is more readily reduced to N2
through staged fuel and staged air mechanisms. However, it is important to recognize that not all
nitrogen volatiles are identical, and they evolve at different time histories. Nitrogen volatiles that
evolve more rapidly are more likely to be released from the fuel particle in a fuel-rich region of
the furnace; nitrogen volatiles that are slow to evolve are more likely to be released in a fuel-lean
environment, where surplus oxygen exists.
Figure 4-12 shows fuel nitrogen concentrations for four biomass fuels extensively analyzed,
along with reference coals. Note the high concentration of fuel nitrogen in the urban wood waste.
Total fuel nitrogen concentration in the sample analyzed is consistent with that measured in the
urban wood waste cofired at the Bailly Generating Station [11]. It results largely from glues in
the plywood and particleboard, OSB, and MDF components of the urban wood waste. It also
results from coatings and coverings, including paints and laminates. The nitrogen contents in the
fresh and weathered switchgrass samples vary significantly, with the weathered switchgrass
showing less fuel nitrogen.
1.8
1.6
1.4
1.2
1
Fuel Nitrogen,
Lb/MMBtu 0.8
0.6
0.4
0.2
0
Urban wood Fresh Weathered Black Pittsburgh
Sawdust Waste Switchgrass Switchgrass Thunder #8
Fuel Type
Figure 4-12
Fuel Nitrogen Concentrations for the Fuel Samples Analyzed in Detail
(Source: Johnson et al., 2001; Johnson et al., 2002)
Figure 4-13 shows the distribution of nitrogen in volatile or char form for these fuels. Note that
the sawdust contains virtually no char nitrogen. Nitrogen in sawdust has been shown to comprise
largely amine functional groups in the living portion of the wood. The char nitrogen is virtually
identical for the two switchgrass samples and for the urban wood waste. Clearly some of the
nitrogen in the glues does not evolve in volatile form, but remains in the char form. Note also
that the char nitrogen for biomass is significantly lower than that for the reference fuels. The
Black Thunder PRB coal shows increased char nitrogen, and the Pittsburgh Seam coal shows the
4-27
10581090
Biomass Energy
highest concentration of fuel nitrogen in char form. These data for coals are well correlated with
the degree of difficulty in controlling NOX emissions during pulverized coal combustion.
1.8
Volatile Fuel Nitrogen (light

1.6
green)
1.4 Char Fuel Nitrogen (dark green)
1.2
Distribution of 1
Fuel Nitrogen by
Type (lb/MMBtu) 0.8
0.6
0.4
0.2
0
Fuel Type
Figure 4-13
Distribution of Volatile and Char Fuel Nitrogen in Selected Biomass and Reference Coal
Samples (Source: Johnson et al., 2001; Johnson et al., 2002)
Maximum volatile nitrogen yield significantly impacts NOX management. It ties structural
considerations to fuel behavior in combustion settings. Figure 4-14 shows maximum nitrogen
volatile yield as a function of fuel sample. Note that sawdust shows the highest maximum
nitrogen volatile yield, while Pittsburgh Seam coal shows the lowest nitrogen volatile yield.
Weathered switchgrass shows the lowest nitrogen volatile yield among the biomass fuels. These
data are consistent with those previously presented in terms of overall fuel volatility and
distribution of fuel nitrogen.
Given these data, it is critical to evaluate the behavior of nitrogen volatiles in biomass and
compare that behavior to the nitrogen volatiles of reference coals. Baxter et al. [23] [24] initiated
investigations into the rate of nitrogen volatile release from coal particles during the pyrolysis or
devolatilization phases of combustion. These investigations recognize that fuel nitrogen in all
coal can be in pyrrole, substituted pyrrole, and pyridine forms, and can be in amine and aniline
forms in the lower rank coals (e.g., lignites) to a very limited extent. These various forms of
nitrogen in fuel volatilize at different temperatures and at different time histories, as the particle
is heated and devolatilized. Volatiles that evolve most rapidly are also most easily controlled.
They tend to evolve in fuel-rich or oxygen-deficient regions of the combustion system. Nitrogen
volatiles retained for longer times in the fuel particle are more likely to be released in fuel-lean
regions of the combustion zone, where excess oxygen is available to oxidize the volatile
nitrogen.
4-28
10581090
Biomass Energy
100.00%
90.00%
80.00%
70.00%
60.00%
Maximum
Nitrogen Volatile 50.00%
Yield (%)
40.00%
30.00%
20.00%
10.00%
0.00%
Fuel Type
Figure 4-14
Maximum Nitrogen Volatile Yields for Selected Biomass and Coal Samples
(Source: Johnson et al., 2001; Johnson et al., 2002)
Drop-tube reactor experiments by Johnson et al. [7] [18] document volatile nitrogen evolution
patterns for four biomass fuel samples evaluated: sawdust, urban wood waste, fresh switchgrass,
and weathered switchgrass. Nitrogen volatilization patterns are shown as a function of
temperature, recognizing that the temperature history of the particle is a reasonable substitute for
the time history of the fuel particle during the devolatilization phase. In these experiments, the
volatile nitrogen evolution rate was compared to the volatile carbon evolution rate and to the
total volatile matter evolution rate. These volatilization patterns were developed over the entire
temperature range of 400°C to 1700°C.
Figures 4-15 through 4-18 present the volatile nitrogen and carbon evolution rates for
representative biomass fuels. Among these samples, nitrogen volatiles evolve most rapidly for
sawdust, where nitrogen volatiles evolved more rapidly than the carbon volatile matter,
particularly during the initial stages of pyrolysis. For urban wood waste, the nitrogen evolution
rate essentially parallels the carbon volatile evolution rate. The nitrogen volatile evolution rate
lags just slightly behind the volatile carbon evolution rate for fresh switchgrass. This is quite
similar to the result for urban wood waste. The nitrogen volatile evolution lags significantly
behind the carbon volatile evolution for the weathered switchgrass, indicating the potential for
more difficulty in using this fuel to reduce NOX emissions from biomass combustion.
4-29
10581090
Biomass Energy
100.00
Percent Carbon or Nitrogen in Volatile Matter

90.00
Nitrogen
80.00
70.00
60.00
Carbon
50.00
40.00
30.00
20.00
10.00
0.00
0 200 400 600 800 1000 1200 1400 1600 1800
Temperature (C)
Figure 4-15
Volatile Nitrogen and Carbon Evolution from Fresh Sawdust
100.00
90.00
Volatile Yield of Carbon and Nitrogen
Carbon
80.00
70.00
60.00
Nitrogen
50.00
40.00
30.00
20.00
10.00
0.00
0 200 400 600 800 1000 1200 1400 1600 1800
Temperature (C)
Figure 4-16
Volatile Nitrogen and Carbon Evolution from Urban Wood Waste
4-30
10581090
Biomass Energy
100.00
90.00
Percent Carbon or Nitrogen Volatilized
80.00
Carbon
70.00
60.00
50.00
Nitrogen
40.00
30.00
20.00
10.00
0.00
0 200 400 600 800 1000 1200 1400 1600 1800
Temperature (C)
Figure 4-17
Volatile Nitrogen and Carbon Evolution from Fresh Switchgrass
100.00%
90.00%
Volatile Yield of Carbon or Nitrogen
80.00%
Carbon
70.00%
60.00%
50.00%
40.00%
30.00%
Nitrogen
20.00%
10.00%
0.00%
0 200 400 600 800 1000 1200 1400 1600 1800
Temperature (C)
Figure 4-18
Volatile Nitrogen and Carbon Evolution from Weathered Switchgrass
4-31
10581090
Biomass Energy
The data presented above can also be normalized to total volatile evolution, as is shown in
Figures 4-19 and 4-20 for the sawdust and weathered switchgrass. Normalizing the data to
proximate analysis, consistent with the Baxter analysis [23] [24], provides a means for relating
volatile nitrogen evolution to traditional fuel analyses. Further, it provides a means for relating
the nitrogen evolution data to typical information used in burner design and operation.
100.00
Percent Nitrogen or Carbon
90.00
Evolved as Volatiles
80.00
Nitrogen Volatiiles Formed
70.00
60.00
50.00
40.00
Carbon Volatiles Formed
30.00
20.00
10.00
0.00
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0%
Percent Volatile Matter Evolved
Figure 4-19
Nitrogen and Carbon Volatilization Normalized to Total Volatiles Formed from Sawdust
100.00%
90.00%
Percent Evolving as Volatile Matter
80.00%
70.00%
60.00%
50.00%
40.00%
Carbon Volatile Yield
30.00%
20.00%
10.00%
Nitrogen Volatile Yield
0.00%
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0%
Percent Total Volatile Yield From Fuel
Figure 4-20
Nitrogen and Carbon Volatile Evolution from Weathered Switchgrass Normalized to Total
Volatile Evolution
4-32
10581090
Biomass Energy
The normalized data demonstrate that the carbon volatile evolution and the total volatile
evolution are virtually identical, with only a slight lag in the carbon volatile evolution caused by
the hydrogen in the fuel.
An alternative approach to evaluating these data is to calculate the nitrogen/carbon (N/C) atomic
ratios of the char remaining from pyrolysis or devolatilization at given temperatures. If the N/C
ratio is declining, then the nitrogen volatiles are evolving more rapidly than the carbon volatiles.
Alternatively, if the N/C ratio is increasing, then the nitrogen is preferentially being retained in
the char. These data can be presented as calculated N/C ratios. These data can also be presented
on a normalized basis, where the initial N/C ratio of the fuel is considered 1.0. N/C ratios in the
char can then be related to the N/C ratio in the fuel on a proportional basis. Both approaches are
presented in Figures 4-21 and 4-22. In Figure 4-21, where normalized data are presented, Black
Thunder and Pittsburgh #8 coals are also presented for comparative purposes. The sawdust
shows the only initial decline in N/C ratios, indicating the preferential evolution of nitrogen
volatiles. The urban wood waste and fresh switchgrass data show that the nitrogen volatiles
evolve at a rate just slightly slower than the carbon volatiles. The weathered switchgrass data
show that the nitrogen volatile evolution lags significantly behind carbon volatile evolution at the
early stages of combustion. The coal data show that the more reactive Black Thunder coal
nitrogen volatile evolution pattern essentially mirrors the carbon volatile evolution pattern, while
there is a significant lag in the nitrogen volatile evolution relative to the carbon volatile evolution
for the Pittsburgh #8 coal.
0.035
Nitrogen/Carbon Atomic Ratio in Solid Fuel or
Urban Wood Waste

0.030
0.025
0.020
Fresh Switchgrass
Char
0.015
Weathered Switchgrass
0.010
Sawdust
0.005
0.000
0 200 400 600 800 1000 1200 1400 1600 1800
Temperature (C)
Figure 4-21
Nitrogen/Carbon Atomic Ratios for Biomass Chars Formed During DTR Pyrolysis at
Various Temperatures
4-33
10581090
Biomass Energy
2.5
Normalized N/C Atomic Ratio in Solid Fuel and

Weathered Switchgrass
Fresh Switchgrass
1.5
Pittsburgh #8
Char
Black Thunder
0.5
Fresh Sawdust Urban Wood Waste

0
0 200 400 600 800 1000 1200 1400 1600 1800
Temperature (C)
Figure 4-22
Normalized Nitrogen/Carbon Atomic Ratios for Biomass Solid Fuel and Char, Compared to
Pittsburgh #8 Bituminous Coal and Black Thunder Powder River Basin Subbituminous
Coal
The result of the nitrogen evolution analysis is an assessment of the relative desirability of the
four biomass fuels tested. Sawdust clearly has the lowest potential for generating NOX emissions
and weathered switchgrass has the most potential. Fresh switchgrass and urban wood waste are
about on equal footing with respect to nitrogen evolution; however, fresh switchgrass has the
advantage of a lower fuel nitrogen content. Actual impact on NOX emissions will vary
significantly, though, depending on the fraction of biomass burned, pre-existing combustion
conditions, etc.
4.2.3.3 Ash Reactivity, Slagging, and Fouling
Ash reactivity, slagging, and fouling also provide significant concerns for biomass fuel
utilization. Ash elemental analyses have been previously presented. More significant data can be
obtained from chemical fractionation of the ash, however. In this procedure (see Miller et al. [25]
[26] and Miller and Miller [27]), the fuel is successively leached in water, ammonium acetate,
and hydrochloric acid (HCl). The successive leaching provides insights into the reactivity of a
given inorganic constituent by revealing mineral associations of the cations. Matter leached with
water and ammonium acetate is highly reactive; matter leached in HCl is less reactive, while the
residue from this procedure is considered to be relatively unreactive. Baxter et al. [24] and
Bryers [28] have used this procedure to evaluate a wide array of biomass fuels, and a very
similar technique is used to assess the deposition probabilities of coals and other mineral effects.
4-34
10581090
Biomass Energy
Table 4-14 presents chemical fractionation of an array of biomass fuels developed by Miller et
al. [25]. This includes wood, herbaceous material (Reed Canary grass), and manure samples.
Table 4-14
Chemical Fractionation Analyses of Various Biofuel Ashes
Parameter Fuel
Reed
Pine Sheep Dairy Tie-
Canary Poultry Litter
Shavings Manure Stall Manure
Grass
Potassium
Water soluble and 65 97 97 96 41
ion exchangeable
Acid soluble 0 2 1 1
Insoluble 35 1 2 3 20
Sodium
ion exchangeable
Acid soluble 0 0 2 0 0
Calcium
ion exchangeable
Insoluble 2 1 1 1 1
Magnesium
ion exchangeable
Aluminum
ion exchangeable
Insoluble 70 72 68 75 100
Silicon
ion exchangeable
Insoluble 71 45 100 100 100
Source: Miller et al., 2002b.
4-35
10581090
Biomass Energy
Table 4-15, from Baxter et al. [24], presents chemical fractionation of switchgrass.
Table 4-15
Chemical Fractionation of Potassium in Switchgrass for Four Samples
Parameter Switchgrass Sample
A B C1 C2
Water Soluble 52 68 70 72
Ion Exchangeable 35 24 20 16
Acid Soluble 9 0 0 2
Residual 4 8 10 10
Source: Baxter et al., 1996b.
Baxter, as reported by Miles et al. [10], performed one complete analysis of switchgrass as
shown in Table 4-16. These results show the high reactivity of the alkali, and alkali earth
minerals along with the low reactivity of the silica, alumina, and titanium dioxide.
Table 4-16
Complete Chemical Fractionation of a Switchgrass Sample
Ash Component Chemical Fraction (wt %)
Water Soluble Ion Exchangeable Acid Soluble Residual

SiO2 0 0 0 100
Al2O3 0 0 0 100
TiO2 0 0 0 100
Fe2O3 0 0 40 60
CaO 2 80 10 8
MgO 32 48 7 13
Na2O 52 0 0 48
K2O 70 20 0 10
P2O5 70 8 15 7
SO3 26 69 5 0
Source: Miles et al., 1993.
Chemical fractionation provides key insights into the reactivity of ash and its potential to cause
slagging and fouling. It provides insights into potential combustion problems of various biomass
fuels when used in conjunction with total ash concentration. Reactivity places a premium on
low-ash biomass fuels and demonstrates some of the difficulties with the herbaceous materials. It
also demonstrates that trees grown for fuel with high doses of fertilizer containing potassium
could generate significant slagging and fouling deposits.
4-36
10581090
Biomass Energy
4.2.4 Wastewater Treatment Gas and Landfill Gas
Unlike solid biomass fuels, gaseous fuels generated by anaerobic digestion either in a wastewater
treatment reactor or in an enclosed landfill cell, exhibit considerable utility and energy density.
All of these gases are produced at atmospheric pressure, and therefore must be compressed
before use in internal combustion engines or combustion turbines. Further, all of these gases are
saturated with water and should be dehydrated to some extent before use.
4.2.4.1 Gas Compositions
Wastewater treatment gas, resulting from the digestion of human waste matter, has a typical
composition of greater than 50% methane on a dry basis, and contains some ammonia, nitrogen,
and oxygen depending upon the integrity of the system.
Table 4-17 presents the composition of wastewater treatment gas produced at the T.E. Maxson
Wastewater Treatment Plant in Memphis, Tennessee. This gas has been fired in the Allen Fossil
Plant of TVA. Note that ammonia was not detected in the analysis. Ammonia concentrations can
be significant when animal wastes are digested.
Table 4-17
Composition and Characteristics of the Wastewater Treatment Gas Produced
at the T.E. Maxson Treatment Facility in Memphis, Tennessee
Parameter Value
Gaseous Composition (volume percent, dry basis)
CO2 31.4
O2 0.4
N2 4.1
CH4 64.1
Molecular Weight 25.32
Higher Heating Value (Btu/ft3) 647.7
3
Lower Heating Value (Btu/ft ) 584.6
Ultimate Analysis
Carbon 45.21
Hydrogen 10.17
Oxygen 40.17
Nitrogen 4.45
Higher Heating Value (Btu/lb) 10,002
Landfill gas typically has a lower calorific value, with CO2/CH4 ratios in the 48/50 to 52/48
range. Heating values are accordingly lower than those shown in Table 4-17. Landfill gas
composition is a function of the age of the landfill, the moisture migration through the cells, and
the composition of the material placed in the cells.
4-37
10581090
Biomass Energy
4.2.4.2 Utilization
Because these biomass gases are low in pressure and have modest heating value, they are not
readily transported significant distances. They can be cleaned of particulate, moisture, and other
contaminants, however, and under such circumstances can be used in numerous generation
technologies. Their reactivity—or rate of combustion—is governed by the kinetics of methane
oxidation. The many combustion issues associated with using solid biomass fuels do not arise for
these fuels.
Use of wastewater treatment gas and landfill gas requires obtaining sufficient quantities in a
single location to make the process economical.
4.3 Characteristics of Biomass Cofiring Technologies
Currently the technologies that are commercially available, or commercially offered, include
various forms of cofiring along with stand-alone (100% biomass-fired) Rankine-cycle generating
systems. Cofiring refers to the practice of firing biomass fuels as a supplement to coal (or very
rarely, gas). Stand-alone biomass plants may either supply superheated steam to condensing
turbines where the only product is electricity, or they may supply steam to backpressure or
automatic extraction turbines in cogeneration (also known as combined-heat-and-power or CHP)
applications.
Cofiring systems are most readily adapted to electric generating stations. Biomass can be
integrated with the fuel supply to existing boilers designed to utility standards. The biomass can
be used in large reheat boilers where steam is used most efficiently. Such systems are less
expensive than stand-alone power plants on a $/kW supported by biomass basis. If the biomass is
unavailable temporarily, the operation of the unit is not compromised. However, integrating
biomass fuel into the coal stream still involves complex issues of materials handling and control.
Further, cofiring does not contribute additional capacity; instead, it displaces coal fuel at the unit.
When gaseous biomass fuels (e.g., landfill gas, wastewater treatment gas, producer gas) are fired,
additional considerations and benefits are available. Figure 4-23 represents a typical
configuration of biomass cofiring in a coal-fired utility boiler.
4.3.1 Overview of Solid Biomass Fuel Cofiring Systems
Biomass cofiring refers to a family of technologies designed to introduce these renewable fuels
into fossil-fuel-fired systems. The common approach is to fire solid or gaseous biomass fuels in
coal-fired boilers. Differences in applicable technology are a function of the following issues:
• The biomass fuel being fired (woody biomass, herbaceous biomass, gaseous fuel from a
landfill, a wastewater treatment facility, or an atmospheric gasifier)
• The boiler type (cyclone, tangentially fired pulverized coal, wall-fired pulverized coal,
fluidized bed)
• The percentage cofiring sought
4-38
10581090
Biomass Energy
• The design and operating parameters of the unit (e.g., the presence of spare pulverizer
capacity, the use of low-NOX burners).
There are certain combinations that are highly useful, but others simply do not work. Technology
selection is based upon putting the most appropriate combination in place.
Mechanical
draft cooling
tower
Cooling
water
Turbine/ Electricity
generator
Steam (2400 psi, 1000 deg F,

Moisture losses reheat to 1000 deg F)
Biomass
710 F
Alternate 269 F
Fuel Utility boiler Air heater ESP
Receiving
610 F 80 F
Combustion Stack
Coal Bottom ash air Fly ash
Ferrous
Figure 4-23
Schematic of Biomass Cofiring with Coal in a Utility Boiler
Numerous facilities have demonstrated cofiring (Table 4-18). Of these, cofiring has continued at
the Greenidge Station, the Big Stone Plant, Plant Gadsden, and the Willow Island Generating
Station at a minimum. It was previously practiced for many years at the King Generating Station
of Northern States Power and several other locations.
4.3.2 Cofiring in Cyclone Boilers
Cyclone firing demonstrates that the key issue of cofiring in either cyclone boilers or pulverized
coal (PC) boilers is fuel handling. Combustion considerations dictate many of the environmental
and technical consequences of cofiring. Materials handling concerns have the major impact on
economics.
Cyclone boilers are commonly considered most appropriate for cofiring dissimilar solid fuels—
biomass fuels, tire-derived fuel (TDF), petroleum coke, and other opportunity fuels. In cyclone
firing, fuel crushed typically to ⅜-inch x 0-inch (9.5 mm x 0 mm) is stored in bunkers. The
outfeed from the bunkers falls by gravity through the downcomers to the feeders, typically Stock
gravimetric feeders, and is then fed by gravity to the cyclone burners. Rather than being burned
in suspension, the fuel is burned in a cyclone, forming a slag layer on the cyclone barrel.
4-39
10581090
Biomass Energy
Table 4-18
Representative Biomass Cofiring Tests and Demonstrations
Owner at Time Biomass Responsible

Plant Boiler Type
of Demo Fired Organization
Allen Fossil Plant TVA Cyclone Sawdust TVA/EPRI/DOE-
NETL
Kingston Fossil Plant TVA T-fired PC Sawdust TVA
Colbert Fossil Plant TVA Wall-fired PC Sawdust TVA
Plant Hammond Southern Co Wall-fired PC Sawdust Southern Co.
Plant Kraft Southern Co T-fired PC Sawdust Southern Co.
Plant Gadsden Alabama Power T-fired PC Switchgrass, Southern Co.,
sawdust DOE, EPRI
Shawville Generating GPU Genco Wall-fired PC Sawdust GPU/EPRI/DOE-
Station T-fired PC NETL(2)
Seward Generating GPU Genco Wall-fired PC Sawdust GPU/EPRI/DOE-
Station NETL(2)
Greenidge Station NYSEG/AES(1) T-fired PC General NYSEG/
wood waste NYSERDA(3)
Blount St. Station MG&E Wall-fired PC Switchgrass DOE/EPRI
Michigan City NIPSCO Cyclone Urban wood EPRI/DOE-NETL(2)
Generating Station waste
Bailly Generating Station NIPSCO Cyclone Urban wood EPRI/DOE-NETL(2)
waste
Albright Generating Allegheny Energy T-fired PC Sawdust Allegheny
Station Energy/DOE-
NETL(2)/EPRI
Willow Island Generating Allegheny Energy Cyclone Sawdust Allegheny
Station Energy/DOE-
NETL(2)/EPRI
Dunkirk Generating Niagara Mohawk T-fired PC Sawdust/ DOE-GFO(4)
Station SRWC
Ottumwa Generating Alliant Energy T-fired PC Switchgrass DOE-GFO(4)
Station
Polk County Generating TECO Coal gasifier SRWC DOE-NETL(2)
Station IGCC
Plant Gadsden Southern Co T-fired PC Switchgrass DOE-GFO(4)
Studstrup(5) Midkraft Wall-fired PC Straw Elsamprojekt
Lahti (gasification) Lahden Wall-fired PC General FWOy/EU Thermie
Lampovoima Oy wood waste Program
1. Many of these units have changed hands since the demonstration (e.g., AES has purchased the Greenidge Station formerly
owned by NYSEG).
2. NETL is the National Energy Technology Laboratory of DOE.
3. NYSERDA is the New York State Energy Research and Development Agency.
4. GFO is the Golden Field Office of DOE.
5. One example of many European cofiring demonstrations.
The flow of fuel to the bunkers, Stock feeders, and cyclones is appropriate for granular and
regularly shaped particles. Wood waste such as sawdust at 18 lb/ft3 works well in cyclone
boilers. However, cyclone technology is not suitable for switchgrass and other herbaceous
4-40
10581090
Biomass Energy
materials. as reported by Bush et al. [29], Jenike and Johanson of Westford, Massachusetts state
that, in evaluating co-feeding and co-milling switchgrass and coal at Plant Gadsden for Southern
Co., they analyzed blends of switchgrass and Pratt Seam coal. Their research documented that a
blend of 5% switchgrass and 95% coal on a mass basis would not flow from the bunker. The
needle-like characteristics of the switchgrass particles, combined with the low bulk densities of
switchgrass (e.g., 5 to 7 lb/ft3), resulted in continuous bridging. As a consequence of the handling
characteristics of herbaceous materials, all cyclone demonstrations have been on woody biomass.
Cyclone firing is unique in that the coal is not pulverized, combustion occurs at very high peak
temperatures (e.g., 3300°F to 3550°F), and inorganic matter is largely converted into a liquid
slag, which is then tapped from the furnace bottom. Cyclone firing typically results in about 70%
of the mineral matter in the coal or fuel mass being removed as bottom ash or slag, leaving only
30% of the mineral matter to exit the boiler as fly ash. Slag or bottom ash is typically sold for
sandblasting grit, roofing granules, and similar products.
Cyclone firing can generate high concentrations of NOX. NOX emissions of 1.2 to 2.0 lb/106 Btu
have been generated depending upon boiler size and configuration (front wall vs. opposed wall
configurations of the cyclone barrels). More recently, most cyclone boilers have been equipped
with overfire air systems to reduce NOX emissions to the 0.3 to 0.5 lb/106 Btu range. No new
cyclone boilers have been built since about 1975, due in large part to the NOX emissions
associated with these units.
Cyclones, originally designed for slagging coals of the Midwest, have had success in firing a
range of coals. Consequently, cyclone boilers were adapted to cofire petroleum coke [19], tire-
derived fuel (TDF) [30], and other opportunity fuels with coal. In the 1970s, some experiments
cofired refuse-derived fuel (RDF) and coal in cyclone boilers. In the 1990s, Northern Indiana
Public Service Co. (NIPSCO) cofired tar-laden soils from manufactured gas plants with coal in
one cyclone boiler. In the early 1990s, Northern States Power also began cofiring dry wood
waste from Andersen Windows with PRB coal at its 600-MWe King Generating Station.
4.3.2.1 Cofiring System Design for Cyclone Boilers
With the exception of the King Station, where dry sawdust was blown into the secondary air
system of three of 12 cyclone barrels, all demonstrations have blended woody biomass with coal
and/or other opportunity fuels on the coal pile or on the conveyor leading to the bunker system.
Consequently, the cofiring system has to have sufficient transport capacity to load the entire
requirement for biomass within a two-to four-hour window depending upon loading schedules at
the plant.
Critical variables associated with cyclone cofiring systems include:

• Project design life
• Degree of automation desired
• Space available, including both location and acreage
• Soil conditions where the cofiring system is intended to go
• Plant standards and preferences.
4-41
10581090
Biomass Energy
Project design life can be between three and 10 years, since the cofiring system addresses
existing plants built before 1975. This impacts the robustness of the design. Degree of
automation is critical. If the plant accepts a low degree of automation, the capital cost can be
minimized; however, labor requirements increase.
As a minimum, cofiring requires a pile and a reclaim conveyor connected to the main belt
conveyor feeding the bunkers or silos. This requires a truck unload location and a conveyor with
a belt scale for control. One to two workers are necessary to operate the system. It also requires
accepting the risk of tramp material in the biomass. Absent accepting this risk, a magnet must be
installed somewhere on the conveyor. This system can be enhanced by a pole barn or similar
structure to contain biomass dust. At the other extreme, the system can be built to accept any
type of truck and automatically screen biomass for oversized particles, with a grinder for
oversized particles, a stock-out bin, a reclaim system, and controls to manage the blend
automatically. This may require one-quarter to one-half of a worker. Space available dictates
whether a hydraulic truck unloader can be successfully installed. It also dictates conveyor
lengths, conveyor selection, and ease of operation. Soil conditions can be critical in determining
whether spread footings are acceptable or whether pile construction is required. Plant standards
dictate whether or not the electrical installation must be explosion-proof, the extent to which fire
suppression systems are required, the method and degree of system controls, and the extent to
which redundancy is built into the system. All of these are issues with significant economic
consequences.
An example of a cyclone cofiring system is the Willow Island Generating Station boiler #2
demonstration hosted by Allegheny Energy Supply Co., LLC. The Willow Island Generating
Station boiler #2, located in Willow Island, West Virginia, is a 188-MWe (net) unit equipped
with five cyclone barrels, all located in two rows or elevations on the front wall of the boiler. The
boiler operates in a positive pressure mode. The cyclones are equipped with scroll feeders.
Willow Island #2 boiler burns a washed Pittsburgh Seam coal. Limestone is periodically added
to the coal to enhance the slag-forming properties of the fuel. The plant is equipped with a “hot
side” electrostatic precipitator (ESP) for particulate control. Overfire air (OFA) was recently
added to the #2 boiler to reduce NOX emissions.
The design criteria for the Willow Island cofiring system included the following:
• The capability to blend 15% sawdust with coal (mass basis) on the main coal belt,
recognizing that normal operation would involve blending 10% sawdust with coal on the
main belt.
• The capability to receive and process at least 150 to 200 tons (six to eight truckloads) of
sawdust daily in walking floor vans.
• The capacity to store more than three days of sawdust on-site in order to supply the boiler
with biomass during weekends, including holiday weekends.
• The construction of a facility with a design life of more than three years.
• The capability to screen the entire mass of biomass fuel in order to protect the crushers and
cyclone barrels from undesirable material.
4-42
10581090
Biomass Energy
• The capability to grind oversized particles to a fuel specification of less than ¼-inch particles,
consistent with maximizing the volatile matter release in the cyclone barrels.
• The capability to meter the flow of biomass to the cyclone barrels in order to measure the
blends being put up in the bunkers.
• In addition, the design was robust in order to minimize maintenance commitments.
Figures 4-24 and 4-25 show the system design. In this system, walking floor vans discharge their
load to a live-bottom hopper. The bottom of the hopper is a conveyor belt carrying the sawdust to
a 50-ton/hr disc screen capable of screening ¼-inch x 0-inch sawdust particles. Oversized
material is discharged from the screen to a slow speed/high torque grinder that reduces the wood
particles to the desired size. Acceptable particles are then carried by mechanical conveyor to a
408-ton live bottom storage bin. This bin accepts the material through three openings. It has the
capability to level the storage pile of sawdust using three leveling screws at the top of the bin.
The floor of the bin is a “walking floor” capable of moving the entire mass of sawdust in the bin.
Figure 4-24
Plan View of the Willow Island Biomass Cofiring System
4-43
10581090
Biomass Energy
Figure 4-25
Elevation View of the Willow Island Biomass Cofiring System
Sawdust is discharged from the bin by twin counter-rotating augers that convey the material to a
weigh belt conveyor. The augers operate at variable speed and meter the sawdust to the weigh
belt conveyor. The weigh belt conveyor carries sawdust to the main coal conveyor from the
crusher house to the power plant bunkers. The system is designed to convey up to 75 ton/hr of
sawdust to the main coal conveyor. Sawdust is added to the conveyor belt before the automatic
belt samplers. TDF is added to the conveyor belt separately, after the belt samplers.
The design was based upon walking floor van truck transportation. Value engineering dictated
that walking floor vans are used due to the costs of hydraulic chip van dumpers. Further, dust
management issues made use of trailer dump trucks less than desirable. Finally, there was
insufficient space at the Willow Island site to house a truck dump other than a walking floor
receiving hopper.
The design then focused upon screen selection. The disc screen provided an acceptable particle
size for combustion without generating the levels of dust associated with trommel screens.
Further, disc screens experience little blinding. Covered mechanical conveyors, selected both for
maintenance and dust management reasons, were used to load the bunker, rather than pneumatic
conveyors previously used to feed silos.
The walking floor bin was chosen over a silo due to storage requirements and height
considerations, in addition to consequent maintenance concerns when dealing with the head
pulley of the conveyor feeding the bin. There were ample reasons for using either type of storage
system, and customer preference ultimately influenced the choice of the storage facility. Covered
contained storage of the prepared fuel is always employed at power plants due to dust
management considerations.
Some additional features were incorporated into the design. A small permanent vacuum system
was added to facilitate general site clean-up. Dock seals were added around the truck unloading
hopper to minimize sawdust spillage at that location. An inspection door was added to the twin
4-44
10581090
Biomass Energy
auger conveyor moving sawdust from the walking floor bin to the weigh belt metering conveyor.
These features improve the overall performance of the system.
The system was installed on a site requiring extensive piling, which added considerably to the
cost. The system also met plant requirements for an explosion-proof electrical system. Figure
4-26 shows this system.
Figure 4-26
Receiving Sawdust at the Willow Island Generating Station
4.3.2.2 Capital Costs of Cofiring Systems for Cyclone Boilers
The capital costs associated with the Willow Island Generating Station are shown in Table 4-19.
Since this unit was built with the potential to supply 75 ton/hr to the main coal belt—equal to
15% of the coal feed on a mass basis, or 7% of the coal feed on a thermal input basis—it had the
capacity to support 13 MWe (net). Normal operation was between 5% and 10% of the total coal
input on a mass basis, or 2.2% to 4.5% on a thermal input basis. The resulting capital cost of the
Willow Island system was approximately $180/kW based on the kilowatt cofiring capacity
provided by the biomass fuel. Assuming a normal operating capacity of 50 ton/hr feed to the coal
bunkers, or 10% of the total coal flow on a mass basis, then the capital cost of this system is
$275/kW supported by biomass.
The capital cost of the Willow Island cofiring system can be compared with that required to
construct the Bailly Generating Station cofiring system, as shown in Table 4-20. The Bailly
Generating Station cofiring system consists of a pole barn, a trommel screen, a pad and
aboveground reclaimer, and a metering conveyor belt. The cost was $1.18 million. It had the
capacity to supply 10% of the mass feed, or 5% of the thermal input for the Bailly Generating
Station #7 boiler, a 165-MWe cyclone unit. On that basis, the capital cost of the Bailly cofiring
system was $140/kW supported by biomass.
4-45
10581090
Biomass Energy
Table 4-19
Capital Cost of the Willow Island Cofiring Demonstration
Number Description Capital Cost

1 Process Equipment $920,200
Walking floor bin
Outfeed conveyor from bin
Cross feed weigh belt conveyor
Scale
Sawdust receiving hopper
Conveyor to disc screen
Disc screen
Overs grinder
Outfeed conveyor
Vacuum system
Motor control center
2 Engineering Studies $194,400
Soils Studies
Engineering
3 Construction $1,114,400
4 Home Office Support and Other $86,800
TOTAL $2,315,800
Source: Tillman and Payette, 2004.
Table 4-20
Capital Cost Summary for the Bailly Generating Station Cofiring System

Trommel Screen
Above Ground Reclaim
Air Slide Metering Conveyor
2 Engineering and Installation $327,400
TOTAL $1,181,600
Source: Tillman, 1999.
The cost range, $140 to $180/kW supported by biomass, is a reasonable estimate of the capital
costs associated with cofiring in cyclone boilers, particularly at the 8- to 12-MWe range. Because
these systems are largely materials handling systems, the exponential scale factor for estimating
larger systems would be 0.65 to 0.8. The $1.18 million cost of the Bailly system could well
represent a minimum cost for cofiring at smaller scales.
4-46
10581090
Biomass Energy
The differential operating costs associated with automation is due to additional labor
requirements. Annual maintenance costs can be estimated at about 4% to 5% of the initial capital
investment.
4.3.2.3 Cofiring Impacts in Cyclone Boilers
Cofiring in cyclone boilers has been tested and demonstrated in a significant number of plants
including the Allen Fossil Plant of TVA, the Bailly Generating Station of NiSource, the
Michigan City Generating Station of NiSource, and the Willow Island Generating Station of
Allegheny Energy.
Efficiency impacts have been modest. At the Willow Island cofiring demonstration, 10% cofiring
(mass basis) resulted in an increase in the net station heat rate (NSHR) of 30 Btu/kWh.
Efficiency penalties of up to 0.5% have been measured when cofiring at 10% mass basis (5%
thermal basis) at the Allen Fossil Plant, the Bailly Generating Station, and the Michigan City
Generating Station. This is about equal to 50 Btu/kWh [11].
Operationally, cofiring increased the duty on the coal feeders to the cyclones, as would be
expected. In no case did cofiring limit unit capacity. Flame temperatures were largely unaffected.
Further, flame intensity of sawdust and finely divided biomass equals that of coal. Furnace exit
gas temperatures (FEGT) decreased in all cofiring tests at Willow Island and at the other cyclone
demonstrations. Figure 4-27 depicts this phenomenon at Willow Island.
3000
FEGT when firing coal alone
2500
Furnace Exit Gas Temperature (F)
2000
1500
FEGT when cofiring coal and sawdust

1000
500
0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0
Percent Sawdust in Fuel (mass basis)
Figure 4-27
Influence of Cofiring in Cyclone Boilers on Furnace Exit Gas Temperature
(Source: Tillman and Payette, 2004)
4-47
10581090
Biomass Energy
Cofiring can also reduce the viscosity of slag formed in the cyclone barrel, increasing the
flexibility of coal chosen. Figure 4-28 shows the impact of cofiring on slag viscosity as
determined by FACTSAGE modeling performed by Miller et al. [25]. Note the significant
change in viscosity from 100% bituminous coal to various cofiring approaches.
10000
1000
Fuel 1 (100% coal)
Viscosity, poise
100
Fuel 1
Fuel 2
Fuel 3
10 Fuel 4
Fuel 5
Fuel 9
1
1200 1600 2000 2400 2800 3200 3600 4000
Temperature, °F
Figure 4-28
Impact of Cofiring on Slag Viscosity. Fuels 2 to 9 are Various Cofiring Scenarios Using
Woody Biomass, Herbaceous Biomass, and Animal Manures with Eastern Bituminous
Coals (Source: Miller et al., 2002)
Typically, cofiring has reduced NOX emissions; however, this has not always occurred. At Bailly
Generating Station, NOX reductions of about 0.1 lb/106 Btu occurred when cofiring at 10% wood
(mass basis), and were far more dramatic when blending petroleum coke with wood and coal.
NOX emissions reductions at Allen Fossil Plant were up to 0.2 lb/106 Btu when cofiring at 10%
(mass basis), depending upon the base coal employed. At Willow Island Generating Station, NOX
emissions did not decrease or increase. SO2 emissions decreased as a function of wood
substitution for coal in the boiler.
Testing at Bailly Generating Station [11] demonstrated that carbon monoxide and hydrocarbon
emissions did not increase or decrease as a result of biomass cofiring. The influence on SO3
emissions was inconsequential (approximately 2 ppmvd @ 3% O2).
In all cases, mercury (Hg) emissions decreased as a function of the concentration of Hg in the
feed fuels.
4-48
10581090
Biomass Energy
4.3.3 Cofiring in Pulverized Coal (PC) Boilers
Cofiring in PC boilers can be accomplished either by blending biomass with coal on the main
belt feeding the coal bunkers or by separately injecting biomass directly into the furnace
equipped with either burner modifications or specially designed biomass burners. Blending on
the belt is limited to woody biomass. Herbaceous biomass such as switchgrass causes significant
problems in this application, as previously discussed [29]. With woody biomass, initial research
by TVA showed that the percentage of biomass in the feed to the bunkers must be less than 5%
(mass basis) in order to minimize problems with the mills. Testing at Shawville Generating
Station of GPU Genco (now Reliant Energy) demonstrated that pulverizers were negatively
impacted even at 3% biomass (mass basis). Ongoing work at the Gadsden Station is determining
the maximum allowable particle size for wood input to the pulverizer. If boilers are equipped
with Riley Atritta mills, the maximum biomass percentage may be about 8%.
Most demonstrations of PC cofiring in the United States in both tangentially fired boilers and
wall-fired boilers have focused upon separately preparing and injecting biomass into the primary
furnace. Internationally, there has been much more interest in co-milling of biomass and coal.
Examples of domestic tests and demonstrations include:
• Greenidge Station
• Blount St. Station
• Seward Generating Station
• Gadsden Station
• Ottumwa Generating Station
• Albright Generating Station
• Dunkirk Generating Station.
4.3.3.1 Cofiring System Designs for Pulverized Coal Boilers
There have been several cofiring demonstrations using separate injection. Of these, the Albright
Generating Station of Allegheny Energy Supply Co., LLC provides key insights into design and
cost considerations.
The process design for the Albright Generating System cofiring demonstration somewhat
paralleled the design for Willow Island in concept. However, numerous differences occurred in
equipment selection, largely due to the following factors:
• Moving to an existing demonstration at Seward Generating Station, where equipment had
already been selected based upon different project criteria
• Favoring pneumatic conveyance of sawdust, rather than mechanical conveyance of sawdust
• Injecting the sawdust directly into the boiler, requiring changes in the final handling and
control approaches.
4-49
10581090
Biomass Energy
The Albright design involved accepting sawdust through a walking floor receiver. This receiver
is slightly wider than a walking floor van, and can hold the entire load of a single truck. This
provides for surge capacity in the unloading process. The discharge of the walking floor unloader
is a series of screw conveyors transporting the sawdust to a 30-ton/hr disc screen.
A disc screen was chosen for Albright Generating Station rather than a trommel screen or deck
screen. Dust management was the critical issue, and a disc screen with covers generates less dust
in the work area than a trommel screen. Disc screens have been successfully deployed in the
forest products industry as well as the waste-to-energy industry. This screen produces a product
for suspension firing at less than 6.35 mm (<¼-inch). The size chosen was based upon extensive
research conducted from 1992 to 1995, and the success of both TVA and Southern Company
cofiring <6.35 mm (<¼-inch) sawdust in PC boilers.
The <6.35 mm (<¼-inch) sawdust from the screen was deposited in a bin and then pneumatically
transported to the top of the agricultural silo. Material rejected as being oversized was originally
ground in a research grinder. When that proved unsuccessful in a commercial application, it was
returned to the vendor and the oversized material—less than 5% of the incoming feedstock—was
disposed of. Some employees used a portion of this as mulch on their own property, while some
was landfilled. Eventually a two-stage grinder and associated dust management system was
purchased to reduce the size of oversized product.
From the silo, sawdust was reclaimed on demand, fed through a surge hopper and onto a weigh
belt conveyor feeding a live bottom bin. The live bottom bin discharged sawdust into two rotary
airlocks, each supported by blowers. This arrangement transported sawdust to opposite corners
of the T-fired boiler. Processing equipment was housed in a metal building in order to manage
the entire process. The building also housed the motor control center and all other electrical
systems.
The Albright system delivered up to 6 tons/hr to the boiler, supporting the generation of
approximately 6.0 MWe or about 4.3% of the net generation of the #3 boiler. Injection into the
boiler was accomplished with two specially designed injectors capable of following the burner
tilts. The boiler has four rows of burners, with the injectors installed between the B and C rows.
The burners and sawdust delivery system were designed with a tip velocity of 25.5 m/s (5,000
ft/min) and with an air/fuel mass ratio of 2 kg air/kg fuel. This achieved a sub-stoichiometric
air/fuel ratio and a fuel speed sufficient to overcome the flame speed of biomass. The first
consideration was essential for NOX control; the second consideration was a safety issue.
The system used two independent control systems: one to manage the flow of fuel to the silo and
a second to manage the flow of sawdust to the boiler. The second set of controls allows the
system operator to set the flow of sawdust up to 5.5 tonnes/hr (6 tons/hr) to the boiler at his
discretion. It also permits the operator to shut down the flow of sawdust with an emergency stop
if, for any reason, the operator believes that such action is in the best interest of the boiler and the
power plant.
The design for Seward Generating Station, also using sawdust, was virtually identical to Albright
with one exception: instead of having a separate biomass injector, the Seward wall-fired burners
were modified to fire sawdust into the center of the coal burner. The sawdust and the coal
interacted in the flame envelope (Figure 4-29).
4-50
10581090
Biomass Energy
Both Albright and Seward used separate blowers and separate transport lines for each biomass
injection point. An alternative used at both Plant Gadsden and Dunkirk Generating Station
acquired biomass from the bin using an eductor and a single transport line. This transport line
then employed a splitter to convey the biomass to individual burners.
Figure 4-29
Design of the Sawdust Injection System at Seward Generating Station
4.3.3.2 Capital Costs of Separate Injection PC Cofiring Systems
As demonstrated at Colbert Fossil Plant and Kingston Fossil Plant of TVA, Shawville
Generating Station of GPU Genco, and Picway Station of American Electric Power, the capital
costs associated with cofiring in PC boilers via blending on the main coal belt are comparable to
those of cofiring in cyclone boilers. The capital costs associated with separate injection are
somewhat higher.
Table 4-21 summarizes the capital costs for the Albright project, assuming that it would be
constructed from completely new equipment. Given that the system could support 6.0 MWe, the
installed capital cost was $300/kW output supported by biomass. It may be surprising that the
cost of reasonably automated cofiring systems for cyclone and separate injection PC boilers are
separated by only about $120/kW. However, it must be remembered that cyclone systems must
deliver 24 hours of fuel in a two- to three-hour span (while bunkers are filling). Separate
injection systems deliver significantly lower quantities of biomass to the boiler (or bunkers) on a
per-hour basis.
4-51
10581090
Biomass Energy
Table 4-21
Estimated Capital Cost of the Albright Cofiring Demonstration if Constructed as a
New Facility

Burners
Disc Screen
Overs 2-stage grinder
Dust control for grinder
Screw conveyor
Sawdust receiving bin, blower, And cyclone
Storage silo and unloader
Paddle conveyor
Surge hopper and live bottom bin
Weigh belt feeder
Rotary airlocks and blowers
Walking floor unloader
Motor starters, Motor Control Centers, and
Related
System controls
2 Engineering Studies $82,100
3 Construction $710,000
TOTAL $1,810,400
Source: Tillman and Payette, 2004.
Capital costs associated with herbaceous crop cofiring systems are likely to be substantially
higher than those associated with biomass. Experience at the Ottumwa Generating Station and
Greenidge stations as well as at the Grena, Avedore, and Studstrup Power Plants of DONG
Energy in Denmark demonstrate the need to commit significant resources to dry storage of the
biomass and to extensive grinding systems. Further, because herbaceous biomass is largely a
seasonal product, investments are required to provide this material on a year-round basis.
4.3.3.3 Impacts of Cofiring Biomass in PC Boilers
The impacts of separate injection cofiring solid biomass fuels in PC boilers are distinct and
separate from those for cofiring in cyclone boilers. Biomass cofiring does not reduce boiler
capacity. In winter, when wet coal reduces capacity, biomass cofiring can facilitate recovery
of some of the lost capacity as demonstrated at Seward Generating Station [8][19][31].
Efficiency penalties can be very modest or somewhat significant depending upon system design
and operation. Typically, biomass is introduced with ambient air. This reduces the combustion
air passing through the air heater and raises the temperature of the gaseous combustion products
exiting the air heater. Further, like cofiring in cyclone boilers, moisture and hydrogen in the fuel
4-52
10581090
Biomass Energy
cause a small boiler efficiency penalty. For example, at Albright Generating Station, the
efficiency penalty was about 3.5 Btu/kWh based on sawdust heat input. This resulted in a penalty
of 35 Btu/kWh at 10% cofiring.
Emissions impacts are clear, and significant. SO2 emissions decline as a function of biomass
Btu substitution in the boiler. Similarly, Hg emissions decrease as a function of Btu substitution.
Assuming the test results at Albright Generating Station can be generalized to all units, CO and
opacity emissions are not impacted by cofiring. NOX emissions decrease, particularly as a
function of the location of the biomass injection.
At Seward Generating Station, NOX emissions decreased dramatically when cofiring sawdust,
and the data are represented by Equation 4-5:
NOX = 0.030 + 0.0017(L) + 0.082(EO2) – 1.917(Wh) Equation 4-5
where NOX = oxides of nitrogen, lb/106 Btu, L = load measured as main steam flow in 1000 lb/hr,
EO2 = excess O2 reported in the control room (total basis), and Wh = wood cofiring percentage,
heat input basis. The r2 for Equation 4-5 is 0.93. Further, the probability that any term is a
random occurrence is small. These probabilities are respectively L = 2.09x10-5; EO2 = 2.53x10-5;
Wh = 8.39x10-7; and the overall probability is 4.36x10-6.
NOX can also be expressed on a mass basis, as is shown in Equation 4-6:
NOX = 0.026 + 0.0017(L) + 0.083(EO2) –0.899(Wm) Equation 4-6
where Wm = wood cofiring percentage, mass basis; r2 = 0.93. The probabilities in Equation 4-6
are essentially identical to those of Equation 4-5.
The NOX reduction at Albright Generating Station indicated in Figure 4-30 is represented by
the equation:
NOX = 0.361 – 0.0043(Cm%) + 0.022(EO2%) – 0.00055(SOFA) Equation 4-7
where NOX is measured in lb/106 Btu, EO2% is percent excess oxygen in the flue gas leaving
the furnace (total basis), and SOFA is the total position of the three SOFA damper systems.
Alternatively, the equation is
NOX = 0.157 – 0.002(W%) + 0.009(O2%) – 0.00024(SOFA) Equation 4-8
where the units of NOX are kg/GJ.
As indicated in these equations, Albright Unit #3 was equipped with a low-NOX firing system,
including three levels of separated overfire air (SOFA). The SOFA values used to create these
equations ranged from 15% to 240%, representing a 0% to 100% opening for each SOFA level.
The mechanisms yielding NOX reduction at Albright included reducing the loading on the
pulverizers, improving their performance; creating a reducing zone in the center of the fireball;
and improving the ability to use the SOFA system. With cofiring, SOFA dampers were opened
up to 240% without increasing unburned carbon or loss on ignition (LOI) in the fly ash.
4-53
10581090
Biomass Energy
Experience from the Studstrup plant, which fires 20 tons/hr straw continuously into a 350-MW
unit, indicates few or no impacts on NOX, LOI in ash, corrosion, or heat rate. Other European
cofiring experience, such as that in the UK and Netherlands, similarly suggests little adverse
impact on boiler operations, emissions, or byproducts.
0.6
0.5
NOx Emissions, lb/MMBtu
0.4
0.3
0.2
0.1
0
0.00 2.00 4.00 6.00 8.00 10.00 12.00
Cofiring Percentage, Mass Basis
Figure 4-30
NOX Reduction at the Albright Generating Station
4.3.4 Cofiring Gaseous Biomass Fuels
Gaseous biomass fuels can be cofired in coal-fired boilers, natural-gas-fired boilers, or

combined-cycle power plants. In the latter case, the design with the most near-term potential
introduces biomass gas in a duct burner between the combustion turbine and the heat recovery
steam generator (HRSG). Both producer gas from biomass gasification and natural gaseous
biomass fuels can be used in these applications.
The most frequently reported example of producer gas cofiring was installed at the Kymijarvi
power plant in Lahti, Finland. The power station is a district heating plant that also generates
electricity. This coal-fired generating station has capacity to produce 167 MWe and 240 MWth
of district heat. The utility, Lahden Lampovoima Oy, owns and operates the plant. The Lahti
gasification project involved installing a Foster Wheeler atmospheric circulating fluidized bed
gasifier capable of converting sawdust, bark, wood chips, plywood trim, particleboard trim, and
recycled fuel (REF) into a producer gas for firing in the coal-fired boiler. The REF consists of
plastics, paper, cardboard, and wood. The gasifier has also blended tire-derived fuel (TDF) into
the overall feedstock.
4-54
10581090
Biomass Energy
Biomass fuels are brought to a central processing facility where they are shredded and prepared
as feedstock for the gasifier. The atmospheric CFB gasifier itself is very simple. The system
consists of a reactor where the gasification takes place, a cyclone to separate the circulating bed
material from the gas, and a return pipe for returning the circulating material to the bottom level
of the gasifier.
The hot gas temperature is 1520°F to 1580°F (827°C to 860°C), and the gas is fed directly to the
boiler to recover both the chemical energy and sensible heat from the gas. In tests, the gasifier
supplied 12% to 16% of the total heat input to the boiler. The gasifier was put on line in
September 1998 and has been operational ever since. It has an average availability approaching
88%, with its highest monthly availability slightly above 95% (Raskin et al., 2001). A similar
system is used by Electrabel at their Ruein Unit 5 to contribute 17 MWe out of 540 MWe.
Other utilities have studied similar applications, using gasifiers supplied by a variety of vendors.
Various feasibility studies have estimated the capital costs for such systems as approximately
$600/kW supported by biomass. Such capital costs include both the material handling systems
for biomass fuels, the gasifier and the hot gas piping. If gas cooling and clean-up are required,
the capital costs increase.
Use of wastewater treatment gas and landfill gas is also frequently proposed. Such gaseous
products are at low pressure and, at 500 to 650 Btu/scf, they can be attractive if the gas is used in
immediate proximity to a suitable power plant. The gas can be used in coal-fired boilers, natural-
gas-fired boilers, or in combined-cycle applications. Again it is most suitable as a duct burner
fuel in a combined-cycle application. One such installation exists south of Sacramento,
California.
Wastewater treatment gas is fired in the Allen Fossil Plant, generating green power for TVA. In
this application, wastewater treatment gas from the Maxson Wastewater Treatment Plant serving
the south side of Memphis, Tennessee. Gas is introduced in separate burners above the cyclone
barrels. Although the Allen Fossil Plant approach does not make special use of the gas, landfill
gas or wastewater treatment gas could well be employed as reburn gas.
The database for these applications is insufficient to provide estimates of capital cost or
operating and maintenance costs. Quantities of gas available, specific application, piping
distances, and the degree of gas clean-up required make each case unique. It is sufficient to note
that this approach can be quite cost-effective depending upon the individual situation.
4.3.4.1 Biomass Pre-Treatments for Cofiring with Coal
There are operational challenges associated with cofiring biomass with coal that can be
addressed either by making modifications to power plants or through downstream actions, such
as replacing corroded equipment. And, to a certain extent, many of the technical challenges
associated with cofiring biomass with coal can be avoided via upstream measures, such as
biomass pre-treatment. However, the more advanced the biomass pre-treatment, the higher the
cost of the feedstock. Common pre-treatment options are sizing, compacting, and drying.
Pelletizing is also applicable in cofiring applications.
4-55
10581090
Biomass Energy
The cost of biomass fuel for cofiring with coal, however, depends not only on the procurement
cost but also on the cost of handling, storage, transportation, operability of the boiler, and end
use of the waste. Therefore, it may not necessarily be economical to use the cheapest fuel
available if the negative impact on the boiler operation or fuel operability is significant. The cost
benefit analysis along the complete cofiring chain should be considered in deciding on the
quality and price of the biomass used in cofiring.
The reasons to consider pre-treatment of biomass include:

• Matching the narrow specifications of the feeding system and boiler with a wide variety of
biomass feedstocks;
• Reducing the costs and improve the characteristics of the biomass fuel for transportation,
storage, and handling;
• Reducing the costs and outage time for plant retrofits;
• Minimizing operation and maintenance costs using a homogeneous fuel suitable for
automatic feeding with existing plant systems;
• Increasing the biomass fraction in the boiler feed.
Washing to remove deleterious materials from biomass is not yet commercially ready, and no
known large-scale tests are underway. Torrefaction and pelletizing is nearly commercial and is
expected to be ready for large-scale deployment by 2012.
Table 4-22 lists the pre-treatment options for addressing cofiring constraints.
4-56
10581090
Biomass Energy
Table 4-22
Pre-Treatment Options for Addressing Cofiring Constraints [46]
Biomass pre-treatment via a combination of pelletization and torrefaction is an interesting option

that can mitigate problems related to high transportation and storage costs, handling, and milling
of biomass to an even greater extent than using conventional wood pellets. Compared with
untreated biomass, wood pellets have more attractive properties with regard to heating value,
grindability, storage, transport, and handling. But issues remain regarding the pellets' resistance
to moisture uptake, mechanical resistance to crushing and dust formation, and, of course, cost.
Biological degradation of wood is decreased after pelletization but does still occur. Moreover,
uniformity of pellets is hard to establish, and there are substantial variations between pellets
made out of softwood, hardwood, and different tree species. Additionally, pelletization is
currently mostly limited to sawdust and cutter shavings as economical feedstock, making the
pellet market closely related to the wood processing industry.
Torrefied pellets have advantages over untreated wood pellets. Biomass is very dry after
torrefaction and its moisture uptake is very limited (1%–6%). Biological degradation is almost
halted, volumetric energy density is about 30% higher than that of wood pellets, and the
torrefaction process can be used to upgrade most wood and herbaceous biomass, thus increasing
feedstock flexibility. Torrefied pellets would seem to allow high biomass cofiring in existing PC
coal boilers with minimum investment and impact on existing hardware. Additionally, due to
energy densification, torrefied pellets can decrease long distance transport costs compared with
green wood chips and wood pellets. An ongoing EPRI test will confirm the grindability and
combustibility of pelletized and torrefied biomass.
4-57
10581090
Biomass Energy
4.3.4.2 The Global Biomass Pellet Market
The wood pellet market is growing rapidly—especially in North America, Europe, and Russia—
but remains immature. To date, its availability is heavily influenced by subsidies, resource
availability, fossil fuel prices, and seasonal influences. Certification of wood pellets is not yet a
hot topic, but it is likely to become one in the years to come, once sustainability requirements for
biomass are provided in Europe and North America.
The main factors driving international pellet trading over the next five years are:
• Increasing CO2 emissions prices;
• Promotion of biomass cofiring with coal in existing power plants;
• RPS policies;
• Increasing oil and natural gas prices.
The primary future producers and exporters of wood pellets are expected to be Canada, the
United States, Russia, Belarus, Ukraine, and Latin America. Meanwhile, the main pellet users
and/or importers are likely to be Canada, the United States, Germany, Austria, the UK, and
Ireland.
The anticipated logistical challenges that must be tackled to support the emergence of a global
biomass pellet market include:
• Development of dedicated pellet terminals at major harbors;
• Commercial development of advanced treatment options (e.g. torrefied pellets);
• Development of loading and unloading equipment;
• Prevailing high ocean long distance shipping rates.
Large facilities for export or ship transport should optimize logistical costs and allow for
international trade in biomass similar to the present coal trade. Coal power plants located close to
the coasts and designed for imported coal may be in the best position to take advantage of large
shipments of pellets by barge at internationally competitive prices from such sources as Brazil
and other Latin American countries, Canada, Russia, Ukraine, among others.
4.3.5 Conclusions Regarding Biomass Cofiring
Depending upon the biomass available and the technology of choice, biomass can be cofired in
cyclone boilers, wall-fired and tangentially fired PC boilers, natural-gas boilers, and combined-
cycle plants. These applications can be cost-effective depending upon the plant, the supply and
cost of the biomass, and the design employed. Cofiring is the least-cost approach to biomass
utilization in electricity generation and can be deployed relatively quickly.
Research is ongoing to address the impact of biomass cofiring on the life of deNOX catalysts and
on fireside corrosion. In both cases, the alkali/alkaline earth constituents prevalent in the biomass
may adversely impact catalyst poisoning or corrosion reactions.
4-58
10581090
Biomass Energy
4.4 Stand-Alone Biomass-Fired Systems
Stand-alone biomass-fired systems designed to produce electricity include:

• Stand-alone wood or wood-waste-fired generating stations producing electricity as an
exclusive product or producing both electricity and process steam in cogeneration or
combined-heat-and-power (CHP) applications.
• Spent pulping liquor (e.g., black liquor from Kraft pulping) combustion systems designed to
recover pulping chemicals while producing steam for both electricity generation and process
heat applications.
• Bagasse-fired boilers used to generate electricity in the sugar industry.
• Other agribusiness systems with biomass either fired directly in a boiler or gasified and gas
fired in a waste heat boiler with steam subsequently used for electricity generation.
• Boilers fired with landfill or wastewater treatment gas, designed to produce electricity.
• Combustion turbines fired with landfill or wastewater treatment gas.
• Internal combustion engines, typically diesel engines, fired with landfill gas, wastewater
treatment gas, or producer gas from woody or herbaceous biomass gasification.
Of these, wood-waste-fired boilers and chemical recovery boilers are the dominant sources of
biomass electricity today. Black liquor from Kraft pulp mills is used to produce more electricity
than any other form of biomass, followed closely by wood waste.
Chemical recovery boilers and associated systems are specific to the pulp and paper industry,
and are not considered further here. The wood-waste-fired boiler is the focus of this discussion,
as it provides the most promise for electricity generation.
4.4.1 Wood-Waste-Fired Boilers
Wood-waste-fired boilers, also known as hog fuel boilers, have long been used in the forest
products and pulp and paper industries to generate process steam and cogenerate both electricity
and process steam. Major integrated forest products mills producing a product mix including
Kraft pulp, plywood, and dimension lumber commonly installed such boilers as power boilers.
Examples abound, including the Longview #11 boiler at the Weyerhaeuser complex in
Longview, Washington and comparable power boilers at Weyerhaeuser facilities in Columbus,
Mississippi, Plymouth, North Carolina, and New Bern, North Carolina. Georgia Pacific,
International Paper, and others have similar installations.
Electric utilities and independent power producers also constructed wood-waste-fired boilers to
generate electricity as their sole product. Large units have been constructed by Washington
Water Power at Kettle Falls, Washington; Burlington Electric at the McNeil Generating Station
in Burlington, Vermont; and at other locations. Independent power producers built similar
systems to fire wood waste and wood-like biomass fuels. Large numbers of such systems were
built in California, Maine, and elsewhere in response to PURPA, the need for capacity in certain
areas of the country, and localized availability of materials. The Shasta Generating Station in
Anderson, California is a prime example of such facilities.
4-59
10581090
Biomass Energy
4.4.1.1 Typical Capacities of Wood-Waste-Fired Plants
While coal is produced at a central mine location and distributed to users, wood waste is
produced as a distributed resource over a large area and must be collected or gathered and
shipped to the user location. This fundamental distinction highlights one capacity limitation
impacting biomass systems. The second limitation results from the typical fuel characteristics:
•
3
15 to 20 lb/ft bulk density
• 40% to 50% moisture
• 8 x 106 to 10 x 106 Btu/ton as-received.
These factors combine to limit boiler and generating capacity. Given typical transportation
distances for wood fuel of up to 50 miles (80 km), wood-fired boilers have been limited to a
nominal 100 to 125 ton/hr firing rate (nominally 10,000 ft3/hr of fuel), or 50 to 70 MWe
depending on system design and operation. This capacity limitation carries with it significant
implications:
• It is difficult to economically justify more than three turbine extractions for feedwater
heaters; the three extractions are typically for a low-pressure heater, the deaerater, and a
high-pressure heater. In some cases only one extraction (deaerater) is justified.
• Reheat cycles that significantly improve cycle efficiencies are not economical at the small
plants available for wood firing.
• Depending on staffing philosophies, such units will achieve ratios of 0.5 to 2.0 MWe per
worker, depending upon unit capacity and staffing approach. This contrasts with
conventional coal-fired staffing ratios of 3 to 6 MWe/worker.
Typical stand-alone wood-fired power plants have been built at capacities ranging from 11 MWe
to 70 MWe, and the larger plants were most successful.
4.4.1.2 Wood-Waste-Fired System Design
Numerous wood-waste-fired systems have been developed for the current market. In order to
evaluate system design, a generic plant is considered. It represents standard practices in wood-
waste-fired technology. Variations on this approach are also discussed below.
4.4.1.2.1 Fuel Handling
Figure 4-31 shows the standard fuel handling approach. Loaded trucks of wood waste are
received and unloaded by hydraulic unloaders that raise the truck sufficiently to dump the load
into a hopper and feed it to a disc screen. Oversized particles are hogged using hammer hogs or
(rarely) knife hogs. The entire fuel pile is then stocked out. Wood waste or hog fuel is then
reclaimed and transported to the plant, where it is discharged into bunkers.
4-60
10581090
Biomass Energy
X-110 F1
Truck Unloader/Scale F-111
Receiving Hopper/Magnet
H-112
F2 Rotary Disc Screen
F3
J-114 F-116 RDS Pile
RDS Conveyor
C-113
Hammer Mill
F4
F1
J-115
HM Conveyor F-117 HM Pile X-120 Dozer F-122 Fuel Storage Pile
F-132
F1 Feed Bin
F-130
X-121 Dozer Reclaim Hopper/Pit
J-133
Feed Conveyor
F6
B
F5
J-131
Reclaim Conveyor/Scale F-134 Excess Pile
Figure 4-31
Representative Wood Fuel Handling System
Because waste wood truckloads typically contain 20 to 25 tons, large plants such as McNeil
Generating Station or Kettle Falls Generating Station require a significant commitment to wood
fuel handling. A 50-MWe plant burning 80 to 100 tons/hr requires four to five trucks per hour, 24
hours per day, seven days per week. Burlington Electric chose another approach to keep truck
traffic out of the plant neighborhood. They receive wood waste off site, load it into open hopper
railroad cars, and transport the biomass to the plant. The wood is discharged at a trestle facility.
This has plant benefits, but increases the cost of materials handling.
Weyerhaeuser at Longview, Washington, uses a wood fuel stacker-reclaimer arrangement
instead of bulldozer reclaiming. This helps manage the 100+ tons/hr burned in its #11 boiler.
Numerous studies have addressed the addition of drying systems to the materials handling as a
means for improving boiler efficiency. These are not generally employed. Dryers require energy
to operate, consequently reducing the efficiency gain. Further, drying technologies introduce the
potential for fires, which have occurred at installations that use fuel dryers. Consequently, the
general approach is to use the boiler as the dryer unless site-specific circumstances favor use of a
separate dryer.
4.4.1.2.2 Boiler and Turbine-Generator Technology
The traditional wood-fired boiler used in electricity generation is a spreader-stoker. Figure 4-32
shows a typical flowsheet for a stoker boiler. Wood is discharged from the surge bin via a twin
counter-rotating auger system and fed to a stoker. Unlike the typical coal stoker, which is a
4-61
10581090
Biomass Energy
paddlewheel device, a wood fuel stoker is an air-swept design that throws the wood fuel onto the
grate. Figure 4-33 depicts a typical boiler and turbine-generator system, along with feedwater
heaters, pumps, and fans.
Mechanical
draft cooling
tower
Cooling
water
generator
Steam (1250 psi, 950 deg F)

Moisture losses Ammonia
Wood 710 F
blend On-site 308 F
wood Stoker boiler Air heater Fabric filter
preparation
638 F 80 F
Combustion Stack
Fines Bottom ash air Fly ash
Ferrous
Figure 4-32
Schematic of Biomass-Fired Stoker Boiler Power Plant
The plan area heat release for wood-waste-fired boilers is typically on the order of 750 x 103 to
1 x 106 Btu/ft2-hr. Most coal-fired stokers are in the 500 x 103 to 750 x 103 Btu/ft2-hr range. The
volatility of the wood fuel allows a higher area heat release rate. The furnace volumetric heat
3
release for wood waste is typically 13,000 to 20,000 Btu/ft /hr, compared to coal at 20,000 to
25,000 Btu/ft3/hr. The lower calorific value of the wood fuel and higher moisture content reduces
the volumetric heat release. Evaporation of fuel moisture to a vapor can require significant
furnace volume.
4-62
10581090
Biomass Energy
Q-210 5
Stoker Furnace 900
967.4
N-300 N-310
STM Turbine 25 MW Generator
4
275
1105 6A
700 G-226
500 Ejector
7A
400
NH4 Grid
100
X-215
E-217 8A 9
Economizer 210 125.5
ATM
10
2
3
250 E-224 10B
1110 Condenser 100
25
10A
80
30
NH4
CA CA E1
ATM
E-218
Air Heater 7A
X-211 400
100
Oil Burner (4X)
200 13
90 175
140
D-223
6A Deaerator 8A
WASTE 700 210
CA CA 500 10
A2
200
CA
E-221 E-225
CA CA FW Heater 8B
FW Heater
B 6B 130
300 4
A1 490
BBD 59 11
H-219 125.5
X-212 1.96
Air Filter
Air Swept Burner (4X) ATM
G-216
ATM Blower 12
X-213 2 DMP
Air Cooled Grate ATM 200 125.5
1160
1 145
200
G-220 L-222 90 G-227
J-214 FD Fan FW Pump C Pump
Bottom Ash Screw Conveyor
WASTE
Figure 4-33
Typical Boiler and Turbine-Generator Approach
Grate selection is of consequence. The generic design presented here uses a water-cooled grate
design such as the Detroit Stoker Hydrograte. Alternatives include air-cooled pinhole grates,
where primary or undergrate air not only provides oxidant to fuel on the grate, but also cools the
grate bars themselves.
Steam conditions are a function of the design capacity of the boiler. For lower-capacity boilers
(e.g., 250,000 lb/h steam), 600 psig/750°F conditions are common. For medium-capacity boilers,
the steam conditions are often 850 psig/825°F, and higher capacity units typically use 1250
psig/950°F and 1450 psig/1000°F steam conditions. The larger units use more feedwater heaters
to improve the thermal efficiency of the unit.
Alternative circulating and bubbling fluidized-bed boilers and gasifiers coupled to waste heat
boilers may also be used. Figure 4-34 shows a typical fluidized bed combustion power plant
flowsheet. Selection of a steam generator depends upon the type and quality of biomass available
as well as the site and project-specific economics. For example, the Tacoma Steam Plant #2 was
repowered using two bubbling bed steam generators coupled to the existing boilers, used as
waste heat boilers. Tacoma Public Utilities fueled the units with a mixture of wood waste,
refuse-derived fuel (RDF), and coal.
4-63
10581090
Biomass Energy
Mechanical
draft cooling
tower
Cooling
water
generator
Steam (1250 psi, 950 deg F)

Moisture losses
Wood 710 F
blend On-site Fluidized 312 F
wood bed Air heater Fabric filter
preparation boiler
642 F 80 F
Combustion Stack
Fines Sand Bottom ash air Fly ash
Ferrous
Figure 4-34
Schematic of Biomass-Fired Atmospheric Fluidized Boiler Power Plant
Turbines are selected based on unit capacity and whether the system is designed as a stand-alone
power-only plant, or whether it is a cogeneration or CHP plant. Steam turbines can be designed
with automatic extractions for process steam or as backpressure turbines exhausting process
steam at 50 psig, 150 psig, or other conditions depending upon the plant requirements.
Alternatively, the turbines can be designed and supplied to exhaust steam at 2 to 3 in HgA if
power is the exclusive product. Typical water rates (lb steam/kWh) for Rankine-cycle turbines
used in wood-waste-fired power plants are shown in Tables 4-23 and 4-24. Table 4-23 presents
water rates for condensing turbines, while Table 4-24 presents water rates for backpressure
cogeneration turbines.
Table 4-23
Typical Water Rates for Condensing Turbines in Wood Waste-Fired Plants
Turbine Throttle Pressure Water Rate

Psig/°F Atm/°C lb steam/kWh kg steam/kWh
600/750 41/400 9.8 4.4
850/825 58/440 8.7 3.9
1250/950 85/510 8.3 3.8
1450/1000 102/540 8.0 3.6
Note: Assumes 3-inch HgA turbine exhaust.
Source: Tillman, Rossi, and Kitto, 1981.
4-64
10581090
Biomass Energy
Table 4-24
Typical Water Rates for Backpressure Turbines in Wood Waste-Fired Plants
Turbine- Turbine Throttle Pressure Water Rate

Generator Cap. Psig/°F Atm/oC lb steam/kWh kg steam/kWh
15 600/750 41/400 31.6 14.3
15 850/825 58/440 23.6 10.7
15 1250/950 85/510 18.4 8.3
20 600/750 41/400 30.9 14.0
20 850/825 58/440 23.4 10.6
20 1250/950 85/510 18.1 8.2
25 600/750 41/400 30.6 13.9
25 850/825 58/440 23.2 10.5
25 1250/950 85/510 17.9 8.1
Note: Assumes 10.2 Atm or 150 psig exhaust pressure.
Source: Tillman, Rossi, and Kitto, 1981.
Given the typical boiler efficiencies of well-run wood-waste-fired boilers, the net station heat
rates (NSHR) for small condensing wood-fired plants are typically in the 15,000 to 17,000
Btu/kWh range, while the NSHR values for the 40+ MWe units are typically in the 13,000 to
14,000 Btu/kWh range. NSHR values depend upon the capital investments made in such
equipment as economizer and air heater surface, feedwater heaters, and like equipment. Balance
of plant becomes a critical concern for these systems. For more severe steam conditions,
feedwater treatment requirements become more substantial.
The incremental net station heat rate for cofiring systems is far more attractive. For backpressure
turbines with throttle steam conditions of 1250 psig/950°F and exhaust conditions of 50 or 150
psig, the NSHR appropriate for the power generation element is on the order of 4,500 to 5,000
Btu/kWh. This is based upon the fact that many process uses of steam capitalize on the heat of
vaporization. Consequently, losses associated with the heat of vaporization are ascribed to the
process user. The losses attributed to cogeneration are related to boiler efficiency (or boiler
thermal losses) and turbine efficiency (or turbine losses). With cogeneration systems based upon
extraction steam, NSHR values can be on the order of 7,500 Btu/kWh or more, depending on the
proportion of steam used in cogeneration applications and the proportion of steam condensed in
power generation application.
4.4.1.2.3 Pollution Control Systems
Air pollution equipment for solid biomass-fired systems includes either fabric filters (FF) or
electrostatic precipitators (ESP), and either selective catalytic reduction (SCR) or selective non-
catalytic reduction (SNCR) systems. Acid gas scrubbers are not required due to the compositions
of typical solid biomass fuels. For this analysis, an ESP was assumed, with multi-clones installed
ahead of the ESP. Multi-clones are useful in scavenging stray glowing embers should they leave
the economizer and air heater sections of the steam generator “hot.” A SNCR system was also
assumed, as firing wood waste typically generates less NOX than firing coal in larger utility
boilers.
4-65
10581090
Biomass Energy
Figure 4-35 depicts the air pollution control system along with a mechanical cooling tower and a
demineralizer system for feedwater treatment.
Air Quality Control System
E1 E2
E1
230
210
14.7 Cooling Tower
14.7
ATM
BBD
H-419 C2 P-417
C2
100
H-420 25
J-421 J-422 G-423 X-424 K-425

C1
C1 C3
WASTE 80
30 L-416 L-418
AQCS - SNCR System

Demineralizer Plant
N3
NH4
N4 N6 C3
H-412 NH4
F-411
N2 M-415
N1 L-413
L-410
DW RW
RW N5
L-414
P-419
Figure 4-35
Air Pollution Control System, Cooling Tower, and Demineralizer for a Typical Wood-Waste-
Fired Plant
4.4.1.3 Comparisons to Other Biomass-Fired Rankine-Cycle Designs
Boilers comparable to the wood-waste-fired boilers described above have been used for
agricultural wastes. Such wastes include nut hulls and shells, stone fruit pits, vineyard prunings,
orchard prunings, rice hulls, corn stover, spent corn cobs, and a host of other materials. Wheat
straws and comparable materials have been fired in stoker and fluidized-bed boilers in both the
United States and Europe. All of these materials can be fundamentally different from wood
waste as follows:
• Higher nitrogen content than wood waste.
• Higher chlorine content than wood waste.
• Higher inorganic matter than wood waste with the wood waste containing significantly
higher concentrations of calcium, potassium, sodium, and phosphorus.
• Higher bulk densities than the biomass fuel, with the exception of stone-fruit pits
(e.g., peach pits).
Designers of boilers firing such materials must take into account the significantly higher slagging
and fouling propensities of these materials when compared to typical wood waste (see Baxter et
al. [23][24]). Experience in the Central Valley of California and elsewhere has shown that the
combination of ash chemistries and chlorine has increased ash deposition in the boiler. Some of
4-66
10581090
Biomass Energy
these biomass materials have been sufficiently different in characteristics that special systems
have been used. Rice hulls, for example, have been suspension-fired. They have also been
gasified and the product gas fired in a waste heat boiler for steam generation and subsequent
power generation.
Rankine-cycle generation has also been used in firing landfill gas and wastewater treatment gas.
For example, the 10 million ton/yr Puente Hills landfill has a 50-MWe Rankine-cycle generating
station. Similar boilers are installed at the Spadra and Coyote Canyon landfills, also in the Los
Angeles County area.
Primary differences include fuel handling and pollution control. The gaseous fuel is produced at
low pressure and saturated with water vapor. Further, it can contain solid contaminants and some
chlorinated compounds. Gas treatment for such biogas-burning boilers may include gas
dehydration, particulate scavenging, and other gas treatments depending upon the design
principles used. Flue gas treatment focuses more on NOX removal than particulate capture.
Operating and maintenance costs for wood-waste-fired generation are typically site-specific,
depending on the owner, the location, and its proximity to related facilities. For example, a
project located on the site of an integrated forest products mill can share certain maintenance
facilities and costs. An independent plant must bear the entire cost of operations and
maintenance.
Typically, a wood-fired plant requires about 20 to 30 employees, including a plant manager,

engineer, business manager, secretary, four shifts with three employees/shift, two fuel yard
employees, and a maintenance crew including both millwrights and electricians. Nominally, a
crew of about 25 is adequate for plant operations. However, larger stations (e.g., 40+ MWe)
generally are more heavily staffed to handle increased fuel receipts and the increased complexity
of the unit. Alternatively, outsourcing can be used to manage the staffing complement. For
smaller generating stations up to 15 MWe capacity, staffing is about 0.5 MWe capacity/worker.
Larger units may achieve better ratios such as 1 to 2 MWe capacity/worker.
Maintenance requirements are comparable to those associated with coal-fired generating stations;
the same types of problems occur. Annual maintenance costs can be estimated at 4% to 5% of
the total capital cost of the facility. However, this is a generalized number, and site-specific
maintenance costs vary significantly.
4.4.2 Repowering Existing Coal Units for Biomass Firing
The Schiller station in Portsmouth, New Hampshire is the site of a novel repowering of an
existing coal unit to wood firing. The station consisted of three 50-MW coal-fired units. PSNH
and its subsidiary, Northern Wood Project, opted to decommission one of the boilers and replace
it with a dual-fuel fluid bed boiler. While the unit is designed for both coal and wood, it is
anticipated that only wood will be fired. The new boiler was integrated with the existing steam
circuit, which kept costs down. Additionally, a wood receiving yard was necessary to
accommodate truck delivery of wood. The unit came on line in December 2006 and early reports
suggest the unit is performing up to expectations. The cost of repowering was about $1,400/kW.
4-67
10581090
Biomass Energy
Repowering smaller and older coal units may offer significant advantages to utilities. It allows
them to utilize existing sites and local established infrastructure. The wood retrofits would
support the local forestry industry, contribute to the utility’s renewable portfolio, and in general
reduce the site emissions profile.
4.4.3 Ownership Models and Institutional Issues
Traditionally, biomass-fired generating stations are owned either by a utility (e.g., Burlington
Electric and Washington Water Power), by a non-utility generator (NUG) such as a forest
products company, or by an independent power producer (IPP). IPPs have installed more than
1,000 MWe of wood-waste-fired capacity in California in response to PURPA and the California
Standard Offer #4 (SO4). As SO4 expired, construction of new facilities halted and several
smaller units were shut down.
An alternative ownership model has been used by selected utilities located where large forest
industry and pulp and paper complexes exist. In this situation, a utility owns a turbine-generator
within the site of a forest industry mill and its cogeneration installation. The utility purchases
high-pressure steam from the boiler and sells low-pressure process steam back to the pulp mill or
integrated complex. The utility has a lower capital investment and a lower net station heat rate.
The turbine is dispatched by the forest industry, consistent with its needs. Such ownership
arrangements are niche applications that are mutually beneficial.
Other institutional issues include regulatory constraints and incentives. There is currently
considerable uncertainty in the regulatory and incentive arenas, which makes definitive
observations difficult if not impossible to make. There are some who would, for political
reasons, consider wood waste “not renewable.” Others argue whether wood waste in various
forms is “green” or “clean” power. Regulatory constraints are equally uncertain.
4.4.4 Alternative Systems for Gaseous Biomass Fuels
Alternative power generation systems for gaseous biomass fuels such as landfill gas and
wastewater treatment gas include internal combustion engines, microturbines, and larger (e.g.,
5000 kW) combustion turbines. Wastewater treatment gas includes the product of municipal
wastewater treatment plants as well as anaerobic digestion of animal manures. Traditionally,
wastewater treatment plants have used internal combustion engines if appropriate for power
generation. These systems can be installed in 1-MWe increments, although both smaller and
larger systems are employed for both landfill gas and wastewater treatment gas. Such
installations require gas compression and are frequently backed up with either natural gas or
propane.
More recently, fuel cells and microturbines have been applied to these gaseous biomass fuels on
a somewhat experimental basis. Small 1- to 5-MWe combustion turbines have also been installed.
The Puente Hills landfill pioneered the use of small combustion turbines for electricity
generation. For all systems, attention must be paid to concentrations of hydrogen sulfide (H2S) at
100 to 2,000 ppmv, particularly in wastewater treatment/digester gas. Siloxanes also are a
contaminant with significant potential for causing combustion and environmental issues.
4-68
10581090
Biomass Energy
The capital costs of these alternatives have been documented by CH2M Hill and others
(see Kitto and Green [32]):
• Internal combustion engine: $3,000/kW (1-MWe system basis)
• Fuel cell: $9,000/kW (300-kW system basis)
• Microturbine: $4,000/kW (30-kW system basis)
• Combustion turbine: $1,000/kW (5-MWe system basis).
The combustion turbine, in particular, is penalized (relative to natural gas turbines) by the small
scale of the equipment and by the need for gas cleaning and compression.
Operating and maintenance costs for these systems can be both site specific and technology
specific.
4.4.5 Near-Commercial Technologies
Hundreds of biomass gasifiers are deployed worldwide (there are at least 50 suppliers), but
almost all of them are unsuitable for utility power applications. There are three primary reasons
for this. First, almost all of them are too small for utility application. Most are rated at less than
1 MW thermal, and would consequently have high operating costs. Second, the gas quality of
many of the small gasifiers is quite low and is suitable for direct combustion as required for
process heat or district heating or, in some cases, running small reciprocating engines with
associated high maintenance costs. And finally, while small gasifiers have been shown to be
robust operationally, scale-up to utility sizes (approaching 10 MW and greater) is problematic.
Demonstrations of larger-scale gasification technology coupled with district heating are

underway throughout Europe and show promise both technically and economically. These are
demonstrations of gasifiers that are either already at utility scale, or could easily be scaled to
larger scale. Larger configurations reduce the operating cost by spreading fixed costs (i.e.,
operations staff) over a larger output. Economies of scale also reduce variable operating unit
capital cost.
The following section briefly describes some key demonstrations of atmospheric gasification in
Europe, and their role in power production. With the exception of the gasifier at Ruien, the
projects highlighted below (and several more) are first-of-a-kind demonstrations. Consequently,
cost and performance information is difficult to acquire and interpret. Other demonstrations, such
as the biomass gasification/combined-cycle demonstration at Värnamo, have been completed and
provide a reasonable technical backdrop, but little cost information. At this time, there are no
active biomass gasification power demonstrations in the United States.
4.4.5.1 Wood Gasification for Power and District Heating in Güssing, Austria
The Güssing Gasifier is a 2-MWe, 4.5 MWth, circulating fluid bed gasifier. Wood harvested from
local forests is the fuel. The village of Güssing has 56 km of district heating supplying the local
hardwood flooring industry, other industries, as well as a residential heating network. Repotec
4-69
10581090
Biomass Energy
owns the gasifier technology, developed by the Technical University of Vienna, and has built a
larger unit at Heiligenkreuz, Austria. The Heilegenkreuz unit is designed to produce
11 MWe/6 MWth, and was commissioned in late 2006.
Figure 4-36 shows the overall process flow diagram. The dimensions of the gasifier vessel are
2.5-m ID, 5.5-m OD, and 8-m tall. The vessel is refractory lined and has an external combustion
zone or combustion leg, operating at nominally atmospheric pressure, with steam injection in the
bottom of the vessel to keep N2 levels low. The ash is stored and, at some point in the future, is
expected to be cleared by environmental authorities to be spread on fields. It takes about 12 to 15
hours to start up the gasifier and two days to shut down, mostly due to thermal stress limitations
on the refractory. The gasifier could be scaled up easily, according to the staff at the site, but
would require multiple combustion legs around the central gasifier rather than a single
combustion leg as currently configured. The bed temperature is 850°C and the gas temperature is
about 830°C at the gasifier exit.
Tar less than 20
150C
830C
850C
Figure 4-36
Schematic of Biomass Gasification Application for Combined Heat and Power
The syngas reports to the gas cooler, where it is cooled to 150°C to generate steam, which goes
to the district heating network. The cooled gas goes to the baghouse, where most of the
particulates are removed and reintroduced into the combustion leg. The baghouse is pulsed with
nitrogen. The particulate-free gas reports to the scrubber, which uses biodiesel to scrub tars and
condense H2O. Incoming tar levels are 500 ppm, and less than 20 ppm tar is in the product gas.
4-70
10581090
Biomass Energy
Then the cleaning liquid is separated, and the middles (tar/water/biodiesel) are introduced into
the combustion leg. The gas composition is 22% CO, 23% CO2, 40% H2, 10% CH4, 4% N2.
4.4.5.2 Wood Gasification at Harboøre
A Vølund-designed updraft gasifier has been installed and operated at Harboøre in western
Denmark since 1988. Figure 4-37 shows a process flow diagram of the system. The gasifier
dimensions are about 2.2-m ID and 12-m high. It operates at atmospheric pressure. The
feedstock is nominally 40 mm green wood chips, delivered from local suppliers. The unit
consumes 1.5 tons/hr at maximum load. The chips are delivered to the top of the gasifier, where
a spreader turning in the middle distributes the chips. When the torque on the spreader decreases,
more wood is added. At the bottom of the gasifier, humidified air enters the gasifier, which
consumes 3 to 4 Nm3 water per day. The temperature is 1200°C at the bottom of the gasifier
vessel and the product gas leaves the top of the gasifier at 75°C. The gas composition is 30%
3
CO, 20% H2, 5% methane, 8% CO2, 3% N2, and the heating value of 6 MJ/nm , compared to
3
natural gas at about 40 MJ/nm . This does not include the dilution of the humidified air. The gas
contains 100 g/nm3 of tar. For firing the gas in IC engines, its tar content needs to be less than
50 mg/nm3. Ash is drawn out of the bottom, and the annual production is about 50 t/year.
Figure 4-37
Process Flow Diagram for the Vølund Gasifier at Harboøre
The gas is cooled in two stages to 35°C, and then enters a wet ESP. The product gas is clean
enough to be burned in Jenbacher engines. The ESP effluent, a water/tar/particulate mixture,
reports to a settling vessel. The heavy tar is stored in a recirculated, heated vessel for combustion
during the winter months as “peaking heat.” The water/soluble tars are heated until vaporized,
then combusted for additional district heating capacity.
4-71
10581090
Biomass Energy
The electric capacity of the unit is 1.3 MW and the thermal service is up to 3 MW. The supplier,
Babcock and Wilcox Vølund, has indicated that it should be feasible to scale this gasification
process to about 20 MW total energy output.
4.4.5.3 Gasification of Wood Pellets at Skive
In Skive in northern Denmark, the local utility, Fjernvarme, has an extensive district heating
network, currently fired predominantly by wood pellets and natural gas. There is some electricity
generation as well. The heating network encompasses 8000 homes and is growing. To meet that
growth, Fjernvarme is building a wood gasifier at its Thorvej site. The gasifier will produce
6 MW electric power in three 2-MW Jenbacher engines, and 20 MW of steam heat in two
10-MW boilers. Figure 4-38 shows the process flow diagram for the plant.
Figure 4-38
Fjernvarme Process Flow Diagram for Coproduction of 6 MW Electric and 10 MW Thermal
The gasifier is a Carbona slightly pressurized gasifier. One of the unique features of this gasifier
is a tar cracker to address the tar problems that other gasifiers have had. Fjernvarme is producing
electricity primarily due to government mandate and would prefer to concentrate on producing
hot water.
The Thorsvej site already has two large wood pellet-fired stoker boilers on site, so the fuel
handling storage and infrastructure is already largely in place.
The gasifier at Skive was under construction at the time of this writing. Originally, the unit was
to have been commissioned in summer 2006, but business issues with the constructor have
introduced delays.
4-72
10581090
Biomass Energy
4.4.5.4 Cofiring Syngas at Ruien
In western Belgium, Electrabel’s Ruien power station has begun to embrace biomass fuels as
part of its goal to displace coal-based carbon. A part of that strategy is to install an atmospheric
fluid bed gasifier to provide syngas to the boiler. The gasifier operates on sized green wood. The
feed is generally coarse 50-mm wood chunks. Wood is delivered via walking floor truck,
screened for oversize and tramp material, and conveyed to a covered storage area. The wood is
extracted and conveyed to the gasifier. Figure 4-39 shows a schematic of the gasification facility
at Ruien. Note that the gasifier has only particulate removal for gas cleanup. This is all that is
required, since the gas is directly injected into the boiler. The injection of the fuel gas occurs
below the lowest coal firing level of Unit 5. The syngas contributes about 17 MW to the 190-
MW rated capacity of the unit.
Figure 4-39
Overall Configuration of the Biomass Gasifier at Ruien
While there have been a few issues with this unit, it has mostly operated very smoothly. There
are indications that using dry wood causes more operational problems than fresh, green wood,
and consequently, Electrabel strives to have mostly fresh wood delivered.
4.4.5.5 Wood Gasification at Kokemäki
The town of Kokemäki, Finland, is the site of a demonstration of the Condens Oy “Novel”
gasifier. The gasifier is a fixed-bed unit, and is fueled by local wood supply. The unit has an
electrical output of 1.8 MWe and a thermal output of 4.3 MWth. The electricity is generated by
syngas-fired reciprocating engines. The unit has been in operation since 2005. The vendor
indicates that capacity could be increased to 10 MW or greater. Figure 4-40 shows a schematic
of the Kokemäki gasification system.
4-73
10581090
Biomass Energy
Figure 4-40
Schematic of Biomass Gasification and Power Generation System at Kokomäki
4.4.5.6 Carbon Negative Effects Associated with Bio-Char from Biomass
Ordinary biomass fuels are carbon neutral. However, bio-char systems can be carbon negative
because they retain a substantial portion of the carbon fixed by plants via photosynthesis. The
result is a net reduction of carbon dioxide in the atmosphere. Under controlled production
conditions, the carbon in the biomass feedstock is captured in the bio-char and the bio-energy co-
products. While it is technically infeasible to capture 100% of the biomass carbon, the optimal
bio-char-producing process can capture roughly half the biomass carbon as bio-char and half as
bio-energy. Bio-char can sequester or store carbon in the soil for hundreds or even thousands of
years. It also improves soil fertility, stimulating plant growth which then consumes more CO2 in
a feedback effect. Moreover, the energy generated as part of bio-char production can displace
carbon-positive energy from fossil fuels.
4-74
10581090
Biomass Energy
Bio-char production processes utilize cellulosic biomass such as wood chips, corn stover, rice
and peanut hulls, tree bark, paper mill sludge, animal manure and most urban, agricultural, and
forestry biomass residues. Bio-char can be produced by pyrolysis or gasification systems, which
can be developed as mobile or stationary units. Outputs beside bio-char are bio-energy in the
form of either syngas, or bio-oils which can be used to produce heat, power, or combined heat
and power.
The co-production of bio-char from a portion of the biomass feed-stock reduces the total amount
of bio-energy that is produced by the technology. But even at today’s energy and fertilizer prices,
the net gain in soil productivity is worth more than the value of the energy that would otherwise
have been derived from the biomass feedstock. As the cost of carbon emissions rises and the
value of CO2 extraction from the atmosphere is also considered, the balance becomes
overwhelmingly more attractive for bio-char co-production
Economics are the biggest barrier to mass commercial adoption of bio-char. Until carbon
markets are established, allowing farmers to earn credits for applying bio-char to their fields, the
bio-char market will not fully develop at a global scale. Bio-char soil management systems can
deliver tradable carbon emissions reductions, and sequester carbon in a manner that is easily
accountable and verifiable. The negotiations for a new post-Kyoto carbon trading standard in
2012 are likely to include bio-char as a key source of high-quality carbon credits.
4.4.6 Non-Commercial Technologies
A host of technologies are commonly proposed for biomass, either to supply fuel or to utilize
that fuel successfully. These technologies include (but are not limited to):
• Energy farms to produce woody biomass.
• Energy farms to produce herbaceous biomass.
• Direct combustion or gasification of animal manures.
• Pressurized biomass gasification for use in integrated gasification/combined-cycle
combustion turbine systems with or without coal gasification systems.
• Technologies for production and utilization of heavy pyrolysis liquids
• Indirect liquefaction of biomass fuels by Fischer-Tropsch or other technologies based upon
gaseous feedstocks from landfills, wastewater treatment plants, and producer gasifiers.
To date, each of these technologies has encountered significant technical, economic, and
institutional barriers sufficient to inhibit implementation on a commercial utility scale. Further,
the traditional commercial test—the ability of vendors to give guarantees—has not been met.
Consequently, the technical and economic information available to provide performance or cost
data is somewhat speculative at this time.
4-75
10581090
Biomass Energy
Some of these technologies may never achieve commercial status. Many biomass technologies
have already been developed but have not succeeded commercially, including:
• Cyclone combustion of wood with the addition of inorganic material
• Various gasification systems
• Mobile pyrolysis in a medium temperature reactor.
It is not the purpose of this report to project which technologies will attain commercial status, but
to provide utilities with sufficient information to understand the commercially available
technologies and near-term potentials.
4.5 Cost and Performance Summary
Biomass is unique among renewable resources. Although the capital costs of biomass
technologies can be lower than those of many other renewable technologies, biomass power
technologies must also account for fuel costs that other renewables such as solar, wind, and (for
the most part) geothermal largely avoid.
Except for landfill gas, municipal solid waste (MSW), and a few other special cases in which the
fuel is free or even comes at a negative cost due to disposal credits, the cost of a biomass
feedstock is a significant fraction of the total cost of generating electricity. Also, compared to
coal and natural gas, biomass power plants have relatively high capital costs and low
efficiencies, especially versus natural gas. Hence, the ability to find low-priced, abundant fuel is
critical to the success of biomass power generation.
The capital and O&M cost estimates presented in the following subsections are from the 2010
EPRI report, Engineering and Economic Evaluation of Biomass Power Plants, 100% Biomass
Repowering, Biomass Cofiring, and Bubbling Fluidized Bed Biomass Combustion [43].
4.5.1 100% Biomass Repowering of a Pulverized Coal Boiler
Modification of coal-fired boilers to fire biomass fuels is achieved by removing the lower
furnace section, including the coal burner systems and much of the secondary air systems. This
strategy maintains the existing heat transfer surfaces located in the upper portion of the boiler
and the existing steam cycle. To allow for the combustion of biomass as the sole fuel for the unit,
a grate or fluidized bed system is installed in place of the coal burner systems. Figure 4-41 is a
schematic showing replacement of the bottom of a pulverized-coal boiler by a bubbling fluidized
bed (BFB) combustion system.
Tables 4-25 to 4-30 summarize plant design assumptions, performance, total capital requirement,
fixed and variable operation and maintenance costs, and levelized cost of electricity estimates for
conversion of a 100-MW pulverized coal unit to a bubbling fluidized bed boiler burning woody
biomass.
4-76
10581090
Biomass Energy
Proposed section to be
added to the existing
furnace
Figure 4-41
Example Bubbling Bed Retrofit to a Pulverized Coal Boiler (Source: Metso)
4.5.1.1 Plant Design Assumptions
The biomass repowering performance and cost estimates are based on the assumptions in Table
4-25. The unit is originally a wall-fired pulverized coal boiler rated at 100-MW net capacity.
Repowering the unit to burn 100% biomass reduces the net rated capacity to 60 MW. It is
assumed that the biomass fuel is undried woody biomass containing 45% moisture with dry-basis
higher heating value of 8670 Btu/lb.
Table 4-25
Plant Design Assumptions for Biomass Repowering Base Case (Source: EPRI [48])
Parameter Biomass Repowering Base Case Scenario

Net Capacity (Coal) 100-MW
Original Boiler Type Wall-fired PC
Net Capacity (Biomass) 60-MW
Biomass Boiler Type BFB
Capacity Factor 85%
Biomass Properties
Fuel Type Undried, blended biomass
Heating Value (dry) 8670 Btu/lb
Moisture Content 45%
4-77
10581090
Biomass Energy
Table 4-26
Raw and Dried Biomass Fuel Composition
Raw Wood, Dried Wood,

Fuel Quality Parameter 45% Moisture 30% Moisture
Higher Heating Value,
Btu/lbm 4770 6070
2
Proximate Analysis
Moisture, % 45.00 30.00
Ash, % 2.18 2.77
Volatile Matter, % 45.32 57.68
Fixed Carbon, % 7.49 9.54
Ultimate Analysis2
Carbon, % 27.67 35.22
Hydrogen, % 2.53 3.22
Nitrogen, % 0.57 0.73
Sulfur, % 0.06 0.08
Chlorine, % 0.02 0.03
Moisture, % 45.00 30.00
Ash, % 2.18 2.77
Oxygen, % 21.96 27.96
4.5.1.2 Plant Performance Estimate
Table 4-27 compares the performance of the original coal-fired PC unit and the 100% biomass
repowered unit. Repowering reduces the gross turbine output by about 30% to 68.2 MW and net
output from 100 MW to 60 MW. Boiler efficiency decreases from 88% to 78%, and net plant
heat rate increases from 10,690 to 12,600 Btu/kWh. Even with the lower rating, fuel mass input
with biomass repowering is almost double that of the original coal unit.
4.5.1.3 Total Performance and Cost Estimates
Table 4-28 summarizes the performance, total capital requirement, and fixed- and variable-O&M
and costs for the 100% biomass repowered plant. All costs are in October 2010 dollars and are
based on the 60-MW net output of the repowered plant.
The total capital requirement estimate is $1970/kW. The scope includes major equipment (fuel
handling and preparation and boiler modification), direct and indirect balance-of-plant costs,
interest during construction, and owner’s costs. Fuel handling/preparation and boiler
modification respectively contribute 16% and 47% of the total plant cost before addition of
interest during construction and owner’s costs.
4-78
10581090
Biomass Energy
The fixed and variable O&M cost estimates are respectively $132/kW-yr and $3.5/MWh. Fixed
O&M costs consist of wages and wage-related overheads for the permanent plant staff, routine
equipment maintenance, and other fees. Variable O&M costs include costs associated with
equipment maintenance during outages, catalyst/reagents, chemicals, water, and other
consumables. Fuel costs are determined separately and are not included in either fixed or
variable O&M costs.
Table 4-27
Plant Performance Summary for PC Coal-fired and Repowered Biomass-fired Bubbling
Fluidized Bed Plants (Source: EPRI [43])
Woody
Coal-fired Biomass-Fired
Parameter Units PC Unit BFB Unit
Turbine Gross Output kW 113,000 68,200
Turbine Heat Rate Btu/kWh 8,218 8,635
Total Auxiliary Power kW 13,000 8,200
% 11.5 12.0
Net Plant Output kW 100,000 60,000
Heat to Steam from Boiler MBtu/h 941 589
Boiler Efficiency % 88.0 78.0
Boiler Heat Input MBtu/h 1,069 756
Coal Consumption tons/h 43.8 0
Biomass Consumption tons/h 0 79.3

Net Plant Heat Rate Btu/kWh 10,690 12,600
Notes:
1. Performance is preliminary and for information only. Not to be used for detailed design.
2. Auxiliary power is assumed.
3. Once through cooling is assumed.
4. Average ambient conditions of 62.1° F dry bulb temperature and 54.8° F wet bulb
temperature. Dew point temperature (49.6° F) is assumed to be the cold water temperature.
4-79
10581090
Biomass Energy
Table 4-28
100% Biomass Repowering Performance and Cost Estimates (3Q 2010$, Source: EPRI
[43])
100% Biomass Repowering
Rated Capacity
Plant Size (units x unit size, MW) 1 x 60
Physical Plant
Unit Life, Years 20
Scheduling
Preconst., License & Design Time, Years 2.0
Idealized Plant Construction Time, Years 0.75
Hypothetical In-Service Date1 January 1, 2011
Capital Costs ($1,000) ($/kW) (%)
Major Equipment Costs
Fuel Handling/Preparation $16,540 $280/kW 16%
Boiler Modification $50,000 $830/kW 47%
Total Major Equipment Costs $66,540 $1,110/kW 63%
Direct Balance of Plant Costs
Site Work, Foundations, Roadways $970 $20/kW 1%
Electrical Equipment, Cable, & Raceway $750 $10/kW 1%
Instrumentation & Controls $1,670 $30/kW 2%
Total Direct Balance of Plant Costs $3,390 $60/kW 3%
Indirect Balance of Plant Costs
Facilities, Engineering, and Const. Mgt. $15,390 $260/kW 15%
Project & Process Contingency & Fees $20,680 $340/kW 19%
Total Indirect Balance of Plant Costs $36,070 $600/kW 34%
Total Costs
Total Plant Costs $106,000 $1,770/kW 100%
AFUDC (Interest during construction) $1,200 $20/kW --
Total Plant Investment (incl. AFUDC) $107,200 $1,790/kW --
Owner’s Cost $10,700 $180/kW --
Total Capital Requirement $117,900 $1,970/kW --
O&M Costs
Fixed, $/kW-yr 132
Variable, $/MWh 3.50
Performance/Unit Availability
Net Heat Rate (Full Load), Btu/kWh 12,600
Equivalent Planned Outage Rate, % 4
Duty Cycle Baseload
Minimum Load, % 40
Emission Rates
NOx, lb/MBtu 0.15 - 0.24
SOx, lb/MBtu fuel dependent
Particulate, lb/MBtu <0.02
Technology Development Rating Commercial
Design & Cost Estimate Rating Simplified
Notes:
1. The “Hypothetical In-Service Date” assumes Notice to Proceed with preliminary activities on January 1, 2011
and assumes no overlap of pre-construction and construction activities.
2. All costs are presented in October 2010 dollars.
4-80
10581090
Biomass Energy
4.5.1.4 Levelized Cost of Electricity Estimates
Table 4-29 summarizes the levelized cost of electricity (LCOE) estimates for the 100% biomass
repowered plant for the base case and two sensitivity cases. The levelized costs are in constant
third-quarter 2010 dollars. The base case assumes that the biomass fuel cost $3.55/MBtu. The
two sensitivity cases assume costs of $2.55 and $4.55/MBtu.
The levelized cost of electricity estimates are $91/MWh for the base case (the delivered biomass
fuel vs. coal is $3.55/MBtu) and $78 and $103/MWh for the two sensitivity cases (biomass fuel
cost is $2.55 and plus $4.55/MBtu.
The other major assumptions include 20-year project life, zero inflation, 55%/45% debt/equity
ratio, 4% interest on debt, 8% return on equity, 15-year debt financing term, zero investment tax
credit, five-year MACRS tax depreciation, and 40% income tax rate. If the 30% investment tax
credit (ITC) is applied, the levelized cost decreases by about $11/MWh. It should be noted that it
is uncertain whether repowering projects would qualify for the ITC.
Table 4-29
LCOE Estimates for 100% Biomass Repowering (3Q 2010 $, Source: EPRI [48])
Sensitivity Sensitivity
1 1
Cost Base Case Case 1 Case 2
Biomass Fuel Cost ($/MBtu) 3.55 2.55 4.55
2
Levelized Cost of Electricity With ITC
($/MWh)
Fixed O&M Component of LCOE ($/MWh) 18 18 18
Variable O&M Component of LCOE ($/MWh) 3 3 3
Fuel Component of LCOE ($/MWh) 45 32 57
3
Capital Charge Component of LCOE
($/MWh) 14 14 14
4
Total ($/MWh) 80 67 92
2
Levelized Cost of Electricity Without ITC
($/MWh)
Fixed O&M Component of LCOE ($/MWh) 18 18 18
Variable O&M Component of LCOE ($/MWh) 3 3 3
Fuel Component of LCOE ($/MWh) 45 32 57
3
Capital Charge Component of LCOE
($/MWh) 22 22 22
4
Total ($/MWh) 88 76 101
Notes:
1
Sensitivity Cases assume biomass fuel costs are $1.00/MBtu above and below the base cost of $3.55/MBtu.
2
Estimate of LCOE assumes an 85% capacity factor, 20-year project life, and constant dollars.
3
Levelized fixed charge is 7.33%/yr with the investment tax credit (ITC) and 8.5%/yr without the credit.
4
Sum of LCOE component values may not equal Total value due to rounding.
4-81
10581090
Biomass Energy
4.5.2 Biomass Cofiring with Coal in a Pulverized Coal Boiler
There are two basic approaches to cofiring. The first is to blend the fuels and feed them together
to the coal processing equipment (i.e., crushers or pulverizers). In a cyclone boiler, generally up
to 10% of the coal heat input can be replaced by biomass fuels. The smaller fuel particle size
requirement of a PC boiler generally limits the fuel replacement to perhaps 3%. Another
approach is to develop a separate biomass processing system, in which high cofiring percentages
(10% and greater) in a PC unit can be accomplished, although at somewhat higher cost.
4.5.2 Biomass Cofiring with Coal
The section addresses cofiring in a 250-MW wall-fired pulverized coal boiler. In addition to the
base case coal fired plant without cofiring, three biomass fuels are considered, undried blended
biomass, dried biomass, and torrefied biomass. The design data, performance, capital and O&M
cost, and levelized cost of electricity estimates are presented in the following sections.
4.5.2.1 Plant Design and Fuel Assumptions
The biomass cofiring performance and cost estimates are based on base-case assumptions in
Table 4-30 and the coal and biomass fuel compositions in Table 4-31. The unit is a wall-fired
pulverized coal boiler rated at 250 MW net capacity. The baseline coal is Central Appalachian
coal containing 12,207 Btu/lb. In the base case, it is assumed that the biomass fuel is undried
woody biomass containing 45% moisture with wet-basis higher heating value of 4768 Btu/lb, the
biomass is injected separately into boiler, and the biomass provides 10% of the heat input to the
boiler. In addition, two other biomass fuels are addressed in the performance estimates in the
next section: dried woody biomass containing 30% moisture and 6068 Btu/lb (wet-basis, high
heating value), and torrefied biomass containing 6.2% moisture and 12,759 Btu/lb.
Table 4-30
Summary of Biomass Cofiring Base Case (Source: EPRI [48])
Parameter Biomass Cofiring Base Case Scenario

Cofiring Method Separate Injection
Net Capacity 250 MW
Boiler Type Wall-fired Pulverized Coal
Capacity Factor 80%
Biomass Heat Input 10%
Biomass Properties
Fuel Type Undried, blended biomass
Heating Value (dry) 8,670 Btu/lb
Coal Properties
Type Central Appalachian
Heating Value 12,207 Btu/lb
4-82
10581090
Biomass Energy
Table 4-31
Woody Biomass Fuel Quality: Heat Content, Proximate, and Ultimate Analyses (Source:
EPRI [48])
Fuel Quality Baseline CAPP Raw Wood, Dried Wood, Torrefied

Parameter Coal 45% Moisture 30% Moisture Wood
Higher Heating
12,207 4768 6068 12,759
Value, Btu/lbm
Proximate Analysis*
Moisture, % 8.47 45.00 30.00 6.21
Ash, % 11.24 2.18 2.77 1.78
Volatile Matter, % 29.57 45.32 57.68 19.66
Fixed Carbon, % 50.72 7.50 9.54 72.35
Ultimate Analysis*
Carbon, % 69.96 27.67 35.22 78.24
Hydrogen, % 4.47 2.53 3.22 3.19
Nitrogen, % 1.36 0.57 0.73 0.56
Sulfur, % 1.03 0.06 0.08 0.01
Chlorine, %** 0.15 0.02 0.03 0.00
Moisture, % 8.47 45.00 30.00 6.21
Ash, % 11.24 2.18 2.77 1.78
Oxygen, % 3.47 21.97 27.96 10.01
Notes:
* As-received values.
** Although not in the ASTM ultimate analysis, chlorine is often reported with it.
4.5.2.2 Plant Performance Estimates
Tables 4-32 through 4-34 present detailed plant performance data for the base coal-fired plant
and cofiring coal with the three biomass fuels: undried, dried, and torrefied biomass fuel, each at
5%, 10%, and 15% heat input.
For undried and dried biomass cofiring, boiler efficiency and net plant heat rate decrease slightly
and net plant heat rate increases slightly as the biomass heat input increases from 0% to 15% of
total heat input. The opposite happens for cofiring co-milled coal and torrefied wood, mainly
because the moisture content of the torrefied biomass is lower than that of coal and thus the
moisture losses are lower.
4-83
10581090
Biomass Energy
Table 4-32
Performance Estimates for Cofiring Coal and Undried Biomass (Source: EPRI [48])
Coal +5% Coal +10% Coal +15%

Undried Undried Undried
Coal Biomass Biomass Biomass
Full Load Performance
Gross Power, MW 261.50 261.50 261.50 261.50
Net Power, MW 250.00 248.46 248.52 247.88

Boiler Efficiency, % 87.38 87.10 86.62 86.13
Net Plant Heat Rate, Btu/kWh 9762.5 9892.5 9951.8 10,037.3
Total Heat Input, MBtu/hr 2440.6 2457.9 2473.3 2488.0

Biomass Heat Input, MBtu/hr 0.0 122.9 247.3 373.2
Coal Burn Rate, ton/hr 99.97 95.65 91.18 86.63
Biomass Burn Rate, ton/hr 0.00 12.89 25.94 39.14
Table 4-33
Performance Estimates for Cofiring Coal and Dried Biomass (Source: EPRI [48])
Coal +10% Coal +15%

Coal +5% Dried Dried
Coal Dried Biomass Biomass Biomass
Gross Power, MW 261.50 261.50 261.50 261.50

Net Power, MW 250.00 248.49 248.58 247.97
Net Plant Heat Rate, Btu/kWh 9762.5 9867.7 9901.6 9958.1


4-84
10581090
Biomass Energy
Table 4-34
Performance Estimates for Cofiring Co-Milled Coal and Torrefied Biomass (Source: EPRI [48])
Coal +5% Coal +10%

Torrefied Torrefied Coal +15%
Wood Wood Torrefied Wood
Coal Biomass Biomass Biomass
Gross Power, MW 261.50 261.50 261.50 261.50
Net Power, MW 250.00 248.40 248.48 248.58

Net Plant Heat Rate, Btu/kWh 9762.5 9830.7 9822.8 9811.5

4.5.2.3 Total Capital Requirement and Operation and Maintenance Cost Estimates
Table 4-35 summarizes total capital requirement, and fixed and variable operation and
maintenance cost estimates for the base case, cofiring coal and biomass at 10% heat input from
undried biomass fuel. All costs are in October 2010 dollars and are based on the contribution of
biomass to the rated capacity with cofiring (10% of 250 MW, or 25 MW).
The total capital requirement estimate at 10% heat input from biomass is $1690/kW. Based on
the total 250 MW total rated capacity of the unit, it is $169/kW. The scope includes major
equipment (fuel handling and preparation and boiler modification), direct and indirect balance-
of-plant costs, interest during construction, and owner’s costs. Fuel handling/preparation and
boiler modification respectively contribute 16% and 47% of the total plant cost before addition
of interest during construction and owner’s costs.
The incremental fixed and variable O&M cost estimates are respectively $21.6/kW-yr and
$4.9/MWh, based on the contribution of biomass to the 250-MW rated capacity. Fixed O&M
costs consist of wages and wage-related overheads for the permanent plant staff, routine
equipment outage maintenance, catalyst/reagents, chemicals, water, and other consumables. Fuel
costs are determined separately and are not included in either the fixed or variable O&M costs.
4-85
10581090
Biomass Energy
Table 4-35
Biomass Cofiring Performance and Cost Summary (3Q 2010$, Source: EPRI [48])
Rated Capacity Biomass Cofiring with Coal
Coal plant size (units x unit size), MW 1 x 250-MW
Biomass fuel feed system Separate Injection
Fraction of plant output from biomass, % 10
Equivalent power output, MW 25
Physical Plant
Unit life, years 20
Scheduling
Preconst., License & Design Time, Years 1.5
Hypothetical In-Service Date1 January 1, 2011
Incremental Total Capital Requirement ($ 1,000) ($/kW)3 (%)
Major Equipment Costs
Fuel Handling/Prep $23,120 $930/kW 61%
Boiler Modification $1,610 $60/kW 4%
Total Major Equipment Costs $24,730 $990/kW 65%
Site Work, Foundations, Roadways $1,040 $40/kW 3%
Electrical Equipment, Cable, & Raceway $340 $10/kW 1%
Instrumentation & Controls $1,220 $50/kW 3%
Total Direct Balance of Plant Costs $2,600 $100/kW 7%
Facilities, Engineering, and Const. Mgt. $4,390 $180/kW 12%
Project & Process Contingency & Fees $6,510 $260/kW 17%
Total Indirect Balance of Plant Costs $10,900 $440/kW 29%
Total Costs
Total Plant Costs $38,230 $1,530/kW 100%
AFUDC (Interest during construction) $250 $10/kW --
Total Plant Investment (incl. AFUDC) $38,480 $1,540/kW --
Owner’s Cost $3,850 $150 --
Total Capital Requirement $42,330 $1,690 --
Incremental O&M Costs (cofiring systems only)
Fixed, $/kW-yr 21.60
Net Plant Heat Rate, Btu/kWh (HHV) 9,760
Duty Cycle Baseload
Minimum Load, % 40
Emission Rates
NOx, lb/MBtu 0.05
SOx, lb/MBtu 0.06
Particulates, lb/MBtu 0.006
4-86
10581090
Biomass Energy
Notes:
3. Costs presented on $/kW basis are determined based on biomass-fired capacity (i.e., equivalent power output
of 25 MW).
Table 4-36 summarizes the levelized cost of electricity estimates for cofiring 10% heat input
from undried biomass and 90% from coal for the base case and two sensitivity cases. The base
case assumes that the biomass fuel cost premium relative to coal at $3.30/MBtu is plus
$0.25/MBtu. The two sensitivity cases assume it is minus $0.75 and plus $1.25/MBtu.
Table 4-36
Levelized Cost of Electricity Estimate for Cofiring Coal and 10% Undried Biomass (3Q 2010
$, Source: EPRI [48])
Sensitivity Sensitivity
1 1
Cost Component Base Case Case 1 Case 2
Incremental Biomass Fuel Cost ($/MBtu) +0.25 -0.75 +1.25
2
Levelized Cost of Electricity With ITC ($/MWh)
Fixed O&M 3 3 3
Variable O&M 5 5 5
Fuel Component of LCOE 2 (7) 12
3
Capital Charge 12 12 12
4
Total 23 13 33
2
Levelized Cost of Electricity Without ITC
($/MWh)
Fixed O&M 3 3 3
Variable O&M 5 5 5
Fuel Component of LCOE 2 (7) 12
3
4
Total 31 21 41
Notes:
1
Sensitivity Cases assume biomass fuel costs different than the base case, at $0.75/MBtu below and
$1.25/MBtu above the $3.30/MBtu cost assumed for coal.
2
3
4
The incremental levelized cost of electricity estimates are $34/MWh for the base case
(incremental cost of delivered biomass fuel vs. coal is plus $0.25/MBtu) and $24 and $44/MWh
for the two sensitivity cases (incremental costs are minus $0.75 and plus $1.25/MBtu). If the
4-87
10581090
Biomass Energy
30% investment tax credit (ITC) is applied, the levelized cost decreases by about $10/MWh. It
should be noted that it is uncertain whether cofiring projects would qualify for the ITC.
The levelized costs are in constant third-quarter 2010 dollars. The other major assumptions
include 20-year project life, zero inflation, 55%/45% debt/equity ratio, 4% interest on debt, 8%
return on equity, 15-year debt financing term, zero investment tax credit, five-year MACRS tax
depreciation, and 40% income tax rate.
4.5.3 Biomass-Fired Bubbling Fluidized Bed Combustion Boiler Power Plants
Tables 4-37 to 4-40 summarize plant design assumptions, performance, total capital requirement,
three cases:
• 50-MW bubbling fluid bed boiler fired by undried woody biomass
• 100-MW bubbling fluid bed boiler fired by undried woody biomass
• 50-MW bubbling fluidized bed boiler fired by a mixture of 80% woody biomass and 20%
switchgrass based on heat input.
The three biomass bubbling bed fluidized bed combustion boiler performance and cost estimate
cases are based on the assumptions in Table 4-37. It is assumed that the biomass fuels are
undried woody biomass containing 45% moisture and 8670 Btu/lb (dry HHV) and as-received
switchgrass containing 15% moisture and 7304 Btu/lb.
Table 4-37
Design Assumptions for Bubbling BFB Boiler Plants
50-MW BFB 100-MW BFB 50-MW BFB

Woody Woody Switchgrass /
Parameter Biomass Biomass Woody Biomass
Net Capacity 50-MW 100-MW 50-MW
Boiler Type BFB BFB BFB
Capacity Factor 85% 85% 85%
20%Switchgrass/
Biomass Heat Input 100% 100% 80% Woody Biomass
Biomass Properties
Woody biomass
Undried blended Undried blended blended with as-
Fuel Type biomass biomass received switchgrass
7,304 Btu/lb
Higher Heating Value (dry) 8,670 Btu/lb 8,670 Btu/lb Switchgrass
Moisture Content 45% 45% 15% Switchgrass
4-88
10581090
Biomass Energy
Table 4-38 compares the performance estimates for the BFB boiler cases. The boiler efficiency is
75% for all three cases. Net plant heat rates are respectively 13,964, 12,894, and 14,119
Btu/kWh for the 50- and 100-MW BFB plants fired by woody biomass and the 50-MW plant
fired by a 80% biomass/20% switchgrass mixture. The principle reason that the 100-MW BFB
plant has a lower heat rate is that the steam cycle is more efficient than that of the 50-MW BFB
plant due to higher steam conditions. In addition, the parasitic power consumption as a fraction
of gross output is lower.
Table 4-38
Plant Performance Estimates for BFB Boiler Plants (Source: EPRI [48])

Woody Woody Switchgrass/
Parameter Biomass Biomass Woody Biomass
Turbine Gross Output, kW 58,824 113,636 59,524
Turbine Heat Rate, Btu/kWh 8,902 8,505 8,895
Total Auxiliary Power, kW 8,824 13,636 9,524
Total Auxiliary Power, % 15 12 16

Net Plant Output, kW 50,000 100,000 50,000
Boiler Efficiency, % 75 75 75
Boiler Heat Input, MBtu/hr 698.2 1289.4 705.9
Biomass Consumption, ton/hr 73.2 135.2 69.8
Net Plant Heat Rate, Btu/kWh 13,964 12,894 14,119
cost estimates for the bubbling fluidized bed boiler cases. All costs are in October 2010 dollars
and are based on net output.
The total capital requirement estimates are respectively $5,588, $4,053, and $5,806/kW for the
50- and 100-MW BFB plants fired by woody biomass and the 50-MW plant fired by an 80%
biomass/20% switchgrass mixture. The scope includes major equipment (fuel handling and
preparation and boiler modification), direct and indirect balance-of-plant costs, interest during
construction, and owner’s costs.
4-89
10581090
Biomass Energy
Table 4-39
Performance and Cost Estimates for Bubbling Fluidized Bed Boiler Plants (3Q 2010$, Source:
EPRI [48])

Net Capacity and Boiler Type Woody Woody Switchgrass /
Biomass Biomass Wood
Location Southeast Southeast Southeast
Rated Capacity
Rated Capacity (units x unit size), MW 1 x 50 MW 1 x 100 MW 1 x 50 MW
Biomass fuel feed system Mechanical Mechanical Mechanical
80%/20%
Fraction of plant output from biomass, % 100% 100%
Wood/Switchgrass
Physical Plant
Unit life, years 30 30 30
Scheduling
Preconst., License & Design Time, Years 1 1.5 1
Idealized Plant Construction Time, Years 2.5 3 2.5
Hypothetical In-Service Date Jan-11 Jan-11 Jan-11
Total Capital Requirement, $/kW
Major Equipment Cost
Steam Generator System $1,350 $1,050 $1,350
Turbine Island System $370 $300 $370
Biomass Handling $350 $270 $480
Environmental Controls $110 $70 $110
Total Major Equipment Cost $2,180 $1,690 $2,310
Direct Balance of Plant Cost
BOP Facilities $840 $550 $840
General Facilities & Site Specific $380 $260 $420
Total Direct Balance of Plant Cost $1,220 $810 $1,260
Engineering Fee & Construction
$1,330 $890 $1,340
Management
Process Contingency $30 $30 $30
Project Contingency $170 $120 $180
Total Indirect Balance of Plant Cost $1,530 $1,040 $1,550
Total Cost
Total Plant Cost $4,930 $3,530 $5,120
AFUDC (Interest during construction) $160 $157 $168
Total Plant Investment (including
$5,090 $3,684 $5,278
AFUDC)
Total Owner’s Cost, $/kW $510 $369 $528
4-90
10581090
Biomass Energy
Table 4-39 (Continued)

Performance and Cost Estimates for Bubbling Fluidized Bed Boiler Cases (3Q 2010 $,
Source: EPRI [48])
50-MW BFB
Operation and Maintenance Cost 50-MW BFB 100-MW BFB Wood/
Woody Biomass Woody Biomass Switchgrass
Fixed, $/kWbiomass-yr 113 63 114
Variable, $/MWhbiomass 5.8 5.1 5.85
Net Plant Heat Rate, Btu/kWh (HHV) 13,964 12,894 14,119
Equivalent Planned Outage Rate, % 4 4 4
Duty Cycle Baseload Baseload Baseload
Minimum Load, % 40 40 40
Emission Rates
NOx, lb/MBtu 0.10 – 0.12 0.1 0.10 – 0.12
SOx, lb/MBtu 0.04 – 0.08 0.01 0.04 – 0.08
Particulates, total, lb/MBtu 0.015 – 0.035 0.018 0.015 – 0.035
Technology Development Rating Commercial Commercial Developing
Notes:
The fixed O&M costs are respectively $113, $63, and $114/kW-yr and the variable O&M costs
are respectively $5.8, $5.1, and $5.9/MWh for the 50- and 100-MW BFB plants fired by woody
biomass and the 50-MW plant fired by an 80% biomass/20% switchgrass mixture. Fixed O&M
costs consist of wages and wage-related overheads for the permanent plant staff, routine
variable O&M costs.
Table 4-40 summarizes the levelized cost of electricity estimates for the 50- and 100-MW BFB
plants fired by undried woody biomass and the 50-MW BFB plant fired by an 80% wood/20%
switchgrass mixture. The costs are levelized over the 30-year project life and are in constant
October 2010 dollars.
The levelized costs are respectively $126, $87, and $148/MWh for the 50- and 100-MW BFB
plants fired by woody biomass and the 50-MW plant fired by an 80% biomass/20% switchgrass
mixture. The major assumptions include 30-year project life, zero inflation, 55%/45%
4-91
10581090
Biomass Energy
debt/equity ratio, 4% interest on debt, 8% return on equity, 15-year debt financing term, zero
investment tax credit, five-year MACRS tax depreciation, and 40% income tax rate.
If the 30% investment tax credit (ITC) is applied, the levelized cost decreases $104, $71, and
$125/MWh for the 50- and 100-MW woody biomass fired plants and the 50-MW plant fired by
the wood/switchgrass mixture. Relative to the repowering and cofiring cases, it is more likely
that stand-alone biomass-fired BFB projects would be eligible for the ITC, provided that the
applying entities (i.e., project owners) are eligible and the project commences operation prior to
the expiration of the ITC.
Table 4-40
Levelized Cost of Electricity Estimates for Biomass-Fired Bubbling Fluidized Bed Plants
(3Q 2010 $, Source: EPRI [48])

Woody Woody Wood/
Cost Component Biomass Biomass Switchgrass
Biomass Fuel Cost, $/MBtu 3.55 3.55 3.55
Switchgrass Fuel Cost, $/MBtu - - 5.05
2
Levelized Cost of Electricity With ITC ($/MWh)
Fixed O&M 15 8 15
Variable O&M 6 5 6
Fuel 50 46 54
3
4
Total 104 84 110
2
Levelized Cost of Electricity Without ITC ($/MWh)
Fixed O&M 15 8 15
Variable O&M 6 5 6
Fuel 50 46 54
4
3
Total 132 104 139
Notes:
1
Sensitivity Cases assume biomass and switchgrass fuel costs are each $1/MBtu above and below the base
cost s of $3.55/MBtu and $5.05/MBtu.
2
3
4
4-92
10581090
Biomass Energy
4.5.4 Biomass-Fired Stoker Boiler Power Plants
Tables 4-41 to 4-44 present the plant design assumptions, performance, total capital requirement,
the 50-MW stoker boiler plant fired by undried woody biomass. The performance and cost data
were generated by adjusting the 50-MW BFB boiler plant data in Table 4-39 based on data for
stoker and fluidized bed boiler plants in the 2007 EPRI report [42].
The biomass stoker boiler power plant performance and cost estimates are based on the
assumptions in Table 4-41. It is assumed that the biomass fuel is undried woody biomass
containing 45% moisture and 8670 Btu/lb (dry HHV).
Table 4-42 presents the performance estimate for the stoker boiler power plant. The boiler
efficiency is 75% and the estimated net plant hear rate is 14,191 Btu/kWh.
Table 4-41
Design Assumptions for Stoker Boiler Plant
50-MW Stoker Plant

Parameter Woody Biomass
Net Capacity 50-MW
Boiler Type Stoker
Capacity Factor 85%
Biomass Heat Input 100%
Biomass Properties
Fuel Type Undried blended biomass

Higher Heating Value (dry) 8,670 Btu/lb
4-93
10581090
Biomass Energy
Table 4-42
Plant Performance Estimates for Stoker Boiler Plant (Source: EPRI [42] [43])
50-MW Stoker Plant

Parameter Woody Biomass
Turbine Gross Output, kW 58,824
Turbine Heat Rate, Btu/kWh 8,902
Total Auxiliary Power, kW 8,824
Total Auxiliary Power, % 15
Net Plant Output, kW 50,000
Boiler Efficiency, % 72.8

Boiler Heat Input, MBtu/hr 709.6
Biomass Consumption, ton/hr 74.4
Net Plant Heat Rate, Btu/kWh 14,191
cost estimates for the stoker boiler plant. All costs are in October 2010 dollars and are based on
the net output.
The total capital requirement estimate is $5188/kW. The scope includes major equipment (fuel
handling and preparation and boiler modification), direct and indirect balance-of-plant costs,
interest during construction, and owner’s costs.
4-94
10581090
Biomass Energy
Table 4-43
Performance and Cost Estimates for Stoker Boiler Plant (3Q 2010$, Source: EPRI [42] [43])
50-MW Stoker Plant

Net Capacity and Boiler Type
Woody Biomass
Location Southeast
Rated Capacity
Rated Capacity (units x unit size), MW 1 x 50 MW
Biomass fuel feed system Mechanical
Fraction of plant output from biomass, % 100%
Physical Plant
Unit life, years 30
Scheduling
Preconst., License & Design Time, Years 1
Hypothetical In-Service Date Jan-11
Total Capital Requirement, $/kW
Major Equipment Cost
Steam Generator System $1,209
Turbine Island System $370
Biomass Handling $350
Environmental Controls $110
Total Major Equipment Cost $2,039
Direct Balance of Plant Cost
BOP Facilities $840
General Facilities & Site Specific $380
Total Direct Balance of Plant Cost $1,220
Engineering Fee & Construction
$1,330
Management
Process Contingency $29
Project Contingency $165
Total Indirect Balance of Plant Cost $1,524
Total Cost
Total Plant Cost $4,783
AFUDC (Interest during construction) $155
Total Plant Investment (including AFUDC) $4,938
Total Owner’s Cost, $/kW $247
Total Capital Requirement $5,185
4-95
10581090
Biomass Energy
Table 4-43 (Continued)

Performance and Cost Estimates for Stoker Boiler Plant (3rd-Quarter 2010$, Source: EPRI
[42] [43])
50-MW Stoker Plant

Operation and Maintenance Cost
Woody Biomass
Fixed, $/kWbiomass-yr 110
Variable, $/MWhbiomass 5.1
Net Plant Heat Rate, Btu/kWh (HHV) 14,191
Duty Cycle Baseload
Minimum Load, % 40
Emission Rates
NOx, lb/MBtu 0.10 – 0.12
SOx, lb/MBtu 0.04 – 0.08
Particulates, total, lb/MBtu 0.015 – 0.035
Notes:
The fixed and variable O&M cost estimates are respectively $132/kW-yr, and $3.5/MWh. Fixed
O&M costs consist of wages and wage-related overheads for the permanent plant staff, routine
variable O&M costs.
4.5.3.4 Levelized Cost of Electricity Estimate
Table 4-44 summarizes the levelized cost of electricity estimate for the stoker boiler plant. The
levelized cost is $122/MWh in constant third-quarter 2010 dollars. The major assumptions
include 30-year project life, zero inflation, 55%/45% debt/equity ratio, 4% interest on debt, 8%
return on equity, 15-year debt financing term, zero investment tax credit, five-year MACRS tax
depreciation, and 40% income tax rate.
If the 30% investment tax credit (ITC) is applied, the levelized cost decreases by about
$20/MWh. Relative to the repowering and cofiring cases, it is more likely that standalone
biomass-fired stoker projects would be eligible for the ITC, provided that the applying entities
(i.e., project owners) are eligible and the project commences operation prior to the expiration of
the ITC.
4-96
10581090
Biomass Energy
Table 4-44
Levelized Cost of Electricity Estimates for 50-MW Stoker Boiler Plant Fired by Woody
Biomass (3rd-Quarter 2010 $, Source: EPRI [47] [48])
Cost Component Base Case

Biomass Fuel Cost, $/MBtu 3.55
Levelized Cost of Electricity1 With ITC ($/MWh)

Fixed O&M 15
Variable O&M 5
Fuel Component of LCOE ($/MWh) 50
2
Capital Charge 32
Total3 102
Levelized Cost of Electricity1 Without ITC ($/MWh)

Fixed O&M 15
Variable O&M 5
Fuel Component of LCOE ($/MWh) 50
2
Capital Charge 52
Total3 122
1
Estimate of LCOE assumes an 85% capacity factor, 30-year project life, and
constant dollars.
2
Levelized fixed charge is 6.4%/yr with the investment tax credit (ITC) and
7.33%/yr without the credit.
3
4.5.5 Landfill Gas Recovery and Power Generation
Table 4-45 presents performance and cost data for a 5-MW generating unit fired by landfill gas.
The plant scope includes the landfill gas collection field, gas cleanup facility to remove sulfur
and other compounds, a flare, and five 1-MW internal combustion engines. The data are from a
2007 EPRI report [47].
According to the U.S. Environmental Protection Agency’s Landfill Municipal Outreach Project
[26], solid waste landfills are the largest source of human-related methane emissions in the
United States, accounting for about 25% of the total emissions in 2004. The landfill gas
composition is typically 50% methane, 50% carbon dioxide, and small amounts of other organic
compounds. Landfill gas is extracted from landfills using a series of wells and a blower/flare (or
vacuum) system. This system directs the collected gas to a central point where it can be
processed and treated depending upon the ultimate use for the gas. From this point, the gas can
be simply flared or used to generate electricity, replace fossil fuels in industrial and
manufacturing operations, fuel greenhouse operations, or be upgraded to pipeline-quality gas.
4-97
10581090
Biomass Energy
Table 4-45
Performance and Cost Estimates for Landfill Gas Combustion
Rated Capacity 5-MW Reciprocating Engine Plant
Plant Size (no. of units x unit size, MW) 5 x 1.0
Physical Plant
Unit Life, Years 20
Land Required —
Scheduling
Available for Commercial Orders, Year —
Hypothetical In-Service Year January 1, 2010
Preconst., License & Design Time, Years 1
Idealized Plant Construction Time, Years 1
Capital Costs, $/kW
Month/Year Dollars December, 2009
Gas Conditioning and Compressor 204
Power conversion equipment 969
General Facilities and Engineering Fee 163
Project and Process Contingency 200
AFUDC (Interest during construction) 0
Total Plant Investment (inc. AFUDC) 1,536
Owner Costs, $/kW
Tax and Fees 0
Total Capital requirement, Hypothetical In-
Service year (includes AFUDC) 1,843
O&M Costs
Fixed, $/kW-yr 56
Net Heat Rate, Btu/kWh, Full Load 13,500
Equivalent Unplanned Outage Rate, % 11
Equivalent Availability, % 85
Duty Cycle Baseload
Minimum Load, % 50
Emission Rates
CO2, lb/MMBtu 227
NOX, lb/MMBtu 0.2
SOX, lb/MMBtu 0.1
Mercury, lb/MMBtu —
4-98
10581090
Biomass Energy
4.5.6 Biomass Technologies Grubb Curve
Figure 4-42 illustrates the Grubb curve, which communicates the maturity of different biomass
technologies. The curve can be used to understand the cost required to complete research either
from concept to finished product or just the cost required to complete a particular stage of
technology development. While early-stage development does not require as much capital as
development at middle stages, technologies in the last stages of development may require
significant capital to purchase major components and conduct full-scale tests over many months
or years.
Technical risk can also be assessed using the Grubb curve. In the first development stages little is
known about the core technology with regard to the size, details, and costs of its final form. As a
result, the technical risk is at its highest in the early stages of development. As the technology
matures the technical risk lessens but the overall risk to development projects can remain high
when considering the likelihood of failure.
Figure 4-42
The Grubb Curve for Biomass Technologies
4.6 Biomass to Electricity RD&D Initiatives for the Future
Substantial industrial participation is essential in all but basic research initiatives if biomass
energy is to effectively contribute to GHG reduction policy objectives. Particular attention needs
to be given to issues relevant to fuel supply, conventional combustion, and gasification/pyrolysis.
4-99
10581090
Biomass Energy
Primary fuel supply issues that need to be addressed include:

• Demonstration of innovative equipment for forestry residue systems;
• Demonstration of machines for establishing, harvesting, transporting, and storing energy crop
systems;
• Basic and applied R&D on grasses and alternative energy crops for areas not suitable for
wood crops;
• Continuous breeding and plant improvement programs for energy crops to improve yields;
• Schemes which can accelerate development of the sustainable biomass fuel supply, such as
standards for certifying sustainable biomass feedstock.
Primary conventional combustion issues that need to be addressed include:

• Demonstration of incremental innovations in technology to improve efficiency, reduce
capital costs, or reduce emissions;
• Demonstration of technical innovations (e.g. biomass pre-treatment processes, Residue
Derived Fuels) that will extend the range of fuels than can be burned;
• Demonstration of small scale CHP with innovative cycles.
Primary gasification/pyrolysis issues that need to be addressed include:

• Demonstrations covering the range of gasification and pyrolysis technologies appropriate for
worldwide regions;
• Applied RD&D of small-scale CHP using advanced systems;
• Demonstration of pyrolysis systems when the current applied R&D shows sufficient
progress.
4.7 U.S. Technical Tax Code Modifications to promote further use of

Biomass
Biomass has the potential to generate up to 230 billion kWh of electricity in the United States by
2030. Current U.S. federal tax laws are designed to encourage greater use of biomass to generate
electricity via production tax credits (PTCs), investment tax credits (ITCs), and Treasury grants.
In addition, accelerated depreciation deductions are available to biomass generating facilities.
However, a number of obstacles in U.S. tax law prevent greater biomass development. Several of
these obstacles are identified below.
• Cofiring: Under IRS Code, an open-loop biomass facility that cofires biomass with fossil
fuels does not qualify for fiscal incentives.
• Modification of Existing Facilities: Facilities that have been converted from fossil fuel units
to open-loop or closed-loop biomass units or cofiring plants do not currently qualify for the
PTC.
4-100
10581090
Biomass Energy
• Definition of Waste: The current definition of open-loop biomass does not assure facility
owners that they can pay for the open loop biomass used as fuel and still qualify for the PTC.
• Biomass to Create Gas or Thermal Energy: The PTC does not currently apply to facilities
that create biomass-derived renewable energy gas and thermal energy.
4.8 Conclusions
Biomass fuels are many and varied, with distinct characteristics. While there are institutional
and political questions concerning the classification of wood waste and other waste products as
biomass or renewable, technically they are biomass resources that are derived from living plants.
Most biomass fuels permit utilities to generate dispatchable renewable power. The fuels can
be extracted and/or stored and then used to meet electricity demand. Utilization systems can
capitalize upon the characteristics of these fuels. They are modest in heating value, highly
reactive, low in nitrogen and sulfur, and of varying ash characteristics. Of the biomass fuels
available, woody biomass is the most commonly used material, but agricultural wastes and
dedicated energy crops appear to be the biomass fuels of the future.
Numerous technologies are available for biomass fuel utilization, including both cofiring options
and stand-alone options. Cofiring provides a means for incorporating biomass utilization into
electricity generating facilities with the lowest cost and the least risk. Cofiring can be used to
enhance combustion and, in most cases, reduce SO2 and NOX emissions. The lack of sulfur in
biomass fuels, coupled with low nitrogen concentrations, reactive nitrogen, and reactive fuel as a
whole, provides the basis for this enhanced combustion.
The forest products and pulp and paper industries, as well as some electric utilities and
independent power producers, have built stand-alone plants. These systems are not inexpensive.
However, they provide a basis for using biomass to generate electricity. They can be
economically justified depending upon localized economics and their use in addressing customer
needs. Institutional arrangements that provide partnerships between industries and utilities can be
particularly useful in such circumstances.
Biomass pre-treatment may open a global biomass trade similar to that of coal. But key to the
expansion of biomass for electricity production will be available tax incentives, RPS mandates,
and GHG-reduction directives that externalize the costs of using fossil fuels. For example,
torrefied biomass may contribute to offsetting fossil-fuel emissions, but consideration of carbon
credits for the char field application will be required to expand its worldwide use.
In summary, biomass is dispatchable renewable power. It can come in solid or gaseous form, and
can be exploited using a wide variety of technologies to supply electricity to internal customers
or to the grid. Moreover, its carbon neutral or negative characteristics position it to potentially
contribute to the target of de-carbonizing electricity production.
4-101
10581090
Biomass Energy
4.9 References
1. Walker, Joseph. 1966. Hopewell Village: The Dynamics of a Nineteenth Century Iron-
Making Community. Philadelphia. University of Pennsylvania Press.
2. Szego, G.C. and C.C. Kemp. 1973. Energy Forests and Fuel Plantations. CHEMTECH. May.
3. Bethel, J. et al. 1976. The Potential of Lignocellulosic Material for the Production of
Chemicals, Fuels, and Energy. National Academy of Sciences. Washington, D.C.
4. Henry, J.F. 1979. The Silvicultural Energy Farm in Perspective. in Progress in Biomass
Conversion, Vol 1. (Sarkanen, K.V. and D.A. Tillman, eds.) Academic Press, New York.
pp. 215 – 256.
5. Walsh, M.E. et al. 2000. Biomass Feedstock Availability in the United States: 1999 State
Level Analysis. Oak Ridge National Laboratory, Oak Ridge, TN.
6. SCS Engineers. 2002. Economic and Financial Aspects of Landfill Gas to Energy Project
Development in California. California Energy Commission. Sacramento, CA. 500-02-020F.
7. Johnson, D.K., D.A. Tillman, B.G. Miller, S.V. Pisupati, and D.J. Clifford. 2001.
Characterizing Biomass Fuels for Co-firing Applications. Proc. 2001 Joint International
Combustion Symposium. American Flame Research Committee. Kuai, Hawaii. Sep 9 – 12.
8. Tillman, D.A. Final Report: EPRI-USDOE Cooperative Agreement: Co-firing Biomass with
Coal. Contract No. DE-FC22-96PC96252. EPRI, Palo Alto, CA: 2001.
9. Jenkins, B.M., L.L. Baxter, T.R. Miles, Jr., and T.R. Miles. 1996. Combustion Properties of
Biomass. Proc. Biomass Usage for Utility and Industrial Power. An Engineering Foundation
Conference. Snowbird, UT. April 28 – May 3.
10. Miles, T.R., et al. 1993. Alkali Slagging Problems with Biomass Fuels. Proc. First Biomass
Conference of the Americas. Aug 30 – Sep 2. Burlington, VT. pp. 406-421.
11. Tillman, D.A. Biomass Cofiring: Field Test Results. EPRI, Palo Alto, CA: 1999. TR-113903.
12. Payette, K., T. Banfield, T. Nutter, and D. Tillman. 2002. Emissions Management at Albright
Generating Station through Biomass Cofiring. Proc. 27th International Technical Conference
on Coal Utilization and Fuel Systems. Clearwater Florida, March 4–7.
13. Hus, P.J. and D.A. Tillman. 2000. Cofiring multiple opportunity fuels with coal at Bailly
Generating Station. Biomass and Bioenergy. 19(6):385-394.
14. Tillman, D.A. 1994. Trace Metals in Combustion Systems. Academic Press, San Diego, CA.
15. Tillman, D.A. Opportunity Fuel Cofiring at Allegheny Energy: Final Report. EPRI, Palo
Alto, CA: 2004. Report 1004811.
16. Campbell, A.G. 1990. Recycling and Disposal of Wood Ash. TAPPI. Sep. pp. 141–145.
17. Prinzing, D.E. EPRI Alternative Fuels Database. EPRI, Palo Alto, CA:1996. TR-107602.
18. Johnson, D.K., et al. 2002. Report to Foster Wheeler on Pyrolysis of Urban Wood Waste and
Fresh Switchgrass. Private communication. Pennsylvania State University, University Park,
PA.
4-102
10581090
Biomass Energy
19. Tillman, D.A. 2002a. Petroleum Coke as a Supplementary Fuel for Cyclone Boilers. Proc.
27th International Technical Conference on Coal Utilization and Fuel Systems. Clearwater
Florida, March 4 – 7.
20. Shafizadeh, F., and W. DeGroot. 1976. Combustion Characteristics of Cellulosic Fuels, in
Thermal Uses and Properties of Carbohydrates and Lignins (Shafizadeh, F., K. Sarkanen, and
D. Tillman, eds.). Academic Press, New York. pp. 1-18.
21. Shafizadeh, F. and W. DeGroot. 1977. Thermal Analysis of Forest Fuels, in Fuels and
Energy from Renewable Resources (Tillman, D., K. Sarkanen, and L. Anderson, eds).
Academic Press, New York. pp. 93 – 114.
22. Broido, A. 1976. Kinetics of Solid-Phase Cellulose Pyrolysis. in Thermal Uses and
Properties of Carbohydrates and Lignins (Shafizadeh, F., K. Sarkanen, and D. Tillman, eds.).
Academic Press, New York. pp. 19-36.
23. Baxter, L.L. et al. 1996a. The Behavior of Inorganic Material in Biomass-Fired Power
Boilers: Field and Laboratory Experiences. Proc. Biomass Usage for Utility and Industrial
Power. Engineering Foundation Conference. Snowbird, UT. April 28-May 3.
24. Baxter, L.L. et al. 1996b. The Behavior of Inorganic Material in Biomass-Fired Power
Boilers—Field and Laboratory Experiences: Vol II of Alkali Deposits Found in Biomass
Power Plants. SAND96-8225 Volume 2 and NREL/TP-433-8142.
25. Miller, S.F., B.G. Miller, and D. Tillman. 2002. The Propensity of Liquid Phases Forming
During coal-Opportunity Fuel (Biomass) Cofiring as a Function of Ash Chemistry and
Temperature. Proc. 27th International Technical Conference on Coal Utilization and Fuel
Systems. Clearwater Florida, March 4 – 7.
26. Landfill Methane Outreach Program, U.S. EPA:
http://www.epa.gov/lmop/overview.htm#methane.
27. Miller, B.G., S.F. Miller, C. Jawdy, R. Cooper, D. Donovan, and J. Battista. 2000. Feasibility
Analysis for Installing a Circulating Fluidized Bed Boiler for Cofiring Multiple Biofuels and
Other Wastes with Coal at Penn State University. Second Quarterly Technical Progress
Report. Work Performed Under Grant No. DE-FG26-00NT40809.
28. Miller, S.F. and B. G. Miller. 2002. The Occurrence of Inorganic Elements in Various
Biofuels and its Effect on the Formation of Melt Phases During Combustion. Proc.
International Joint Power Generation Conference. Phoenix, AZ. June 24 – 27.
(IJPGC2002-26177).
29. Bryers, R.W. 1993. Analysis of a Suite of Biomass Samples. Foster Wheeler Development
Corporation, Livingston, NJ. Report FWC/FWDC/TR-94/03.
30. Bush, P. V. et al. 2001. Evaluation of Switchgrass as a Cofiring Fuel in the Southeast.
USDOE Cooperative Agreement No. DE-FC36-98GO10349. Southern Research Institute,
Birmingham, AL.
31. Harding, N.S. 2002. Cofiring Tire-Derived Fuel with Coal. Proc. 27th International
Technical Conference on Coal Utilization and Fuel Systems. Clearwater Florida,
March 4 – 7.
32. Tillman, D.A. 2002b. Cofiring Technology Review. Report to USDOE – NETL. Pittsburgh,
PA.
4-103
10581090
Biomass Energy
33. Kitto, B. and D. Green. 2003. Electricity Generation Using Digester Gas at Clean Water
Services. Proc. 28th International Technical Conference on Coal Utilization and Fuel
Systems. Clearwater Florida, March 5–8.
34. Green, A. 2003. Opening Remarks. International Conference on Co-Utilization of Domestic
Fuels. Gainesville, FL. Feb 5–6.
35. Tillman, D.A., A.J. Rossi, and W.D. Kitto. 1981. Wood Combustion: Principles, Processes,
and Economics. Academic Press. New York.
36. Thimsen, D, Lindgren, G. Assessment of Biogas-Fueled Electric Power Systems. EPRI, Palo
Alto, CA: 2004. 1009450.
37. Haq, Z. 2002. Biomass for Electricity Generation. U.S. DOE Energy Information
Administration.
38. Perlack, R.D., et al. Biomass Fuel from Woody Crops for Electric Power Generation. Oak
Ridge National Laboratory, Oak Ridge, TN: 1995. ORNL-6871.
39. Campbell, J.E., Lobell, D.B., et al. Greater Transportation Energy and GHG Offsets from
Bioelectricity than Ethanol. Science. 2009.
40. Georgia Pacific. 2009. Processed Engineered Fuel-Effective Waste to Energy Solution.
41. A. Maciejewska. 2006. Cofiring of Biomass with Coal: Constraints and Role of Biomass Pre-
treatment. European Commission EUR 22462 EN.
42. Engineering and Economic Evaluation of Renewable Energy Technology, EPRI, Palo Alto, CA:
2007. 1012726.
43. Engineering and Economic Evaluation of Biomass Power Plants, 100% Biomass
Repowering, Biomass Cofiring, and Bubbling Fluidized Bed Biomass Combustion. EPRI,
Palo Alto, CA: 2010. 1019762.
4-104
10581090
5
SOLAR PHOTOVOLTAICS
Installed Capacity • 21,500 MW worldwide

(Jan. 1, 2010 est.)[32]
• 2000 MW in United States
Technology Readiness • Flat-plate crystalline and some thin-film types are commercial.
• Concentrator modules and most thin-film products are in pilot-scale

production.
• Research (emerging materials and manufacturing processes).
• Moderate-high confidence in cost estimates and projections.
Environmental Impact • Manufacturing: Manageable waste streams.
• Installation: Minimal; potential visual and land-use issues.
• Operation: Minimal.
• Decommissioning: Material recycling and disposal needs.
Economic Status • Currently competitive only in high-value applications; cost (without

subsidy) remains several times that of grid power in most areas.
• As cost drops, the number of potential markets and uses will increase.
• Tax and cash incentives still necessary to facilitate most projects.
Policy Status • Encouraged by RPS and energy policies in most jurisdictions.
Trends to Watch • Short-term supply constraints from silicon shortage, though less
problematic than predicted as global output of PV cells rose more than
55% in 2007 and growth rates are forecasted to continue high, ranging
from 30%-50% per year, over the next five years. Emerging from a
brief period of supply constraints due to silicon shortage, PV supply
has caught up with demand. At the same time, the global economic
downturn has reduced demand and caused module prices to drop,
effectively returning PV to its historical trend of declining costs and
putting pressure on emerging thin-film technologies.
• Emerging dominance of Asian production and module export has

contributed to global downward pricing pressure.
• Emergence of thin-film technology.
• More building-integrated PV (BIPV) and energy-efficiency bundling.
• Continued growth as a distributed resource, including microgrids.
• Increasingly automated manufacturing leading to lower cost.
5-1
10581090
Solar Photovoltaics
• Large corporations, some with expertise in different but related

businesses such as electronics or chemicals, are angling for strategic
position and investing heavily in PV as the market has grown.
Examples include GE, Dow Chemical, Sharp, Kyocera, and Norsk
Hydro. At the same time, some corporations such as BP and Shell that
were once significant players in PV have retreated.
5.1 Introduction
Solar photovoltaic (PV) modules are solid-state semiconductor devices that convert sunlight into
direct-current electricity. PV was initially developed for the space program in the 1950s to power
the first satellites, an application for which simplicity and reliability were tremendously
important and cost was of much lesser consequence. In the early 1970s, PV began to be used for
terrestrial power applications, which led to a shift in R&D goals and funding sources away from
the space program to private/public sector collaborative efforts. This work, led primarily by the
United States, Europe, and Japan, has resulted in a generally steady, dramatic decrease in module
costs and a concurrent near-doubling in energy conversion efficiency. The average module sales
price (in 2009 dollars) dropped from $77,000/kW in 1976 to $2,500/kW in 2009. Over the same
period, the performance of high-end commercial modules has increased from 8% conversion
efficiency to about 20%. A spike in PV prices driven by a shortage of high-purity silicon
feedstock in 2004–2008 has been resolved. As a result, module prices dropped by about 40%
between 2008 and 2009 and inventory levels in early 2009 were high. One consequence of the
short-term silicon shortage was to accelerate growth in the thin-film PV sector, whose market
share increased from 11% in 2007 to 14% in 2008 to 17% in 2009 despite a softening market.
Notably, thin-film manufacturer First Solar produced up to 1100 MW in 2009, making it the top
PV producer in the world.
As of the end of 2009, there were approximately 21,500 MW (21.5 GW) of PV capacity installed
worldwide, with about 2000 MW deployed in the United States. In 2009, annual global PV
module shipments exceeded 7900 MW [29], a 44% increase over 2008, which translates into
billions of dollars per year of business for both module producers and system installers. Recent
mergers and acquisitions, which are consolidating PV manufacturing companies and system
integrators into large multinational energy companies, may mark PV’s emergence into the
mainstream. With growing environmental concerns and the emerging green-power market, PV
appears to be entering a new era in which it will play an increasingly important role in meeting
the world’s energy needs. Industry analysts believe PV shipments will substantially exceed
10,000 MW (10 GW) in 2010, with usually conservative Navigant Consulting estimating as high
as 15,600 MW (15.6 GW) [41].
5-2
10581090
Solar Photovoltaics
18,000
16,000
15612.9
14,000
12,000
Annual Sales, MW peak
10,000
7913.3
8,000
6,000
5491.8
4,000
3073
2,000
1049.8
504.9
42.7 71.5 114.1
0
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
Figure 5-1
Worldwide PV Industry Growth and Forecast for 2010 (Source: Navigant Consulting) [41]
The growth of PV applications since 1992 has occurred in three major application groups:
remote off-grid commercial PV, grid-connected PV, and indoor consumer applications such as
watches, calculators, and other household electronics. Data compiled by Navigant Consulting
and others through 2009 [27] clearly indicate tremendous growth in grid-connected PV since the
mid-1990s. This trend has continued such that the grid-connected market segment accounted for
more than 95% of PV installed in 2009, up from just 8% in 1991, as well as 85% to 90% of total
installed capacity worldwide [31]. Despite a market downturn and given policy considerations,
grid-connected PV systems for commercial, industrial, and residential uses appear well
positioned for long-term growth.
According to Navigant Consulting’s 2010 PV forecast [29], if conservative market and policy
assumptions apply, future annual output will continue to rise following historical trends to about
17.6 GW per year by 2014. If more accelerated growth conditions apply, as the organization
believes most likely in regions of strong demand such as Europe, annual PV production could
increase to nearly 45 GW, more than five times its 2009 levels. A scenario in which reduced
incentives predominate and result in lower annual PV production is considered less likely.
5-3
10581090
Solar Photovoltaics
5.2 Cost and Economic Issues

This edition of the RETG continues to reflect an emphasis on relatively small-scale distributed
PV as opposed to utility-scale centralized PV that a utility might, for example, build near a
substation to provide grid support or in a rural field to connect directly to the transmission
system. One significant project of the latter type is the 25-MW DeSoto Next-Generation Solar
Energy Center commissioned by FPL in October 2009 to provide peaking power, whose cost
only became competitive when PV prices declined significantly in 2009 (Figure 5-2). EPRI
believes that such options will prove increasingly practical as the cost of PV falls and becomes
competitive generally with other intermediate and peaking supply technologies. The 2008
extension of the federal 30% Investment Tax Credit to utilities seems to have accelerated this
trend.
Figure 5-2
PV Field of 25-MW DeSoto Next-Generation Solar Energy Center (Source: FPL)
Though the fundamental PV technology may be the same, cost and economic considerations
related to distributed PV are significantly different from those related to centralized utility-
owned PV. In the former case, some or all of the cost may be born by customers interested in
benefiting from on-site distributed generation and supporting a clean, renewable technology.
Depending on local electricity prices and any rebates, subsidies, or incentives available, the
payback period for a typical residential-scale PV system may be as short as five years or could
well exceed a decade, depending on variables such as technology selected, site latitude and
climate, and system location and orientation. A distributed PV system may also be owned by an
energy company interested in providing grid support in critical areas, exploiting new business
opportunities, or complying with policy demands such as renewable portfolio standards. Because
PV is highly modular, distribution companies can precisely deploy it in optimal locations and
capacities or work with customer-owners to do so. Other purchase or lease arrangements may be
beneficial to customers, utilities, or both.
The U.S. government, acting through the DOE, has instituted several programs through the years
to support PV research and development, subsidize its use, and plan for its future. The federal
“Million Solar Roofs” initiative, begun under the Clinton administration, aimed to install one
5-4
10581090
Solar Photovoltaics
million solar energy systems—including PV, water heating, and space heating—on the rooftops
of American homes and businesses by 2010. The initiative included federal procurement
programs, technology grants, and lending programs. Briefly renamed “Solar Powers America,”
the Million Solar Roofs program effectively concluded in 2006 after investing $16 million to
help install approximately 370,000 solar projects comprising 200 MW of grid-connected PV and
another 200 MW of solar water heating capacity. A final report on the program was issued in
October 2006 [33]. Since then, states such as California and Delaware have established their own
Million Solar Roofs programs and goals. In July 2010, Sen. Bernie Sanders (Ind.-Vermont)
introduced the “Ten Million Solar Roofs Act of 2010” (S. 3460) to support similar efforts
through 2021.
In 2006, the Bush administration launched the DOE Solar America Initiative (SAI), whose goals
were to accelerate PV deployments and conduct RD&D targeted to bring PV costs to “grid
parity” by 2015. The DOE PV budget was roughly doubled as a result of the SAI program and
many new projects were launched, consisting primarily of industry-led short-term efforts to more
rapidly scale up current PV technologies. They also included public-outreach educational
projects, a “technology incubator” program to help startup companies move from laboratory-
scale pilot lines into early commercial-scale production, and a small amount of more exploratory
research into novel concepts. The SAI concluded in 2009, but the activities developed during its
tenure were incorporated into the portfolio of DOE’s Solar Energy Technologies Program.
Funding provided through the American Recovery and Reinvestment Act (ARRA) of 2009
greatly expanded DOE’s budget for solar energy research, development, and demonstration. In
addition, the DOE Loan Guarantee Program was established to support “qualified projects in the
belief that accelerated commercial use of these new or improved technologies will help to sustain
economic growth, yield environmental benefits, and produce a more stable and secure energy
supply.” In March 2009, DOE awarded a $535 million loan guarantee to PV developer Solyndra
to advance its cylindrical PV panel design and build a 500-MW-per-year module factory in
Fremont, California. In October 2009, DOE announced that it would draw on ARRA funds to
provide guarantees for up to $8 billion in loans for conventional renewable energy projects such
as solar, wind, biomass, geothermal and hydropower. At the same time, DOE pledged to
streamline the selection standards and lending process. Dozens of PV companies are currently in
the queue for DOE support.
Several states also have programs encouraging the development of renewable energy. In many
cases, a combination of state and federal incentive, rebate, or loan programs can pay a significant
share of a PV system’s cost. Most notably, beginning in 1998 California’s Emerging Renewables
Program (ERP) and Self-Generation Incentive Program (SGIP) led to an estimated installed grid-
connected PV capacity totaling one-quarter of the nation’s installed PV in 2005. As of January 1,
2010, SGIP had helped fund more than 890 PV projects representing nearly 136 MW of
generating capacity [34]. California’s PV momentum was further boosted in 2006 by the passage
of SB-1 (Senate Bill 1) and the formation of the California Solar Initiative (CSI), a $3 billion
program to incentivize the installation of one million solar rooftops by 2017. Through mid-
September 2010, the CSI had helped fund nearly 37,000 installed PV projects with a cumulative
capacity of 390 MW (Figure 5-3) [38].
5-5
10581090
Solar Photovoltaics
Figure 5-3
California Solar Initiative Installed Projects by Month [38]
As of the end of 2009, California led the United States in total installed PV capacity with 768
MW, followed by New Jersey with 128 MW, Colorado with 59 MW, Arizona with 46 MW,
Florida with 39 MW, and Nevada with 36 MW [39]. In work funded by DOE, the Interstate
Renewable Energy Council (IREC) concluded that residential installed capacity more than
doubled between 2008 and 2009, and represented 36% of all new grid-connected PV capacity.
Despite the quick expansion of residential PV over the past few years, non-residential
commercial projects continue to dominate the U.S. grid-connected PV market.
Figure 5-4
Annual U.S. Installed Grid-Connected PV Capacity by Sector [39]
5-6
10581090
Solar Photovoltaics
In addition, solar requirements embedded in renewable portfolio standards are also fueling PV
development. The National Database of State Incentives for Renewable Energy (DSIRE) at
www.dsireusa.org provides up-to-date information on state financial and regulatory incentives
that are designed to promote renewable energy technologies.
Other nations, such as Germany, Spain and Japan, have had active and generous government-
sponsored and -mandated incentive programs. As a result of its aggressive programs, Germany
leads the world in both annual grid-connected PV installations, with 3800 MW added in 2009
and a total installed capacity of approximately 9 to 10 GW, nearly half the world’s total [31, 36].
Spain is second in the world in installed capacity with 3.4 GW despite a large drop-off in new
installations in 2009 when incentives were greatly reduced; for example, Spain added 2430 MW
of PV in 2008 but only 70 MW in 2009. Third in the world in total installed capacity is Japan
with an estimated 2.2 to 2.6 GW, followed by the United States with about 2 GW. Italy notably
launched itself onto the PV scene with favorable incentives that doubled its new installations
from 338 MW in 2008 to about 700 MW in 2009 [32, 40].
5.2.1 Japan
Japan had approximately 320 MW of PV in place at the end of 2000, with more than two-thirds
of it installed through a residential capital cost buy-down program that began in 1994 and was
administered by the New Energy Foundation, part of the Ministry of Economy, Trade, and
Industry (METI) until the program’s termination in 2005 [15]. In the following five years,
another 1.1 GW was installed, nearly all under the METI program, launched in the late 1990s
and run by METI’s New Energy and Industrial Technology Development Organization (NEDO).
The purpose of the METI program is to develop a full range of new energy technologies,
including PV, by subsidizing both R&D and market penetration to bring costs down.
METI PV systems are grid-connected and net-metered, which increases their attractiveness given
Japan’s high-priced electricity. The maximum subsidy provided to system buyers was initially
about 50% of installed costs. It has decreased annually, reaching 10% in 2003, and was phased
out completely in 2005. Despite the decline and eventual termination of end-user incentives, the
program remained popular and successful to its end, in part because net prices in Japan have
continued to fall, aided by ongoing Japanese government market-development assistance to the
industry.
After three stagnant years, new incentives that took effect in January 2009 paid ¥70,000/kW
(approximately $820/kW U.S.) for residential PV systems and led to Japan’s installing more PV
than ever in 2009, approximately 484 MW [32, 40]. In November 2009, METI’s “New Purchase
System for Solar Power Generated Electricity” established a new feed-in tariff of ¥48/kWh
($0.56/kWh U.S.) for electricity generated by residential PV smaller than 10 kW and ¥24/kWh
($0.28/kWh U.S.) for non-residential PV, which is expected to further Japan’s recently refined
goals to install 28 GW of PV by 2020 and 56 GW by 2030 [40].
5-7
10581090
Solar Photovoltaics
5.2.2 Germany
Germany had approximately 110 MW of PV installed as of the end of 2000. Following a “1000
Solar Roofs” program in 1989, Germany initiated a “100,000 Solar Roofs” program in 1999 with
the goal of installing 300 MW of PV by 2004. The program provides 10-year low-interest loans
that feature no money down and no interest payment for two years toward the installation of
rooftop PV systems. Over the term of the program that ended in June 2003, loans were approved
for approximately 350 MW.
Germany’s Renewable Energy Sources Act, which took effect in April 2000, also encouraged
PV development by offering high financial incentives. The act established a PV feed-tariff which
initially offered solar power producers roughly €0.5062 per kWh ($0.66 at April 2006 exchange
rates, about $0.75 in late 2007, $0.69 in 2008), guaranteed for 20 years. Since 2002, this
guaranteed price has been reduced by 5% to 6.5% per year, depending on the type of system, to
encourage cost reductions. New regulations that took effect in January 2004 provide a tiered
incentive structure ranging from €0.469 to €0.632/kWh of PV generation, depending upon the
system size and whether it is free-standing, rooftop, or building integrated. In addition to national
incentives, some individual German municipalities offer production incentives funded by
surcharges on utility bills. All told, these programs have helped catapult the country’s PV
capacity from 76 MW in 2000 to nearly 8900 MW in 2009 [36], with about 3000 to 3800 MW of
that total installed in 2009 alone [29]. Much of the year-end rush was due to a steep reduction in
Germany’s feed-in tariff (Figure 5-5).
Figure 5-5
German PV installations between January 2009 and March 2010 showing surge of projects,
particularly larger than 10 kW, at year’s end in anticipation of declining feed-in tariff.
(Source: Solarenergie-Förderverein Deutschland e.V.)
5-8
10581090
Solar Photovoltaics
In January 2010, the feed-in tariffs were €0.4301/kWh ($0.56/kWh) for a typical 30-kW rooftop
installation, €0.4092/kWh ($0.53/kWh) for a 30- to 100-kW rooftop system, and €0.3194/kWh
($0.42/kWh) for ground installations. Tariff reductions in 2010 are being debated and negotiated
by legislators, and are expected to result in cuts of up to 16% in October 2010 and another 10%
to 12% in January 2011 [30]. Discussions reflect the worldwide decline in PV prices in 2009,
which put pressure on the German government to decrease incentives at a similar pace. Critics
argue that a large contributor to PV price declines is competition from China, and that cutting
incentives too steeply will harm the German PV industry to the benefit of China. The German
PV market appears robust in 2010 despite the declining feed-in tariff and political uncertainty.
5.2.3 Example PV Project Costs
As solar photovoltaic technology continues to evolve, significant cost reductions are expected to
continue as a result of improving power conversion efficiencies, development of low-cost cell
fabrication processes, increasing cell production volume with attendant economies of scale, and
reduced balance-of-plant costs through standardized modular power blocks.
5.2.4 Technology Performance and Cost Tables
The following subsections address the performance and cost of PV technologies using a two-
track approach. Distributed and utility-scale PV systems demand very different economic and
technical considerations. In particular, for distributed PV systems, the costs are typically covered
by a building owner or operator with significant support in the form of rebates or tax credits in
most cases. These systems require a more flexible and tailored approach. The information and
examples that follow are intended to illustrate some of the variables involved and an approach to
estimating system performance and cost through a range of options.
5.2.4.1 Distributed PV Performance and Cost Estimates
Many customers considering installation of a distributed PV system are relatively sophisticated

energy consumers. In many cases, they have investigated their energy consumption patterns,
taken measures to improve energy efficiency and reduce usage, and developed a working
knowledge of PV operating principles, utility interconnection practices, net metering tariffs, and
financial incentive programs for solar electric power.
Nonetheless, due to the complexity (and novelty) of PV production models and financial
calculations, their investment decisions are often made without a complete understanding of the
true economic effects. To accurately evaluate the costs and benefits of a possible PV project, the
consumer must:
• Obtain detailed cost and performance data from the PV system or module manufacturer
• Acquire meteorological data from a representative ground station
• Simulate the production of the PV system, taking into account sun angles and module
orientation
• Determine hourly impacts on utility consumption
5-9
10581090
Solar Photovoltaics
• Calculate monthly energy and/or demand savings using their specific utility tariff structure
• Calculate the financial impacts and applicability of various buy-down programs, tax credits,
and financing terms.
In recent years, the rooftop PV market has evolved so that many capable installers take
responsibility for such concerns on behalf of their clients, who need do little more than sign a
contract and write a check. Such customer service has minimized the effort that might have
deterred potential PV owners in the past and helped increase sales. Because the factors involved
vary so widely, it is impractical to attempt to construct a single table representative of all
distributed PV systems. One consumer tool that has been designed to overcome these difficulties
is the “Clean Power Estimator.” Section 5.2.6.3 provides sample calculations using the
Estimator.
5.2.4.2 Utility-Scale PV Plant Performance and Cost Estimates
The technology performance and cost tables for utility-scale central-station PV power plants are
based on the performance and cost estimates presented in the May 2010 EPRI report,
Engineering and Economic Evaluation of Central-Station Solar Photovoltaic Power Plants [35].
Tables 5-1 through 5-7 summarize the design and site assumptions and performance, capital and
operation and maintenance costs, and levelized cost of electricity estimates for conceptual 10-
MW central-station PV power plants for 22 combinations of six PV technologies and four sites
in the United States. The assumptions and results are addressed in the following sections.
5.2.4.2.1 Design and Site Assumptions
Table 5-1 describes the six PV technologies. They are fixed-tilt flat plate modules using
crystalline silicon (c-Si), thin-film amorphous silicon (a-Si) and thin-film cadmium telluride
(CdTe) modules; single-axis and tilted single-axis tracking modules using c-Si cells; and two-
axis tracking, multijunction concentrating PV modules (CPV). The system designs are based on
current best practices for North American utility-scale photovoltaic plants.
Table 5-2 provides information on the four PV project sites: Las Vegas, Nevada; Alamosa,
Colorado; Jacksonville, Florida; and Columbus, Ohio. They were chosen to be representative of
a range of site conditions in different regions of the United States. Region 1 is a location with
high direct normal insolation (DNI), little cloud cover and high ambient temperatures, typical of
the arid zones lower than 5000 feet in the desert Southwest (Arizona, California, Nevada, Texas
and New Mexico). Region 2 is a location with high DNI, little cloud cover and low ambient
temperatures, typical of Rocky Mountain areas with locations above 5000 feet in Arizona,
Colorado, Utah and New Mexico. Region 3 is a location with high global horizontal insolation
(GHI), dense cloud cover, high humidity and high ambient temperature. This region is
considered typical of southeast-coastal Texas, the southern States and Florida (excluding the
southern most coastal areas). Region 4 is a location with low GHI, cold winters with variable
skies and hot humid summers with frequent dense cloud cover. This region is considered typical
of New England and the Northern and Midwestern states.
5-10
10581090
Solar Photovoltaics
Table 5-1
PV Technology Summary
PV Technology Abbreviation Description
Fixed-Tilt Thin Film – Amorphous silicon modules mounted at a fixed

a-Si
Amorphous Silicon 30 degree tilt facing south
Fixed-Tilt Thin Film – Cadmium telluride modules mounted at a fixed
CdTe
Cadmium Telluride 30 degree tilt facing south
Single-Axis Tracking – Monocrystalline modules on a north-south axis tracker
SAT c-Si
Crystalline Silicon with backtracking
Tilted-Axis Tracking – Monocrystalline modules on a north-south axis tracker
TAT c-Si
Crystalline Silicon tilted south at 20 degrees with backtracking
Fixed-Tilt Polycrystalline Polycrystalline modules mounted at a fixed 30 degree
c-Si
Silicon tilt facing south
Concentrating
CPV Two-axis tracked high concentration photovoltaics
Photovoltaic
Table 5-2
Photovoltaic Site Assumptions
Ambient
Region # Description Site Irradiance1 Temperature Cloud Cover
Las Vegas,
1 Desert Southwest High DNI High Low
Nevada
Alamosa,
2 Rocky Mountain High DNI Low Low
Colorado
South-Central Jacksonville,
3 High GHI High and humid High
U.S. Florida
North-Eastern and Columbus, High in summer, High in summer,

4 Low GHI
Mid-Western U.S. Ohio Low in winter low in winter
1
DNI is direct normal insolation, GHI is global horizontal insolation
Coastal areas of the Pacific Northwest are considered less attractive for utility-scale PV as they
present low insolation and dense cloud cover for a significant part of the year.
The terrain conditions of each project site are assumed to be either flat or a location with a
topography and cover that does not incur a significant cost to clear and grade flat in preparation
for the construction of the PV power plant.
5-11
10581090
Solar Photovoltaics
5.2.4.2.2 Plant Performance Estimates
Table 5-3 summarizes the estimated first-year power generation and annual capacity factors of
each of the 22 conceptual PV power plants. The annual capacity factor estimates represent the
first year of operation, after which they decline gradually as the PV modules age. They range
from 14.4% for the fixed-flat plate CdTe system in Columbus, Ohio to 27.9% for the
concentrating PV system in Alamosa, Colorado. The assumed degradation rate is 0.75% per
year.
Table 5-3
First-Year Electricity Generation and Capacity Factors
First Year AC Capacity
Technology Location MWh Factor
Las Vegas, NV 19,760 22.6%
Alamosa, CO 16,770 19.1%
Fixed a-Si
Jacksonville, FL 14,910 17.0%
Columbus, OH 12,810 14.6%
Las Vegas, NV 19,450 22.2%
Alamosa, CO 18,150 20.7%
Fixed CdTe
Las Vegas, NV 21,940 25.0%
Single Axis Tracking Alamosa, CO 22,200 25.3%

Crystalline Jacksonville, FL 16,150 18.4%
Las Vegas, NV 23,590 26.9%
Tilted Single Axis Alamosa, CO 23,470 26.8%

Tracking Crystalline Jacksonville, FL 17,160 19.6%
Las Vegas, NV 19,640 22.4%

Alamosa, CO 18,490 21.1%
Fixed Crystalline
Concentrating Las Vegas, NV 24,294 27.7%

Solar PV Alamosa, CO 24,416 27.9%
5-12
10581090
Solar Photovoltaics
5.2.4.2.3 Total Capital Requirement
Table 5-4 summarizes the Total Capital Requirement estimates for the conceptual solar PV
plants. The reference date of the cost estimates is 4th quarter 2009.
The Total Capital Requirement consists of the Total Plant Cost, escalation and interest during
construction, and owner’s costs. The Total Plant Cost includes PV equipment procurement and
transportation, balance-of-plant direct labor and material, and indirect and EPC construction
costs. The Total Plant Investment is the sum of Total Plant Cost and escalation and interest
during construction (also known as allowance for funds used during construction, or AFUDC).
The Total Capital Requirement is the sum of Total Plant Investment and owner’s costs (due
diligence, permitting, legal, development, land, taxes and fees). Larger projects generally benefit
from economies of scale and this is reflected in the cost estimates. The assumed land cost is
$10,000 per acre.
The Total Capital Requirements of the six PV technologies are assumed to be the same at the
four locations. As shown in Table 5-4, they range between $3630/kW (net ac) for the fixed-flat
plate CdTe systems and $4,942/kW for tilted one-axis tracking systems.
5.2.4.2.4 Operation and Maintenance Cost
Operation and Maintenance (O&M) cost consists of scheduled maintenance and cleaning,
unscheduled maintenance, inverter replacement reserve, and insurance, property taxes and
owner’s costs. As shown in Table 5-5, the O&M costs range from $47/kW-yr for the fixed-flat
plate c-Si system to $65/kW-yr for the two-axis tracking system. Scheduled maintenance and
module cleaning are the largest components of the O&M cost.
5.2.4.2.5 Levelized Cost of Electricity
The levelized cost of electricity (LCOE) is a measure of the life-cycle cost of electricity, and
provides a consistent basis for comparing the economics of the projects evaluated across all of
the technologies and locations. The LCOE is calculated using a model that sums the present
values of the annual O&M and capital-related costs and amortizes the result to yield the levelized
cost. The capital component of LCOE is a function of the Total Capital Requirement and the
economic assumptions.
Table 5-6 summarizes the LCOE estimates for each of the technologies and locations, and both
with and without the 30% Federal Investment Tax Credit (ITC). The 30% ITC expires at the end
of 2016 while the parallel non-expiring 10% Investment Tax Credit for solar facilities will
continue beyond 2016. With the 30% ITC, the LCOE varies between $160/MWh for the Fixed
Flat-Plate CdTe/Las Vegas, Nevada case and $277/MWh for the Tilted One-Axis Tracking
c-Si/Columbus, Ohio case. Without the tax credit, the LCOE varies between $241/MWh and
$421/MWh for the same technologies and locations.
All costs are in 4th quarter 2009 U.S. dollars, and the current-dollar levelized costs assume 20-
year plant life, 55/45 debt-equity ratio, 6.5%/yr interest on debt, 11%/yr return on equity, 3%/yr,
inflation and escalation of operation and maintenance costs, five-year MACRS tax depreciation,
40% income tax rate, 30% Investment Tax Credit, and no Production Tax Credit.
5-13
10581090
Solar Photovoltaics
Table 5-4
Utility-Scale Solar Photovoltaic Power Plant Total Capital Requirement Estimates (4th Quarter 2009 $)
2-Axis
Fixed-Flat Fixed-Flat Fixed-Flat 1-Axis Tilted 1- Tracking
Plate a-Si Plate CdTe Plate c-Si Tracking AxisTracking Conc. PV
PV Technology $/kW $/kW $/kW C-Si $/kW C-Si $/kW $/kW

Site Prep $319 $286 $192 $269 $285
Inverter + Installation $332 $332 $332 $332 $332
DC Wiring $155 $157 $123 $118 $118
Substation + MV System $144 $141 $139 $144 $144
DAS/SCADA $15 $14 $13 $34 $34
Mounting System $310 $410 $219 $625 $907
Total Direct Balance of Plant Costs $1,276 $1,339 $1,019 $1,521 $1,820
Engineering + Management $160 $160 $160 $160 $160
Contingency + Profit $174 $169 $185 $209 $232
Total Indirect Balance of Plant Costs $334 $329 $345 $369 $392
Modules $1,876 $1,720 $2,333 $2,281 $2,427 $2,400
Total Plant Cost $3,486 $3,389 $3,696 $4,171 $4,639 $4,516
Total Owner's Cost $254 $241 $233 $275 $303 $301
5-14
10581090
Solar Photovoltaics
Table 5-5
Utility-Scale Solar Photovoltaic Power Plant Operation and Maintenance Cost Estimates (4th Quarter 2009 $)
2-Axis
Plate a-Si Plate CdTe Plate c-Si Tracking C- AxisTracking Conc. PV
PV Technology $/kW-yr $/kW-yr $/kW-yr Si $/kW-yr C-Si $/kW-yr $/kW-yr
Scheduled Maintenance and Cleaning $25 $25 $20 $30 $30 $35
Unscheduled Maintenance $2 $2 $2 $5 $5 $5
Inverter Replacement Reserve $10 $10 $10 $10 $10 $10
Subtotal O&M $37 $37 $32 $45 $45 $50
Insurance, Property Taxes, Owner's Costs $15 $15 $15 $15 $15 $15
Total O&M $52 $52 $47 $60 $60 $65
5-15
10581090
Solar Photovoltaics
Table 5-6
Levelized Cost of Electricity With and Without 30% Investment Tax Credit (4th Quarter 2009 $)
LCOE With 30% ITC LCOE Without 30% ITC

Technology Location ($/MWh) ($/MWh)
Las Vegas, NV $161 $243
Alamosa, CO $191 $287
Fixed a-Si
Jacksonville, FL $214 $322
Columbus, OH $249 $375

Alamosa, CO $172 $258
Fixed CdTe
Jacksonville, FL $210 $316
Single-Axis Alamosa, CO $170 $257
Tracking
Crystalline Jacksonville, FL $234 $353
Tilted Single- Alamosa, CO $175 $266
Axis Tracking

Fixed Alamosa, CO $176 $268

Concentrating Las Vegas, NV $168 $254

Solar PV Alamosa, CO $167 $252
5-16
10581090
Solar Photovoltaics
A probabilistic analysis of the levelized cost of energy is performed to generate a cumulative

probability distribution of the LCOE for each combination of PV technology and site. The first
step in the process is to perform a sensitivity analysis to identify the parameters that have the
most impact on the LCOE. The five highest sensitivity parameters are the capacity factor, capital
cost, return on equity, interest on debt, and O&M cost. Then, a probability distribution is
developed for each of the selected parameters based on the range of uncertainty of the parameter.
Finally, a 10,000-iteration Monte Carlo simulation is performed using the Crystal Ball risk
analysis software and the LCOE model to generate the cumulative probability as a function of
the LCOE. Figure 5-6 is an example cumulative probability distribution for the fixed flat-plate,
crystalline-silicon cell/Las Vegas, Nevada case. The chart shows the minimum, maximum and
25th, 50th, and 75th percentile values of the LCOE.
Figure 5-6
LCOE Probabilistic Analysis: Fixed a-Si Modules, Las Vegas, NV (4th Quarter 2009 $)
5.2.4.2.6 Performance and Cost Summary
Table 5-7 summarizes the performance, capital, and O&M cost estimates for the six PV
technologies.
5-17
10581090
Solar Photovoltaics
Table 5-7
Utility-Scale Solar Photovoltaic Power Plant Performance and Cost Estimate Summary (4th Quarter 2009 $)
2-Axis
Plate a-Si Plate CdTe Plate c-Si Tracking c- AxisTracking Conc. PV
PV Technology $/kW-yr $/kW-yr $/kW-yr Si $/kW-yr c-Si $/kW-yr $/kW-yr
Plant Size (no of units x unit size, MW) 10 x 1 MW 10 x 1 MW 10 x 1 MW 10 x 1 MW 10 x 1 MW 10 x 1 MW

Available for Commercial Orders, Year 2009 2009 2009 2009 2009 2009
First Commercial Service, Year 2010 2010 2010 2010 2010 2010
Hypothetical In-Service Year Jan-09 Jan-09 Jan-09 Jan-09 Jan-09 Jan-09
Direct Balance of Plant Cost $1,276 $1,339 $1,019 $1,521 $1,820 $2,116
Indirect Balance of Plant Cost $334 $329 $345 $369 $392
PV Modules $1,876 $1,720 $2,333 $2,281 $2,427 $2,400
Total Plant Cost $3,486 $3,389 $3,696 $4,171 $4,639 $4,516

Owner's Cost $254 $241 $233 $275 $303 $301
Operation and Maintenance Costs
Costs for Hypothetical In-Service Year
Fixed, $/kW-yr $52 $52 $47 $60 $60 $65
Power Generation
First-Year Annual Capacity Factor
Las Vegas, NV 22.6% 22.2% 22.4% 25.0% 26.9% 27.7%
Alamosa, CO 19.1% 20.7% 21.1% 25.3% 26.8% 27.9%
Jacksonville, FL 17.0% 16.9% 16.9% 18.4% 19.6%
Columbus, OH 14.6% 14.6% 14.4% 15.8% 16.9%
5-18
10581090
Solar Photovoltaics

Utility-Scale Solar Photovoltaic Power Plant Performance and Cost Estimate Summary
2-Axis
Plate a-Si Plate CdTe Plate c-Si Tracking c- AxisTracking Conc. PV
PV Technology $/kW-yr $/kW-yr $/kW-yr Si $/kW-yr c-Si $/kW-yr $/kW-yr
Unit Availability
Equivalent Planned Outage Rate 3.8% 3.8% 3.8% 3.8% 3.8% 3.8%
Equivalent Unplanned Outage Rate, % 1.0% 1.0% 1.0% 2.0% 2.0% 3.0%
Equivalent Availability, % 95% 95% 95% 94% 94% 93%
Preconst., License & Design Time, Months 8 8 8 8 8 8
Idealized Plant Construction Time, Months 6 6 6 6 6 6
Unit Life, Years 20 20 20 20 20 20
Technology Development Rating Commercial Commercial Commercial Commercial Commercial Commercial
Design & Cost Estimate Rating Simplified Simplified Simplified Simplified Simplified Simplified
5-19
10581090
Solar Photovoltaics
5.2.5 Performance and Cost History and Projections
Figure 5-7 shows the evolution of worldwide PV module average selling price from 1976
through 2009 in constant year-2009 dollars [20, 22]. The figure presents these data in the form of
an “experience curve,” where the average sales price is plotted versus cumulative capacity of
modules sold. This format was chosen to show the clear power-law behavior of historical module
prices, where the average selling price has declined by about 20% with each doubling of
capacity. Note that the right end of the figure’s abscissa, at 1000 GW, is about equal to the total
installed U.S. electric generation capacity today.
In the 33-year timeframe of the historic data in Figure 5-7, the total market size has grown more
than 10,000-fold while prices have fallen more than 95% and typical system efficiency has
roughly doubled. In contrast to the four orders of magnitude of historical progress, looking ahead
to achieving 1,000 GW of installed capacity entails less than 500-fold new growth.
Straightforward extrapolation of this trend for the next 30 years, while by no means guaranteed
to be an accurate forecast, implies another efficiency doubling and order-of-magnitude cost
reduction while reaching terawatts of PV deployments. For specific projections of future costs
and performance, see Tables 5-15 to 5-17.
PV Power-Module Global Average Sales Price

$100,000
Year 2009 Dollars per Peak Kilowatt
Average Sales Price
$10,000
$1,000
$100
0.0001 0.001 0.01 0.1 1 10 100 1000
Cumulative Sales, GWP
ASP data source: P. Mints, Navigant Consulting PV Service Practice T.M. Peterson 09/23/10
Figure 5-7
Worldwide Average PV Module Selling Price vs. Cumulative Sales
5-20
10581090
Solar Photovoltaics
5.2.6 PV Performance and Cost Estimate Model
The procedure and data tables that follow provide solar resource estimates and cost and
performance information for both distributed and central-station solar PV power generation.
5.2.6.1 Resource Estimates
Location is a fundamental consideration when estimating PV performance and cost. Table 5-8
presents site data, as described below, for 27 representative locations in the continental United
States.
2
The global latitude-tilt insolation rates (kWh/m /yr) represent the annual solar energy that
reaches an equator-facing solar array oriented at the latitude tilt angle relative to horizontal; i.e.,
the angle equals the latitude coordinate of the site.
The global one-axis tracking insolation rates are 20% higher than the latitude-tilt rates, to
include the additional energy collected as the solar panel rotates around a single axis to follow
the sun. The axis is oriented in the north-south direction at the latitude tilt angle relative to
horizontal.
The direct-normal insolation rate excludes diffuse solar energy and includes only the direct
solar energy that can be collected by a concentrating lens and solar module that tracks the sun
via a two-axis tracking system.
Other site conditions include: dry-bulb temperature, which affects the PV cell operating
temperature and therefore conversion efficiency; maximum wind speed, which affects design
of the structural components; and the city cost factor, which affects capital and O&M costs.
5.2.6.2 Distributed PV Cost and Performance Estimates
As described in Section 5.2.4.1, the Clean Power Estimator evaluates energy investments
for a variety of technologies, including PV, wind, solar thermal, energy efficiency, and fuel
cells. Although the discussion that follows focuses on the Estimator, EPRI is not endorsing this
particular tool over similar products that serve the same purpose. However, it provides a useful
illustration of the methods and variables involved.
The Estimator takes into account the specific characteristics of the customer purchasing the
system (such as the utility rate structure, local solar and environmental conditions, and local,
regional, and federal financial incentives) to provide a directly relevant “personalized” analysis.
5-21
10581090
Solar Photovoltaics
Table 5-8
Site Data for 27 Locations in the Continental U.S.
Solar Insolation Average Ambient Temperature

Global
Global Direct Maximum Dry City
1-Axis Wet Bulb City
Location Latitude Tilt
Tracking
Normal Wind Speed Bulb
(C)
Cost
Index
2 2
(kWh/m /yr) (kWh/m /yr) (mph) (C) Factor
(kWh/m2/yr)
El Paso, TX 2434 2921 2631 70 17.4 20.5 0.850 1
Albuquerque, NM 2401 2881 2566 90 13.8 18.9 0.927 2
Phoenix, AZ 2394 2873 2496 75 21.3 24.4 0.977 3
Carrisa Plains, CA 2345 2814 2445 70 20.1 26.5 1.100 4
Ely, NV 2230 2676 2377 74 6.7 15.6 1.062 5
Fresno, CA 2169 2603 2243 43 16.8 22.2 1.112 6
Dodge City, KS 2055 2466 2103 78 12.7 23.3 0.950 7
Santa Maria, CA 2039 2447 1947 60 13.8 18.3 1.130 8
Fort Worth, TX 1877 2252 1764 68 19.0 25.6 0.951 9
Columbia, MO 2058 2470 1552 54 12.5 25.6 0.987 10
Apalachicola, FL 1866 2239 1639 67 19.8 26.6 0.900 11
Brownsville, TX 1868 2242 1587 106 18.2 26.7 0.860 12
Omaha, NE 1767 2120 1652 109 9.7 25.6 0.950 13
Great Falls, MT 1715 2058 1641 82 7.2 17.8 0.952 14
Bismarck, ND 1707 2048 1641 72 5.2 22.8 0.880 15
Medford, OR 1699 2039 1582 60 11.6 21.1 1.080 16
Cape Hatteras, NC 1747 2096 1504 110 16.5 25.5 0.820 17
Miami, FL 1802 2162 1418 132 24.1 26.1 0.930 18
Charleston, SC 1690 2028 1354 76 18.2 27.2 0.836 19
Lake Charles, LA 1652 1982 1285 69 20.2 26.7 0.920 20
Nashville, TN 1589 1907 1279 73 15.2 25.6 0.843 21
Madison, WI 1554 1865 1304 77 7.2 25.0 0.939 22
Washington, DC 1559 1871 1287 78 12.1 25.6 0.970 23
Caribou, ME 1418 1702 1183 76 3.8 21.7 0.885 24
Boston, MA 1421 1705 1160 87 10.7 23.9 1.073 25
New York, NY 1402 1682 1087 113 12.5 24.4 1.127 26
Seattle, WA 1299 1559 999 65 10.6 19.7 1.051 27
Source: EPRI [2]
A critical aspect of evaluating economic feasibility is obtaining the correct solar resource, utility
tariff, and incentive data applicable to the specific customer. The Estimator contains databases
for this purpose that are regularly updated to ensure that Estimator licensees are using accurate
data. Table 5-9 describes the contents of the Estimator databases.
Many of the Estimator databases are indexed to locations. Once a location is selected, the
databases provide available utility tariffs for that location. The Estimator is normally configured
to use a default tariff, but it allows the user to select from all tariffs offered by the utility. In
5-22
10581090
Solar Photovoltaics
addition, the selected location determines the available federal, state, and utility economic
incentive programs, the solar resource data, income tax rates, and air emissions rates.
Table 5-9
Information Included in Clean Power Estimator Databases
Database Description
Solar Resource Hourly “typical” solar irradiance by month (12 x 24 values); U.S.
database is derived from the TMY-2 data and includes 237 locations
Utility Incentives The “buy-down” and incentives provided by utilities; can include a
variety of cost, size, and tax treatment constraints
State Incentives Incentives provided by state governments or other jurisdictions below
the federal level
Federal Incentives Incentives provided at the federal level
Electricity Tariffs Over 1300 tariffs for residential and commercial customers, including
net metering tariffs; provides wide flexibility in capturing the range of
tariff structures, including seasonal variations, time of day variations,
metering charges, energy pricing, demand pricing, energy and demand
tiers, sell-back pricing, fuel cost adjustments, public programs, and
other special charges; tariffs are updated using the Estimator’s semi-
automated monitoring tools
Electricity Load Hourly profiles based upon independent utility research showing
Profiles consumption patterns for various locations and customer types, such as
all-electric heating, gas water heating, etc.
State Income Tax Rate Tax rate schedules reflect tax filing status, such as personal income for
Schedules married couples filing jointly, individuals, corporations, etc., for all states
Federal Income Tax Similar information as above, except for federal taxes
Rate Schedules
Depreciation Depreciation schedules such as the Modified Accelerated Cost
Schedules Recovery System (MACRS) of the U.S. tax code used by businesses to
depreciate their PV capital investment
Environmental Air emissions (CO2, SO2, and NOX) offset rates by state based on
Emissions Factors regional mix of sources, such as coal, gas, diesel, hydro, and nuclear
The case studies below show how cost estimates are derived for rooftop PV systems of both
residential and commercial sizes in five different geographical, regulatory, and tariff situations in
the United States. These examples demonstrate that factors such as incentives and rate structures
can be as important as local solar resource to customer economics when deploying a distributed
PV system.
5.2.6.3 Hypothetical Case Studies
Assumed capital costs are based on the median total system cost (per kWdc nameplate rating) for
completed California Solar Initiative (CSI) incentive applications with a reservation request
review date of April through September 2010, rounded to the nearest $100 per kW.18 Residential
18
CSI published Working Data Set (http://www.californiasolarstatistics.ca.gov). For the 5-kWdc residential case
($6,500/kWdc), all systems between 4 and 6 kWdc (628 systems) were included. For 10-kWdc commercial
($6,000/kWdc), all systems between 9 and 11 kWdc (109 systems) were included.
5-23
10581090
Solar Photovoltaics
systems are assumed to have a total installed cost of $6500 per kWdc, and commercial systems
$6000 per kWdc. Table 5-13 illustrates the impacts of various utility, state, and federal incentives
and their cost impacts. The net capital cost on a CEC-rating basis ($/kWac) is also included for
reference. The tabulated values are the net costs to consumers of the investments, after all
incentives and tax impacts are taken into account.
The Estimator methodology was applied to 10 case studies from five utilities across the country.
These represent a range of meteorological conditions, electric rates, net metering availability,
and incentive policies. The 10 cases are defined in Table 5-10.
Table 5-10
Rooftop Case-Study Electric-Rate Scenarios
State Case Name Tariff

NY LIPA Residential Residential Service (Codes 180, 183, 186)
NY LIPA Commercial General Service-Large
Secondary (Code 281)
OH FirstEnergy Residential Residential Service
OH FirstEnergy Commercial General Service
AL Alabama Power Residential Residential Service (Rate FD)
AL Alabama Power Commercial Medium Light and Power (Rate LPM)
AZ SRP Residential Basic Plan (Rate E-23)
AZ SRP Commercial General Service (Rate E-36)
CA PG&E Residential Residential Service (Rate E1 Area X)
CA PG&E Commercial Medium General Service (Schedule A-10-Secondary)
The assumed annual electric bill for each case-study customer was $1200 for residential
systems and $20,000 for commercial systems. Using these values along with specific rate details
(fixed charges, demand charges, energy charges), seasonal variations, and hourly load profiles,
19
the corresponding annual consumption was calculated. Table 5-11 shows annual consumption
and comparative average electricity prices (including all cost components) for the 10 scenarios.
Systems are assumed to have a “dc rating” (i.e., the sum of all the PV modules’ Standard Test
Condition [STC] nameplate ratings) of 5 kW for residential systems and 10 kW for commercial
systems. System performance was modeled using actual equipment characteristics assuming
south facing module orientation with tilt angle equal to the local latitude and no shading.20
CEC-AC system ratings and estimated annual performance by location are shown in Table 5-12.
19
Load profiles by region are included in the Clean Power Estimator database. The data include relative hourly
values by month (12 month x 24 hour matrices). All days within a given month are assumed to be identical.
20
PV modules selected were BP Solar model MST-50, rated at 50 W (100 modules for residential and 200 modules
for commercial). Inverters were Xantrex model PV-5208, rated at 5 kW, and PV-10208, rated at 10 kW, for
residential and commercial systems, respectively.
5-24
10581090
Solar Photovoltaics
Table 5-11
Rooftop Case-Study Utility Rates
Residential Commercial
First Alabama First Alabama

LIPA SRP PG&E LIPA SRP PG&E
Energy Power Energy Power
Assumed Annual Electric Bill ($) 1,200 1,200 1,200 1,200 1,200 20,000 20,000 20,000 20,000 20,000
Annual Energy Consumption

6,215 6,576 10,319 10,620 7,511 126,434 142,407 199,559 219,838 130,435
(kWh)
Average Price ($/kWh) 0.193 0.182 0.116 0.113 0.160 0.158 0.140 0.100 0.091 0.153
Net Metering Offered
Table 5-12
Rooftop Case-Study System Performance

Latitude (°N) 42 40 33 33 37 42 40 33 33 37
System Rating
4.47 4.47 4.47 4.47 4.47 8.57 8.57 8.57 8.57 8.57
(kWac, CEC-AC*)
Annual Energy Produced (kWh) 6,162 5,545 6,624 8,322 7,274 11,805 10,623 12,691 15,997 14,056
Capacity Factor (%) 15.7 14.2 16.9 21.3 18.6 15.7 14.2 16.9 21.3 18.7
2
* The “CEC-AC” rating is calculated from the “PV-USA Test Conditions” (PTC) module dc rating (1000-W/m insolation, 20°C ambient temperature, 10-m above grade, 1-m/s
wind) and the system inverter efficiency. Although the CEC-AC rating does use the more “real world” PTC measure of module performance, rather than the industry-standard
“STC” rating, it ignores other typical system losses, such as module mismatching and field-wiring resistance. Therefore, it is still an optimistic measure of total system output in
most cases.
5-25
10581090
Solar Photovoltaics
Table 5-13
Rooftop Case-Study Capital Costs
Assumed Capital Cost
6,500 6,500 6,500 6,500 6,500 6,000 6,000 6,000 6,000 6,000
($/kWdc) (STC basis)
Assumed Capital Cost
7,270 7,270 7,270 7,270 7,270 7,000 7,000 7,000 7,000 7,000
($/kWac) (CEC-rating basis)
Capital Cost ($) 32,500 32,500 32,500 32,500 32,500 60,000 60,000 60,000 60,000 60,000
Incentives
LIPA Buydown (17,500)
OH PV Rebate (15,000) (30,000)
CSI Buydown-PG&E EPBB (1,549) (2,993)
AZ SRP SolarWise Energy (15,000) (10,037)
AZ Tax Credit (1,000) (6,000)
NY Tax Credit (5,000)
Federal Tax Credit (9,750) (5,250) (9,750) (8,250) (9,285) (18,000) (18,000) (18,000) (18,000) (18,000)
Taxes on Incentives
LIPA Buydown 13,633
CSI Buydown-PG&E EPBB 4,457
AZ SRP SolarWise Energy 11,778
AZ Tax Credit 2,465
NY Tax Credit 1,650
Net Capital Cost
Net Capital Cost ($) 19,400 12,250 22,750 18,580 21,666 31,316 23,883 42,000 31,877 40,025
Net Capital Cost ($/kWac) 4,340 2,740 5,089 4,157 4,847 3,654 2,787 4,901 3,720 4,670
5-26
10581090
Solar Photovoltaics
Finally, Table 5-14 shows the economic results, in terms of both simple payback and the internal
rate of return. These results vary enormously, with simple payback ranging more than a factor of
4, from 13 to 57 years.
The best possible results would benefit from all four primary driving factors:
• High retail electricity price to offset
• Availability of net metering
• High system performance (good local solar resource)
• Availability of incentives
The least economic residential case of those studied is the one at Alabama Power. While its
performance (capacity factor of 16.9%) is in the middle of the range, it has nearly the lowest
average retail electricity price ($0.116/kWh), net metering is not offered, and there are no
incentives other than the Federal Investment Tax Credit that is available to all systems.
Table 5-14
Rooftop Case-Study Economic-Analysis Results

Utility Bill Savings

1122 970 402 779 1146 2024 1659 1406 1419 2285
First Year ($)
Simple Payback
17 13 57 24 19 15 14 30 22 18
(years)
IRR (%) 7.0 10.4 <0 4.2 6.1 15.3 20.9 7.6 11.8 11.7
The most economically attractive case studied is the commercial system in First Energy’s
territory. While its performance is the lowest of the group (14.2% capacity factor), the utility
does offer net metering, the incentives available are the most generous, and the retail electricity
price is above the commercial-case average. It is particularly interesting to compare the LIPA
and SRP commercial cases. They have comparable incentive levels so that their net costs are
similar, and they both offer net metering. But SRP’s far better solar resource (21.3% versus
LIPA’s 15.7% capacity factor) fails to offset the “advantage” of LIPA’s higher electric rate so
that the LIPA system is a rather more attractive investment.
5.2.6.4 Central Station PV Cost and Performance Methodology
This section presents and illustrates a methodology to estimate the performance and cost of
central-station PV power plants using fixed flat-plate, one-axis, and two-axis tracking. Tracking
systems maximize the power output of a particular PV array but also add to its cost and
mechanical complexity. The source of the 2010 PV cost and performance estimates is the 2010
5-27
10581090
Solar Photovoltaics
EPRI report, Engineering and Economic Evaluation of Central-Station Solar Photovoltaic

Power Plants [35].
The resulting estimates represent technology available for commercial order in a given year and
site conditions that include annual insolation, average ambient temperature, and a city cost index.
Because solar PV technology is continually evolving, the projections are based on an assumed
future technology development and market scenario. Hence, these performance and cost
projections are subject to considerable positive and negative uncertainty, especially in the out
years. To evaluate the economics of future central-station PV, three possible technologies are
evaluated below and each is evaluated for various years. For fixed flat-plate PV systems, the
example assumes a scenario of thin-film PV technology evolution, based initially on single-
junction copper indium-gallium diselenide or cadmium telluride technology and evolving to
multi-junction technology involving copper indium diselenide, cadmium telluride, and perhaps
other materials in later years. The overall system efficiency (solar energy to ac power) will
increase from approximately 10% in 2010 to 14.2% by the year 2030. This assumes that the
commercial module efficiency increases from 12% in 2010 to 17% by 2030.
For the one-axis tracking flat-plate PV systems, the development scenario assumes crystalline
silicon PV technology. The overall system efficiency (solar energy to ac power) will increase
from 14.7% in 2010 to 20% by the year 2030. This assumes that the commercial module
efficiency increases from 17% in 2010 to 24% by 2030. Note that the addition of one-axis
tracking also increases the solar input by 20%, relative to the fixed flat-plate system.
For the two-axis tracking, high-concentration PV system, the development scenario assumes that
the technology uses a point-focus, gallium arsenide cell system with 1000x concentration. The
overall system efficiency will increase from 21.7% in 2010 to 40% by 2030. The corresponding
cell efficiencies range from 27% in 2010 to 50% in 2030. The rated outputs are 17,000 kWac in
2010 and 68,000 kWac in 2030.
The capital cost of solar PV power plants is a function of the performance of the solar collectors,
the plant rating and number of modules, and the delivered cost of the modules, arrays, trackers,
controls, power conversion system, and the other balance-of-plant components. O&M cost is a
function of the solar collector area, the complexity of the solar tracking and control system, and
the failure and replacement rates of the plant components. As the technology advances over time,
performance improves, component costs drop, and the market for PV technology will grow,
significantly reducing unit capital and O&M costs.
Tables 5-15 through 5-17 provide estimates of cost and performance for each the three PV
system examples as a function of the year of commercial order. The data are extrapolated from
the May 2010 Engineering and Economic Evaluation of Central-Station Solar Photovoltaic
Power Plants report [35] using staff judgment based on historic PV cost and performance trends.
Performance data include system conversion efficiencies and rated outputs (Wac/m2). Cost data
2 2
include collector area and projections of total plant cost ($/m and $/kWac), O&M cost ($/m /yr
and $/kWac/yr), and land lease cost (% of revenues, $/kWh) for two kWac output ratings: 1) the
nominal output rating of plants to be installed in a given year; and 2) scaled to an output rating of
5 MWac. Table 5-18 lists scaling factors that are used to scale the solar collector area, total plant
cost, and O&M cost estimates to different plant ratings.
5-28
10581090
Solar Photovoltaics
Table 5-15
Solar PV Power Plant Capital, O&M, and Lease Cost Projections for
Fixed Flat-Plate Thin-Film Photovoltaic Power Plants (Dec-09 $)
Commercial Order Year: 2015 2020 2030

System Efficiency: 10% 13% 15%
2
Rated Output (Wac/m ): 100 130 150
Plant Rating (kWac): 20,000 50,000 200,000
2
Collector Area (m ): 200,000 380,000 1,300,000
Total Installed Cost ($/m2) $250 $180 $110
TPC ($/kWac): $2,500 $1,400 $730
O&M ($/m2/yr): $4.0 $3.0 $2.0
O&M ($/kWac/yr): $40 $23 $13
Costs Scaled to 5 MW
Collector Area (m2): 50,000 38,000 33,000
Total ($/kWac): $2,600 $1,500 $810
O&M ($/kWac/yr): $49 $32 $23
Land Lease, % of Revenues: 3% 3% 3%
Table 5-16
Solar PV Power Plant Capital, O&M, and Lease Cost Projections for One-Axis Tracking
Crystalline Silicon Flat-Plate Photovoltaic Power Plants (Dec-09 $)

Rated Output (Wac/m2): 160 180 200
Plant Size (kWac): 20,000 50,000 200,000
Collector Area (m2): 130,000 280,000 1,000,000
2
Total Installed Cost ($/m ) $475 $250 $150
TPC ($/kWac): $3,000 $1,400 $750
2
O&M ($/m /yr): $6.00 $4.50 $3.00
O&M ($/kWac/yr): $38 $25 $15
Collector Area (m2): 31,000 28,000 25,000
Total ($/kWac): $3,100 $1,500 $830
O&M ($/kWac/yr): $46 $35 $26
5-29
10581090
Solar Photovoltaics
Table 5-17
Solar PV Power Plant Capital, O&M, and Lease Cost Projections for Two-Axis
Tracking High Concentration Photovoltaic Power Plants (Dec-09 $)

Rated Output (Wac/m2): 240 270 340
Plant Size (kWac): 20,000 50,000 200,000
2
Collector Area (m ): 84,000 180,000 590,000
Concentration Ratio 1000 1000 1000
2
Total Installed Cost ($/m ) $600 $390 $250
TPC ($/kWac): $2,500 $1,400 $740
2
O&M ($/m /yr): $10.00 $6.50 $5.00
O&M ($/kWac/yr): $42 $24 $15
Collector Area (m2): 21,000 18,000 15,000
Total ($/kWac): $2,800 $1,700 $920
O&M ($/kWac/yr): $51 $33 $25
Table 5-18
Solar PV Power Plant Capital and O&M Cost Scaling Factors vs. Rated Plant Output
Relative to a 5-MWac Plant
Total Plant Cost Factor O&M Cost Factor

Rated
Fixed 1-Axis 2-Axis Fixed 1-Axis 2-Axis
Output
Flat Plate Flat Plate HCPV Flat Plate Flat Plate HCPV
(kWac)
10 2.33 2.33 2.55 2.75 2.75 2.51
50 1.87 1.87 2 2.12 2.12 1.98
100 1.7 1.7 1.8 1.89 1.89 1.78
500 1.37 1.37 1.41 1.45 1.45 1.41
1000 1.24 1.24 1.27 1.3 1.3 1.27
5000 1 1 1 1 1 1
10,000 0.97 0.97 0.95 0.9 0.9 0.91
20,000 0.95 0.95 0.89 0.82 0.82 0.82
50,000 0.92 0.92 0.83 0.72 0.72 0.73
100,000 0.9 0.9 0.8 0.63 0.63 0.65
200,000 0.9 0.9 0.8 0.57 0.57 0.59
5-30
10581090
Solar Photovoltaics
5.2.6.4.1 Example Performance Calculations
As described in the steps below, the annual power generation is calculated by multiplying the
annual insolation estimate for the site from Table 5-8 by the system efficiency from Tables 5-15
through 5-17. The system efficiency allows for soiling, energy collection, power conversion,
step-up transformer, availability, and other losses within the power plant, but excludes substation
losses.
The performance estimating procedure involves the following steps:

1. Obtain insolation and other site data from Table 5-8 and the system efficiency and rated
output corresponding to the PV technology and year of commercial order from Table 5-15
for fixed flat-plate, Table 5-16 for one-axis tracking flat-plate, and Table 5-17 for two-axis
tracking concentrator technologies.
2
2. Calculate the required solar Collector Area (m ) by dividing the kW rating of the plant at
standard conditions (kWac) by the Rated Output (Wac/m2) corresponding to the year of
commercial order:
Collector Area (m2) = [(Plant Rating, kWac)/(Rated Output, Wac/m2)] x 1000

3. Calculate the annual power generation by multiplying the annual insolation (kWh/m2/yr)
by the System Efficiency (%) and the Collector Area (m2):
Annual Power Generation (kWh/yr) =

(Annual Insolation, kWh/m2/yr) x (System Efficiency, %) x (Collector Area, m2)
For a 1000-kWac one-axis tracking flat-plate solar PV power plant, ordered in 2010 and a plant
site near El Paso, Texas:
1. The one-axis tracking insolation rate for El Paso is 2921 kWh/m2/yr % (from Table 5-8).
For the one-axis tracking flat-plate PV system ordered in 2010, the Rated Output is 147
2
Wac/m and the system efficiency is 14.7% (from Table 5-16)
2 2
2. The required solar collector area is: Collector Area (m ) = (1000 kWac)/(147 Wac/m ) x 1000 =
2
6,803 m
3. The net annual power generation is:
Annual Power Generation (kWh/yr) =
(2921 kWh/m2/yr) x (14.7%) x (6,803 m2) = 2,921,000 kWh/yr
5.2.6.4.2 Example Capital and O&M Cost Calculations
The Total Plant Cost includes all direct and indirect construction costs, sales tax, engineering and
procurement costs, and contingencies. The O&M cost includes all labor and material required to
operate and routinely maintain the plant as well as allowances for periodic repair and
replacement of failed modules and other components. If applicable, the lease cost covers the cost
of leasing the land for the plant site.
5-31
10581090
Solar Photovoltaics
The cost estimating procedure involves the following steps:

1. Estimate the Total Plant Cost (TPC, $/kWac) by multiplying the base Total Plant Cost
(Base TPC, $/kWac) corresponding to the PV technology and year of commercial order
(from Table 5-15 for fixed flat-plate, Table 5-16 for one-axis tracking flat-plate, and Table
5-17 for two-axis tracking concentrator technologies) by the City Cost Index from Table 5-8
and the TPC Scale Factor from Table 5-18:
TPC ($/kWac) = (Base TPC, $/kWac) x (City Cost Index) x (TPC Scale Factor)
2. Estimate the total O&M Cost ($/kWac/yr) by multiplying the Base O&M Cost ($/kWac/yr)
from Tables 5-15, 5-16, or 5-17 by the City Cost Index from Table 5-8 and the O&M Scale
Factor from Table 5-18: O&M ($/kWac/yr) = (Base O&M, $/kWac/yr) x (City Cost Index) x
(O&M Scale Factor)
3. If the land is leased by the project, estimate the annual land lease cost by multiplying the
average power revenue rate ($/kWh) by the lease rate from Tables 5-12 through 5-14: Lease
Cost ($/kWh) = (Power Revenue, $/kWh) x (Lease Rate, %)
Back to the example for a 1000-kWac one-axis tracking flat-plate solar PV power plant, ordered
in 2010, and a plant site near El Paso, Texas:
1. The City Cost Index for El Paso is 0.850 (from Table 5-8) and the TPC and O&M Scale
Factors are 1.24 and 1.30 (from Table 5-18)
2. For a 2010 commercial order, one-axis tracking flat-plate technology, and 10,000-kWac
uniform plant rating, the Total Plant Cost is $4,171/kWac, the O&M cost is $60/kWac/yr,
and the annual land lease cost is 3% of power revenue (from Table 5-17).
Thus: TPC ($/kWac) = ($4,171/kWac) x (0.85) x (1.24) = $4,396/kWac
3. O&M ($/kWac/yr) = ($4.43/kWac/yr) x (0.85) x (1.22) = $4.59/kWac/yr. Assuming the power
revenue averages $0.05/kWh or $50/MWh, the land lease cost is:
Land Lease ($/MWh) = ($50/MWh) x (3%) = $1.50/MWh

Although PV operation emits no gases or sounds and generally presents less environmental
impact per deployed megawatt than any other known generation technology, it is not completely
free from impacts and potential hazards. Most notably, some impacts may be expected during
system manufacture involving handling of potentially toxic or flammable materials. These
resemble the hazards encountered in the semiconductor industry, and the PV industry has taken
advantage of many of the approaches practiced there. Other issues exist for polycrystalline thin-
film systems in terms of ultimate disposal or recycling. These issues may prove problematic for
deployment of those specific PV technologies in BIPV applications in the long term, but for
large-scale utility deployments they may be readily addressed in “cradle-to-cradle” recycling
schemes, as outlined in [4].
One significant benefit of PV is its lack of noise. PV is a solid-state technology that operates
completely silently, making it unique among generation options and particularly among
competing distributed generation technologies. The fact that PV does not require noise
5-32
10581090
Solar Photovoltaics
abatement could, in some cases, help offset its cost or argue for its use when other generation
technologies would not be permitted or desirable.
Land-use concerns for most PV deployments are minimal. In operation, PV produces no liquid
or solid waste and requires no fuel handling. In most applications, PV is installed on a building
rooftop, a parking structure, or available nearby land. Building-integrated PV is incorporated
directly into a building’s roofing, siding, or glass surfaces.
Only large utility-scale, industrial, or military PV installations demand large areas of dedicated
land. In most cases, such land consists of inhospitable, desert-like terrain of little use for
alternative development but which nevertheless raises concerns about habitat disruption. In late
2009, potential conflicts between solar power and wildlife habitat were spotlighted with Sen.
Dianne Feinstein’s (D.-Calif.) introduction of the California Desert Protection Act of 2010
(S.2921), which at this writing in October 2010 is still making its way through Senate
committees. The act sets aside approximately 2.5 million acres of the Mojave Desert for
conservation by expanding Death Valley National Park, Joshua Tree National Park, and the
Mojave National Preserve, as well as establishing the new Mojave Trails National Monument
and Sand to Snow National Monument. Several solar power projects—mostly solar thermal but
also photovoltaic—had been slated for siting on land that will be off limits to development if the
bill passes. Press reports claimed that as many as 13 solar power project developers cancelled
proposals rather than work around the restrictions. However, of possible benefit to solar projects,
Section 203 of the bill addresses renewable energy and requires the Bureau of Land
Management, Department of Defense, and U.S. Forest Service to undertake environmental
studies and identify federal land where such projects could be expedited. As of May 2010, the
Bureau of Land Management was considering more than 350,000 acres for potential renewable
energy development—much more than sufficient for any proposed use. In general,
environmental impacts on native plants and wildlife can be minimized with proper planning and
management.
Interestingly, the land-use impacts of PV can compare very favorably with those of other
generation options such as hydroelectric power. For example, one can contrast the output of
Hoover Dam to that of a hypothetical PV installation occupying the same land area as Lake
Mead (which is formed by the Colorado River behind Hoover Dam and provides the water to
turn its hydro turbines):
(Typical Annual Insolation Value) x (PV Module Efficiency) x (Area of Lake Mead) =
2 2
(2000 kWh/m ) x (10%) x (688.6 million m ) = 138 billion kWh per year
Hoover Dam generates an average of four billion kWh per year; its maximum annual net
generation was 10.35 billion kWh in 1984. Thus, on average, Hoover Dam/Lake Mead produces
just 3% as much electricity as would PV occupying an equivalent amount of land. This
comparison is merely illustrative of the potential energy that could be generated using a given
amount of land area, and obviously does not take into account the many other costs and benefits
involved.
5-33
10581090
Solar Photovoltaics
5.4 Design and Deployment Issues
Since distributed and utility-scale PV systems employ essentially the same technology at
different scales, many of their design and deployment issues are the same. Several fundamental
details must be considered when planning a PV project of any scale:
• Project size. For most power generating technologies, large systems are more cost-effective
per unit. However, due to its modular construction, economies of scale have a much smaller
effect on PV system cost than on the cost of conventional power plants such as combustion
turbines.
• Demonstration versus production systems. Demonstration projects are more expensive per
unit of capacity than production systems and, because they are research oriented and employ
new or untested equipment, they can have lower reliability as well.
• Overall project plan. While most structures with the proper orientation and design (roof
support, drainage, etc.) can be readily retrofitted to accept PV, it is preferable to design a
building, subdivision, or business park to optimize the placement of PV from the start.
Structures can be designed with the future deployment of PV in mind even if it is not
included at the time of construction. While it is also possible to retrofit structures for
building-integrated PV, incorporated into roofing materials, window glass, facades, etc., its
full economic and technological benefits are best realized when it is included from design
through completion.
• Type of installation. Ground-mounted systems may benefit from economies of scale because
they can be larger and heavier than roof-mounted systems. However, roof-mounted systems
or building-integrated PV may benefit from savings in land costs or provide ancillary
benefits such as shade or insulation. More sophisticated mounting systems that move a PV
array to track the sun’s motion on either one axis or two axes can increase the array’s power
output but also add to its cost and mechanical complexity.
• PV technology. Alternative technologies should be evaluated to determine which best suit the
system needs.
• Inverter technology. Because PV devices generate dc power, electronic interfacing devices
(inverters) are used to convert the power into ac for connection to the grid. Inverter reliability
has been an issue for grid-connected PV systems for the past two decades, making inverter
replacement or repair a leading O&M cost component. This situation has been improving,
but careful selection of the specific inverter for any large installation is warranted.
• Use of storage. Because PV is an intermittent resource, storage technologies are necessary
to produce dispatchable or “firm” solar power, but they add cost to the system. Where net
metering policies are in place, grid-tied systems often forego storage because they can
transfer excess energy to the grid, essentially using it as a 100%-efficient storage system.
• Type of developer. PV projects can be designed and installed by electric utilities, energy
service companies, solar companies, and private individuals. The system developer’s
background and experience can make a significant impact on a project’s success. For
example, using high-quality components and proper installation practices, an experienced
system integrator can improve the likelihood that a project will be trouble free.
5-34
10581090
Solar Photovoltaics
• Ownership and advocacy. PV has a variety of champions. It makes a difference who initiated
the project under consideration and who are the project owners and team members.
• Non-energy benefits (ancillary benefits). Beyond power, PV can provide additional value in
some applications. Building-integrated photovoltaics can provide insulation, shading, hot
water, and extended roof life among other benefits.
• Distribution system issues. Utility distribution companies have concerns about the safety and
power-quality impact of connecting PV to distribution feeders. These interconnection
questions involve such issues as islanding protection, fault contributions, and voltage
regulation. For most small PV systems, recent IEEE and state utility commission standards
have made interconnection more straightforward.
• Interconnection and inspection requirements. Though they are increasingly standardized,
such requirements still vary greatly among utilities and jurisdictions. Some critics contend
that utility distribution companies and some jurisdictions create barriers to PV by imposing
unnecessarily stringent interconnection and inspection requirements.
• Availability and competitive cost of grid-supplied electricity. PV is recognized as cost-
effective for many off-grid energy applications that are remote from a distribution line. The
technology has advanced to the point where some on-grid applications may also prove cost-
effective in high-price utility locations when coupled with various tax incentives.
• Financial Assumptions. PV system energy cost is impacted by issues involving insurance,
taxes, incentives, rebates, green pricing, return on investment, O&M, and capital costs. For
example, depending on the lifespan assumptions, maintenance assumptions, and the discount
rate used in an analysis, PV energy costs can vary by as much as a factor of two for the same
basic system design at the same location. Costs must be compared on an equivalent basis. Of
course, the local cost of utility-supplied electricity is also frequently part of the economic
analysis as mentioned above.
5.4.1 Utility-Scale Issues
The design and deployment of utility-scale PV systems resembles a simple distribution system
construction project in terms of wiring complexity and civil engineering. The main differences
are that most of the wiring involved is at lower voltages and currents than those in typical
distribution systems. Modern inverters typically do not require an additional transformer, at least
up to tens of kilowatts; however, for larger-scale and higher-than-distribution-level voltages, a
transformer is desirable.
5.4.2 Operating and Maintenance Labor Requirements
Distributed and utility-scale PV systems share many of the same O&M requirements. In general,
PV is the lowest-maintenance generation technology available. Except for tracking systems,
which may require periodic inspections to ensure proper operation of a few moving parts, PV
hardware generally requires no significant scheduled maintenance. As mentioned above, inverter
replacement or repair can be a leading contributor to O&M cost.
5-35
10581090
Solar Photovoltaics
Although occasional panel washing may be necessary, in most climates and locations natural
precipitation limits losses due to dust and grime accumulation to about 10% to 15%. PV modules
are manufactured to withstand a hail-impact test. For glass-encapsulated modules, the primary
environmental hazard is vandalism in cases where its potential is not minimized by the site
design. Lightning is also an infrequent hazard. A typical rooftop PV system located on a home or
business can operate with literally no maintenance for years (as most do).
PV O&M strategies focus on three areas:
• Preventive Maintenance, which entails routine inspection and servicing to prevent

breakdowns and unnecessary production losses. Although increasingly popular, preventive
maintenance programs entail moderate upfront costs and extra labor expense, as well as
possibly contributing themselves to wear and tear on site.
• Corrective or Reactive Maintenance, which addresses equipment breakdowns after the

fact. The current industry standard “break-fix” strategy allows for low upfront costs but also
greater risk of component failure and higher back-end costs. It can be lessened by effective
preventive maintenance and condition-based maintenance strategies.
• Condition-Based Maintenance (CBM), which uses real-time data to prioritize and optimize
maintenance and resources. Increasing numbers of third-party providers are developing CBM
plans to improve O&M efficiency, but with some upfront costs for communications and
monitoring software and hardware.
A major distinction between the three approaches is that preventive maintenance requires O&M
action perhaps once or twice per year, while corrective or reactive maintenance as needed is
typically less frequent and condition-based maintenance is continuous. Overall, the PV industry
is trending toward O&M approaches that promote greater oversight and management capability.
An EPRI survey of PV O&M knowledge and practices [42] concluded that the O&M costs for
PV systems smaller than 1 MW have ranged from $6/kW to $27/kW, or from less than 1% to 5%
of total system cost. For utility-scale PV plants, O&M costs were estimated to vary with the type
of technology and tracking schemes involved. Those estimates are summarized in Table 5-19.
5-36
10581090
Solar Photovoltaics
Table 5-19
Utility-Scale PV Power Plant O&M Cost Estimates
Tilted Single- Single-Axis

O&M Costs Fixed-Tilt Fixed-Tilt Fixed-Tilt Axis Tracking Tracking
($/kW-yr) c-Si CdTe a-Si c-Si c-Si
Scheduled
20 25 25 30 30
Maintenance/Cleaning
Unscheduled
2 2 2 5 5
Maintenance
Inverter Replacement
10 10 10 10 10
Reserve
SUBTOTAL O&M 32 37 37 45 45
Insurance, Property
15 15 15 15 15
Taxes, Owner’s Costs
TOTAL O&M 47 52 52 60 60
5.4.3 Interconnection
Only a very small proportion of PV-equipped facilities operate completely disconnected from the
electric grid, most often for reasons of distance, cost, or the owner’s personal philosophy. Most
residential, commercial, and industrial PV systems are connected to the grid. The interconnection
between a distributed resource and the larger utility grid is a critical factor affecting a project’s
safety and financial and technological viability. The 2003 promulgation of IEEE 1547, the
“Standard for Distributed Resources Interconnected with Electric Power Systems,” provided a
universal standard to help resolve the current patchwork of interconnection practices, which
currently vary from utility to utility and project to project.
PV penetration on the grid is a growing concern among utilities, both because of its novelty as
small distributed generation and the fact that it has different ramping characteristics than other
generators. Customers of San Diego Gas and Electric (SDG&E) in California have installed
more than 33 MW of PV capacity, so that PV now provides 20% or more of the current on nine
of SDG&E’s 950 circuits. The utility is currently studying the growth of PV in its territory and
assessing options, including energy storage, to better balance and manage its output. The effects
of localized PV variability are also the subject of increasing numbers of DOE national
laboratory, industry, and university studies. [43, 44] Preliminary conclusions indicate that the
grid effects of many smaller systems distributed over a large area are significantly smaller than
might be expected from a simple extrapolation of individual system behaviors.
Local, state, and federal policies increasingly allow and even encourage interconnecting
distributed generation with the grid. Net metering programs in many states allow PV owners
to feed short-term excess power they generate to the grid, which is colloquially called “running
the meter backward.” It is important to note that net metering policies vary greatly among
jurisdictions and that only some allow customers to sell electricity back to their host utility
should they generate more power than they consume. Some programs offer payment or credit at
rates less than the retail price, such as the avoided cost rate.
5-37
10581090
Solar Photovoltaics
Net metering is best understood as a method for customers to store self-generated energy that
must be consumed within a relatively short time, typically on the order of one month, effectively
permitting these customers to use the grid as a cost-free 100%-efficient storage battery. Some
distribution companies have taken an understandably conservative approach to supporting
distributed photovoltaic systems. Several potential safety and power quality problems may result
from customer-sited PV projects. Specific power quality challenges associated with
interconnecting PV and other distributed generation (DG) to the utility T&D system include
unintentional islanding, changes in radial feeder power flows, reverse power flow in distribution
networks, loss of effective voltage regulation, harmonic injection and distortions, voltage
fluctuation and flicker, and overcurrent-protection device coordination.
Unintentional islanding is perhaps the most significant and commonly raised concern with
respect to DG, primarily because it can endanger the safety of line workers and the public.
Islanding is a situation in which a portion of the utility system operates separately from the rest
of the system. Line crews working on a section of line they believe to be de-energized may
unexpectedly encounter line voltage and could be electrocuted. Similar danger extends to the
public in situations with downed conductors or other live wires within reach that would have
normally been de-energized by upstream utility switchgear had an island not developed. The
most common way to prevent unintentional islanding is to use voltage and frequency relays on
DG units, set to trip whenever voltage or frequency migrate outside a selected window. This
form of islanding protection is called “passive” protection, and prevents islanding in most cases.
Several utilities are planning to test the interaction of a diverse number of inverters that are
affixed to large individual PV arrays. Other potential issues related to interconnection include
changes in radial feeder power flows, possible reverse power flow in distribution networks, loss
of effective voltage regulation, harmonic injection and distortions, voltage fluctuation and
flicker, and overcurrent protective device coordination. All such issues may be successfully
addressed by appropriate regulatory policies and interconnection practices.
Most existing and proposed interconnection requirements stem from fundamental issues related
to providing a compatible interface. These include safety, reliability, quality, and potentially
damaging interactions. Some requirements originate with distribution system operators and take
the form of interconnect agreements, permits, or public service commission rulings. Industry
standards, such as those promulgated or being developed by IEEE, may be adopted in whole or
in part through such rulings. Local inspection authorities may impose other requirements to
fulfill their obligation to protect the public and comply with building and electric codes. Five
basic functions include electrical isolation via power transformer, controlled connection and
disconnection, a visible and secure disconnect, short-circuit protection, and surge protection and
are common to most interconnection policies or agreements.
In the United States, the PV market is also evolving rapidly. State government mandates for PV,
the promulgation of uniform interconnection standards that in most cases allow net metering, and
recent utility involvement in the sector are making PV more attractive to manufacturers,
investors, and potential buyers. Some private companies have established business units to target
specific markets, such as California or New Jersey. In some cases, demand had outstripped
5-38
10581090
Solar Photovoltaics
supply and resulted in transient delays and price increases. This situation was repeated on a
global scale starting in 2005 and continued into 2008 as module demand outstripped the world’s
current production capacity of polysilicon feedstock for wafers. That situation was fully resolved
by 2009, and in fact led to oversupplies of PV inventory at the beginning of the year that
contributed to greatly reduced module prices. Emerging from the short-term silicon feedstock
shortage, PV pricing, shipment and deployment appeared to resume their historical patterns.
5.5.1 Equipment Markets
In the domestic PV market, five major market sectors can be identified:

1. Grid-connected residential, commercial, and industrial rooftop or building-integrated
applications that rely on incentives offered by state programs as well as the 30% Federal
Investment Tax Credit (ITC), authorized by the Energy Policy Act of 2005. In addition,
ARRA allows project owners to opt for a 30% cash grant in lieu of the ITC through 2010. In
the United States, the commercial sector alone now accounts for nearly half of all PV sales,
and remains dependent on incentives [16].
2. U.S. DOE plus other civilian government agencies committed to using PV and renewables
through efforts such as the Federal Energy Management Program.
3. U.S. Department of Defense, which has been installing both grid-independent and grid-
connected PV systems at military facilities for years. DOD has deployed PV at several
military bases, including a 14-MW system at Nellis Air Force Base near Las Vegas. The
department also procures several megawatts of PV per year for mobile applications for use
by personnel in the field. In late 2009, Secretary of the Navy Ray Mabus set the goal of
producing at least 50% of Navy and Marine Corps shore-based energy from renewable
sources by 2020.
4. Grid-independent buyers of PV systems for off-grid homes, ranches, or farms. An estimated
100,000 off-grid residential PV installations are operating today, most of them supplied by
retail companies, service providers, system integrators, or mail-order organizations.
However, as the number of grid-connected commercial, industrial, and residential
installations grows, this sector is becoming much smaller in proportion to the entire market.
5. Grid-independent commercial buyers, which typically use PV to power remote
communications installations, environmental measurement and monitoring sites, and security
systems.
Figure 5-8 illustrates how PV cells and modules made in the U.S. are distributed by market
sector and end use, according to the most recent data available. According to the EIA, the total
amount of PV produced for all uses in the U.S. in 2008 was 524.3 MW, nearly twice the 2007
output of 280.5 MW [16].
5-39
10581090
Solar Photovoltaics
Residential
Industrial
Electric Power
Transportation
Commercial
Grid-Tied
Consumer Goods & Health
Transportation
Water Pumping
Communication
Sales to OEMs
Remote
Figure 5-8
U.S. Shipments of PV by Market Sector (top) and End Use (bottom) in 2008. (Source: EIA
[16])
Internationally, the World Bank, the Global Environmental Facility, regional development
banks, and other national and international programs have made funds available for developing
renewable technology systems, particularly in developing countries such as Indonesia and Brazil.
This financial and technical support has helped provide PV-based systems to areas where electric
service has never been available. The mix of PV applications in different countries varies
according to their particular electricity infrastructures, local demand, incentives and policy
support, and other variables as demonstrated by Figure 5-9. For example, note the overwhelming
fraction of PV dedicated to grid-connected applications in Spain (ESP) and Germany (DEU) due
to those countries’ aggressive solar programs.
5-40
10581090
Solar Photovoltaics
Figure 5-9
Distribution of PV Applications in Countries Participating in the International Energy
Agency (IEA) Photovoltaic Power Systems Program, 1992–2008.
(NOR = Norway, MEX = Mexico, TUR = Turkey, MYS = Malaysia, CAN = Canada, ISR = Israel,
AUS = Australia, SWE = Sweden, USA = United States, DNK = Denmark, FRA = France, AUT =
Austria, NLD = Netherlands, CHE = Switzerland, GBR = Great Britain, PRT = Portugal, JPN =
Japan, ITA = Italy, KOR = South Korea, ESP = Spain, and DEU = Germany.) (Source: IEA [45])
The international PV market served by U.S. suppliers has historically been two or three times the
size of the domestic PV market and is still somewhat larger, a fact that can have significant
impact on domestic supply, price, and delivery times.
5.5.2 Key Participants
The PV industry is made up of a large number of organizations that can be separated into two
categories: component manufacturers and infrastructure organizations. The first category designs
and fabricates PV cells, modules, and balance-of-system hardware. The second category consists
of integrators that help identify customers, design and install systems, and provide maintenance
services. There may be a few hundred such companies nationwide. Most serve off-grid
customers, but the number and size of those involved in the grid-connected market is growing.
Today, hundreds of PV manufacturers and suppliers are providing components to customers

around the world. The majority are located in Europe, Japan, and the U.S., but China is
experiencing a booming PV industry, as well. In 2009, for the first time, China took the lead as
the world’s leading PV supplier, led by companies such as Suntech Power, Yingli Solar, and JA
Solar. China and Taiwan produced 46%, Europe 18%, Japan 16%, and the United States 5% of
5-41
10581090
Solar Photovoltaics
the nearly 8000 MW of PV shipped in 2009 [29]. Note that shipment by region did not coincide
with demand. In 2009, Europe—responsible for 18% of shipments—accounted for 86% of
demand. The United States was responsible for 5% of shipments and 6% of demand. Many of the
leading PV producers are vertically-integrated organizations that refine basic cell materials,
produce cells, and assemble modules. Another type of organization is the module assembly firm,
which purchases PV cells from various manufacturers and adds value by producing modules. A
list of the leading PV manufacturers is provided below.
5.5.3 Internet Resources
In addition to the vendors and organizations listed in the monitoring guide, the reader is referred
to the following Internet websites. Note that while these sites contain information and contacts
that may be of interest, EPRI cannot vouch for the accuracy of their content.
U.S. DOE Energy Office of Efficiency and Renewable Energy

www.eere.energy.gov.
U.S. National Renewable Energy Laboratory (NREL)
www.nrel.gov/csp/.
Solar Electric Power Association (SEPA)

Formerly the Utility PhotoVoltaic Group (UPVG), SEPA is a non-profit collaboration
comprising more than 700 utilities, energy service providers, and the PV industry working to
create and encourage commercial use of new solar electric power technology and business
models. SEPA also created and maintains the “Solar Data and Mapping Tool,” a web-based
utility that provides project data for solar power projects in the United States.
(www.solarelectricpower.org/solar-tools/solar-data-and-mapping-tool.aspx).
www.solarelectricpower.org
American Solar Energy Society (ASES)

ASES is a national organization dedicated to advancing the use of solar energy for the benefit
of U.S. citizens and the global environment. The society sponsors conferences and publishes a
magazine and white papers on the subject. ASES is the U.S. section of the International Solar
Energy Society (ISES).
www.ases.org
www.ises.org
National Center for Photovoltaics (NCPV)

NCPV is headquartered at the National Renewable Energy Laboratory (NREL) in Colorado.
Its primary members comprise PV researchers affiliated with NREL and Sandia National
Laboratories.
www.nrel.gov/pv/ncpv.html
Solar Energy Industries Association (SEIA)

Established in 1974, SEIA is the national trade association of the U.S. solar energy industry. As
the voice of the industry, SEIA works with its 1000 member companies to make solar a
mainstream and significant energy source by expanding markets, removing market barriers,
5-42
10581090
Solar Photovoltaics
strengthening the industry and educating the public on the benefits of solar energy.
www.seia.org
Energy Information Administration

Part of DOE, the EIA collects, analyzes, and disseminates information about energy sources,
usage, forecast, and other data, generally focused on the United States.
www.eia.doe.gov
University Centers of Excellence for Photovoltaics Research and Education

The U.S. DOE established PV centers of excellence at Georgia Tech and the University of
Delaware to improve the fundamental understanding of the science and technology of advanced
PV devices, create breakthrough PV technology, provide training, and give the U.S. a
competitive edge in PV development.
www.ece.gatech.edu/research/UCEP
www.udel.edu/iec/
NOTE: In rankings compiled by Solar Outlook and Photon International, the companies listed
below were the six largest PV manufacturers in 2009, responsible for roughly half the total PV
production worldwide. The ranges of production estimates reported here reflect differences in the
source data. These companies are included in this list of contacts for informational purposes
only; no endorsement is implied.
First Solar
A manufacturer of CdTe cells/modules, First Solar increased its production from 60 MW in 2006
to approximately 1060–1100 MW in 2009, becoming the first pure thin-film manufacturer to top
the PV production list. It is based in Tempe, Arizona.
www.firstsolar.com
Suntech Power
Suntech, based in Wuxi, China, is one of the fastest growing PV companies in history. It wasn’t
even in the top 10 companies as recently as 2005, but became the world’s third largest shipper
of PV cells in 2007 and produced an estimated 670–704 MW in 2009.
www.suntech-power.com
Sharp Solar
Based in Osaka, Sharp initially focused its PV product line on the Japanese domestic market,
but has more recently begun dramatic expansion of its production together with more significant
U.S. and European marketing. Its residential PV systems employ both single-crystalline and
polycrystalline technology. Sharp produced an estimated 570–595 MW of PV in 2009.
http://sharp-world.com/solar/index.html
Q-Cells AG
This fast-growing German-based company jumped from nowhere in 2002 into the top tier in just
three years, nearly doubling annual production between 2004 and 2006. Q-Cells was one of the
top two producers of PV cells in 2007 and produced approximately 518–586 MW in 2009.
www.q-cells.com
5-43
10581090
Solar Photovoltaics
Yingli Solar
A new addition to the roster of top PV producers, Yingli is headquartered in Baoding, China.
Although founded in 1998, the company only began producing PV modules in 2002, and in 2009
produced around 525 MW.
www.yinglisolar.com
JA Solar
Another new addition to the list, JA Solar is also based in China. In 2009 the company produced
approximately 509–520 MW of multi-crystalline PV cells, and plans to manufacture 1000 MW
in 2010.
www.jasolar.com
5.5.4 Applicable Codes and Standards
Both customers and distribution companies anticipate that adoption of universal standards will
address interconnection requirements, ease the introduction of PV into the marketplace, and
eliminate the complex and sometimes time-consuming or expensive tangle of practices currently
followed. Several respected organizations are leading the drive to create such standards,
including IEEE, Underwriters Laboratories (UL), and the Federal Energy Regulatory
Commission (FERC).
A trend in this area is growing consensus that earlier codes and standards addressing PV
inverters and interconnection were better suited for an environment in which PV on the grid was
rare. As PV becomes commonplace, it appears that more robust and intelligent inverters could be
very helpful in stabilizing the grid during disturbances, rather than immediately disconnecting as
most standards require. This perspective is expected to inform future work on relevant codes and
standards, although it has not yet emerged in the form of new drafts or working group activities.
The list below summarizes the content and status of key interconnection standards already
promulgated or under development. Each standard relies on the precedents the others have set. In
many instances, standards issued by different organizations have been developed cooperatively.
IEEE 929-2000: the “IEEE Recommended Practice for Utility Interface of Residential and
Intermediate Photovoltaic Systems,” which describes the interface, functions, and requirements
necessary to interconnect a PV power system with the electric grid. It also describes acceptable
and safe practices for accomplishing those functions.
UL 1741: the “Standard for Safety for Static Inverters and Charge Controller for Use in
Photovoltaic Power Systems,” is closely related to IEEE 929-2000. In fact, UL and IEEE worked
together to ensure that testing procedures described in UL 1741 ensure inverter compliance with
the guidelines established in IEEE 929-2000.
IEEE 1547: the “Standard for Distributed Resources Interconnected with Electric Power
Systems.” It is seen as a critical milestone for the DG industry. IEEE 1547 provides a uniform
interconnection standard that states and energy companies throughout the U.S. and the world are
expected to use as the basis for their own interconnection practices. The recently Energy Policy
5-44
10581090
Solar Photovoltaics
Act of 2005 also includes language encouraging states to adopt IEEE 1547 as the basis for their
interconnection standards.
5.6 References
1. Renewable Energy Technology Characterizations. EPRI and U.S. Department of Energy:

2. Engineering and Economic Evaluation of Central-Station Photovoltaic Power Plants.

3. “Capital Requirement and Economic Summary, 50 MW Photovoltaic Fresnel Lens

Concentrator Plant,” Scientific Analysis, Inc., EPRI Project RP 2948-8, March 5, 1991.
4. Moskowitz, P.D., National PV Environmental, Health and Safety Information Center:

Bibliography, Brookhaven National Lab, Upton, NY: 1993. Report 11973.
5. Renewable Energy 2000: Issues and Trends, U.S. Energy Information Administration:
February 2001. DOE/EIA-0628.
6. Renewable Energy Annual 2000, U.S. Energy Information Administration: March 2001.
DOE/EIA-0603.
7. Renewable Power Industry Status Overview. EPRI, Palo Alto, CA: December 1998.
TR-111893.
8. Photovoltaic Five-Year Market Forecast 2000-2005, Strategies Unlimited: April 2000.

Report PM-48.
9. 2000 Photovoltaic Industry Competition Analysis, Strategies Unlimited: July 2000.

Report PC-11.
10. “Annual Solar Thermal and Photovoltaic Manufacturing Activities Tables, 2000,”
U.S. Energy Information Administration, www.eia.doe.gov.
11. Distributed Renewable Energy Generation Impacts on Microgrid Operation and Reliability.
EPRI, Palo Alto, CA: February 2002. 1004045.
12. CERTS Customer Adoption Model, Consortium for Electric Reliability Technology
Solutions, prepared for the Transmission Reliability Program, Office of Power Technologies,
U.S. Department of Energy by Lawrence Berkeley National Laboratory, Berkeley,
California: March 2001. LBNL-47772.
13. Assessment of Rooftop and Building-Integrated PV Systems for Distributed Generation.

EPRI: March 2003. 1004204.
14. Photovoltaic Manufacturer Shipments 2000-2002, Strategies Unlimited: August 2002.

Report PM-51.
5-45
10581090
Solar Photovoltaics
15. Bolinger, M., and Wiser, R., “Support for PV in Japan and Germany,” Case Studies of State
Support for Renewable Energy, Lawrence Berkeley National Laboratory, September 2002.
16. "Annual Energy Review 2009, U.S. Energy Information Administration (EIA): August 2010.
DOE/EIA-0384(2009). See also:
http://tonto.eia.doe.gov/cneaf/solar.renewables/page/solarreport/solarpv.html.
17. Hoff, T.E., TAG Case Studies for Clean Power Estimator Model of Distributed PV.
Draft report to EPRI, Clean Power Research, October 2002.
18. Sun Screen II: Investment Opportunities in Solar Power, Credit Lyonnais Securities Asia:
July 2005.
19. Case Studies of Grid-Connected Photovoltaic Systems, Volume 2. EPRI, Palo Alto, CA:
April 2003. 1004203.
20. Analysis of Worldwide Markets for Photovoltaic Products & Five-Year Application Forecast
2006/2007, Navigant Consulting: August 2007, Report NPS-GLOBAL2.
21. Sharp, T., “New Energy for Japan,” Cogeneration & On-Site Power Production, James &
James Ltd., London, U.K.: September-October 2003.
22. Engineering and Economic Evaluation of Central-Station Photovoltaic Power Plants. EPRI,
Palo Alto, CA: 2009. 1017600.
23. Photovoltaic Manufacturer Shipments & Profiles, Strategies Unlimited: September 2003.
SUPM 53.
24. Photovoltaic Balance-of-Systems and U.S. Grid-Connected System Price Analysis 2005,
Strategies Unlimited: June 2005. Report PM 56.
25. Green, M.A., Third Generation Photovoltaics: Advanced Solar Energy Conversion,
Springer-Verlag, Berlin (2003).
26. Solar Outlook, Navigant Consulting: June 2006. Issue 2006-1.
27. Solar Outlook, Navigant Consulting: May 7, 2008. Issue SO2008-2.
28. Solar Outlook, Navigant Consulting: February 23, 2009. Issue SO2009-1.
29. Solar Outlook, Navigant Consulting: April 30, 2010. Issue SO2010-2.
30. Solar Outlook, Navigant Consulting: February 29, 2010. Issue SO2010-1.
31. REN21 2010, Renewables 2010 Global Status Report (Paris: REN21 Secretariat).
32. U.S. Solar Industry Year in Review, Solar Energy Industries Association (SEIA), April 2010.
33. Laying the Foundation for a Solar America: The Million Solar Roofs Initiative, DOE Office
of Energy Efficiency and Renewable Energy (EERE), October 2006.
5-46
10581090
Solar Photovoltaics
34. CPUC Self-Generation Incentive Program Ninth-Year Impact Evaluation: Final Report,
Itron Inc., June 2010.
35. Engineering and Economic Evaluation of Central-Station Solar Photovoltaic Power Plants,
EPRI, Palo Alto, CA: May 2010. 1021320.
36. Development of Renewable Energy Sources in Germany in 2009, Federal Ministry for the
Environment, Nature Conservation and Nuclear Safety: March 2010.
37. PV News, Greentech Media: July 2010.
38. California Solar Initiatives website, http://www.californiasolarstatistics.ca.gov/reports/9-15-

2010/ApplicationStatsByMonth.html.
39. U.S. Solar Market Trends, Interstate Renewable Energy Council (IREC): July 2010.
40. National Survey Report of PV Power Applications in Japan 2009, International Energy
Agency Cooperative Program on Photovoltaic Power Systems and New Energy and
Industrial Technology Development Organization (NEDO): May 2010.
41. Mints, P., Navigant Consulting, Photovoltaics World, PennWell: September 2010. Volume
2010 Issue 5.
42. Addressing Solar Photovoltaic Operations and Maintenance Challenges: A Survey of

Current Knowledge and Practices, EPRI, Palo Alto, CA: July 2010. 1021496.
43. Mills, A. and R. Wiser, Implications of Wide-Area Geographic Diversity for Short-Term
Variability of Solar Power, Lawrence Berkeley National Laboratory, LBNL-3884E.
September 2010.
44. Hoff, T. and R. Perez, Quantifying PV Power Output Variability, conditionally accepted for
publication in Solar Energy, 2010.
45. Trends in Photovoltaic Applications: Survey Report of Selected IEA Countries between 1992
and 2008, International Energy Agency, Paris, France: September 2009. IEA-PVPS T1-
18:2009.
5-47
10581090
10581090
6
GEOTHERMAL ENERGY
Installed Capacity (est. • 10,715 MW worldwide

as of December 2010) • 3,086 MW in the United States
Technology Readiness • Ranging from commercial for shallow hydrothermal to early research for
hot dry rock.
• Moderate-low confidence in cost estimates and projections beyond
commercial range of technology.
Environmental Impact • Some atmospheric emissions and solid waste that can be controlled.
• Unknown long-term impacts on local hydrology and possible related
subsidence.
• Can help meet local wastewater disposal challenges by pumping treated
water to recharge subterranean aquifers.
Economic Status • Hydrothermal is competitive in many markets, although in the United
States it is limited geographically to the Western states.
• Binary technology has potential to be more widespread and economical,
but needs further experience.
• Subsurface technology needs improvement to move beyond prime areas
in the West.
• Re-injection of treated water into geothermal reservoir has been shown
to be economically feasible to prolong life of the wells.
• High-grade Enhanced Geothermal Systems (EGS) (Hot dry rock) are
likely to become competitive in the near future, with reasonable
investments in R&D for improvements and cost reductions in resource
development and deep drilling.
Policy Status • Currently being assessed in most states where potential growth of
available resources adds significant capacity to state RPS.
• Included with other non-hydro renewables.
• Public RD&D funds in critical areas to promote geothermal technologies.
Trends to Watch • Improved drilling and new exploration technologies.
Lower cost for binary-cycle conversion technologies, lower effective
temperatures.
• Better characterization and identification of geothermal resources.
• Enhancement of geothermal systems' waste water re-injection to
increase capacity of current plants.
• Re-injection of treated water into geothermal reservoir.
• Limit to near-term growth potential driven primarily by economics,
permitting challenges, and the lack of economically feasible resources
near transmission facilities.
• Development of plant designs for higher resource temperatures to the
supercritical water region that would lead to an order-of-magnitude gain
in both reservoir performance and heat-to-power conversion.
• Short-term development of shallow, high-grade EGS sites at the margins
of hydrothermal reservoirs
• Use of co-produced hot water from existing oil and gas operations using
binary cycle technology.
6-1
10581090
Geothermal Energy
6.1 Introduction
Geothermal energy is literally energy in the form of heat contained within the Earth. It arises
from the heat of the Earth’s molten interior and occurs mainly in geologically active areas where
the planet’s continental plates meet. It is very accessible in certain areas, such as the “Ring of
Fire” that bounds the Pacific Ocean. These regions are often characterized by volcanism,
earthquakes, hot springs, and/or geyser activity. However, in recent years geothermal resources
are also being used in central China, Utah and other areas considerably distant from geologic
plate boundaries.
Figure 6-1
Map Illustrating the "Ring of Fire" [38]
Geothermal power production is a renewable technology with a worldwide operating capacity of

more than 10,000 MW [28]. Geothermal reservoirs have been a commercial reality in Italy,
Japan, the United States, Iceland, New Zealand and Mexico for several decades. According to
the Energy Information Administration (EIA), the United States is the world leader in electricity
generation from geothermal resources with approximately 14,860 GWh (in 2008). California
alone has approximately 2,605 MW of geothermal power plants that provide over one-third of
the state’s non-hydro renewable capacity [3], exceeding that of every country in the world.
However, the known high-grade pure steam (i.e., not water-steam) reservoir (The Geysers) is
already largely exploited. Future geothermal power generation will depend on locating and
tapping new water-steam, hot pressurized water, and enhanced (or engineered) geothermal
systems (EGS) or hot dry rock (HDR) resources.
Geothermal power plants are dispatchable and range in size from 0.28 MWe to 180 MWe, with
30 to 60 MW considered standard for steam or flashed-steam plants. For binary cycle plants,
smaller sizes are more common, in the range of 15 to 45 MW. Ramp rates of 5 MW per hour
6-2
10581090
Geothermal Energy
are typical of all geothermal power plants. Energy production is not affected by daily or seasonal
resource supply fluctuations, but may degrade over time due to reservoir degradation.
Atmospheric emissions are low or non-existent, and solid wastes are readily managed by
conventional techniques. In addition to use for electric power generation technologies, which are
the subject of this report, hot geothermal fluids can be used directly in aquaculture, greenhouses,
space heating, wood drying, and vegetable drying. Today the U.S. non-electric energy uses
consists of a very small portion of geothermal usage compared to electric power applications.
Two limiting factors in geothermal power development are the high risk of resource exploration
and development, particularly when geothermal power competes against lower-cost fossil fuels,
and time delays from leasing and permitting requirements, which can add significant costs to
projects.
6.2 Resources and Current Installed Capacity

As a general rule, geothermal resources are classified in three categories: hydrothermal-
convection, geo-pressured, and HDR resources. Hydrothermal-convection resources are
subdivided further into vapor and liquid-dominated resources, which produce mostly steam
and hot water, respectively. They occur as a result of heat transfer from geologically active
high-temperature belts to aquifers in close proximity. Geo-pressured resources are hot water
containing dissolved methane under a high subsurface pressure about twice that of normal
hydrostatic pressure. HDR resources are hot rock masses that lack fluid content but are close
enough to the surface to have potential for commercial heat extraction.
6.2.1 Hydrothermal Geothermal Resource Assessment
Currently, hydrothermal-convection resources account for all electricity generation from

geothermal energy in the United States. Its major physical characteristics are the resource
temperatures, depths, and subsurface permeability, which controls the flow of hot water from
the hydrothermal aquifer into wells drilled into a subsurface geothermal reservoir.
6.2.1.1 Resource Temperature
How a resource is developed depends on its temperature: hot, defined as greater than 400°F
(204°C); moderate, 300-400°F (149-204°C); or low, 250-300°F (121-149°C). Hot resources are
generally developed via “dual flash” (two-stage flash) technology and all low-temperature and
many moderate-temperature resources will be developed via “binary” technology. Both
technologies are described in detail in the technology section.
In 2005, a new proposed temperature-based geothermal resource classification system was
developed at the request of the Geothermal Energy Association (GEA) and DOE [29]. With a
new national resource assessment on the horizon, this proposal would further subdivide the
resource temperatures into seven categories as shown in Table 6-1.
The seven resource classes in Table 6-1 are based on temperature as the primary criterion, and
steam fraction in the mobile fluid phase in the reservoir (but not necessarily steam saturation in
the reservoir) as the secondary criterion. The attributes of the reservoirs in the various classes, as
listed in Table 6-1, reflect general industry experience.
6-3
10581090
Geothermal Energy
Table 6-1
Proposed Geothermal Resource Temperature Classification
Mobile Applicable Unusual

Well Productivity and
Class of Reservoir Fluid Production Fluid State at Power Development or
Controlling Factors Other
Resource Temperature Phase in Mechanism Wellhead Conversion Operational
than Temperature
Reservoir Technology Problems
Class 1: < 100°C Liquid Artesian Liquid water Well productivity dependent Direct Use
Non-electrical water self-flowing on reservoir flow capacity
Grade wells; and static water level
pumped
wells
Class 2: 100°C to Liquid Pumped Liquid water Typical well capacity 2 to 4 Binary
Very Low < 150°C water wells (for pumped wells); MWe; dependent on reservoir
Temperature steam-water flow capacity and gas content
mixture(for self- in water; well productivity
flowing wells) often limited by pump
capacity
Class 3: 150°C to Liquid Pumped Liquid water Typical well capacity 3 to 5 Binary; Two- Calcite scaling
Low < 190°C water wells; self- (for pumped wells); MWe; dependent on reservoir stage Flash; in production
Temperature flowing wells steam-water pressures, reservoir flow hybrid wells and stibnite
(only at mixture(for self- capacity and gas content in scaling in binary
the higher flowing wells) water; productivity of pumped plant are
temperature wells typically limited by occasional
end of the pump capacity and pump problems
range) parasitic power need;
productivity of self-flowing
wells strongly dependent on
reservoir flow capacity
Class 4: 190°C to Liquid Self-flowing Steam-water Well productivity highly Single-stage Calcite scaling in
Moderate < 230°C Water wells mixture(enthalpy variable (3 to 12 MWe); Flash; Two production wells
Temperature equal to that of strongly dependent on stage Flash; and occasional
saturated liquid reservoir flow capacity Hybrid problem; alumino-
at reservoir silicate scale in
temperature) injection system
a rare problem
6-4
10581090
Geothermal Energy

Proposed Geothermal Resource Temperature Classification
Mobile Applicable Unusual

Well Productivity and
Class of Reservoir Fluid Production Fluid State at Power Development or
Controlling Factors Other
Resource Temperature Phase in Mechanism Wellhead Conversion Operational
than Temperature
Reservoir Technology Problems
Class 5: 230°C to Liquid Self-flowing Steam-water mixture Well productivity highly Single-stage Silica scaling in
High < 300°C water; wells (enthalpy equal to or variable (up to 25 MWe); Flash; Hybrid injection system
Temperature Liquid- higher than that of dependent on reservoir flow occasionally
dominated saturated liquid at capacity and steam corrosion;
two phase reservoir saturation occasionally high
temperature) NCG content
saturated steam
Class 6: 300°C + Liquid- Self-flowing Steam-water mixture Well productivity extremely Single-stage High NCG content;
Ultra-High dominated wells (enthalpy equal to or variable (up to 50 MWe); Flash silica scaling in
Temperature two phase higher than that of dependent on reservoir flow injection system;
saturated liquid capacity and steam occasionally
at reservoir saturation corrosion; silica
condition); saturated scaling potential
steam; super heated in production wells
steam at lower wellhead
pressures
Class 7: 240°C Steam Self-flowing Saturated or Well productivity extremely Direct Steam Occasionally high
Steam Field (33.5 bar-a wells superheated steam variable (up to 50 MWe); NCG content or
pressure; dependent on reservoir flow corrosion
2,800 kJ/kg capacity
enthalpy)
6-5
10581090
Geothermal Energy
Figure 6-2 illustrates the calculated distribution of geothermal energy as a function of the
resource temperature worldwide. Based on this distribution, it is possible to evaluate the
potential for low-temperature electricity production and direct heat utilization.
Figure 6-2
Distribution Curve of Geothermal Energy as a Function of Worldwide Temperature [39]
6.2.1.2 Depth
Resource development also depends on the depth to which wells must be drilled to tap
underground reservoirs of hot water or steam as well as the permeability of the rock formation
that constitutes the reservoir and, therefore, the rate of flow from the reservoir rock into the
wells. Depth may range from less than 1,000 feet (300 m) to over 7,000 feet (2,000 m). The
combined effects of permeability and well design/performance can be measured as flow rate per
unit of pressure difference that drives the flow. Some new projects are designed to produce from
depths of 10,000 feet or more.
6.2.1.3 Resource Permeability
Permeability of a hydrothermal reservoir has a significant impact on the economics of

geothermal energy projects. Low resistance to fluid flow (hot water or steam) from the
hydrothermal aquifer to the well bore(s) results in high well production rates and therefore,
lowers unit cost to deliver hot fluid to the geothermal power plant or heat-using facility.
Permeability is a function of the degree of fractures in the rock and the porosity of the rock
matrix. Geothermal resources are associated with volcanic and seismic regions of the Earth.
Therefore, fractures due to seismic (i.e., earthquake) faults are often associated with
hydrothermal reservoirs and can often significantly contribute to their permeability. Often, the
pattern of productive wells in a hydrothermal geothermal field follows a fault line—for example,
at Roosevelt Hot Springs near Milford, Utah, where PacifiCorp has its 20-MWe Blundell
6-6
10581090
Geothermal Energy
Geothermal Plant. Or, the limit of the production zone of a geothermal field may be defined by a
rather pronounced line where the change in rock type at a fault ends the zone of a permeable rock
formation containing the water and/or steam.
6.2.2 Geo-Pressured Geothermal Resource Assessment
Geo-pressured resources are natural deposits of hot brine found in conjunction with oil and gas
deposits. Methane gas dissolved in the hot brine, plus the heat of the brine itself, gives these
deposits value as potential energy and power resources. Use of geo-pressured resources depends
upon the economics of natural gas. This is because the low temperature of the hot brine produced
and the great depth, pressure, and hence cost of the producing wells drilled into geo-pressured
zones make electricity generation alone uneconomic. In the United States, geo-pressured
resources are found for the most part in the Texas-Louisiana coastal region near the Gulf of
Mexico. These geo-pressured resources are expensive to exploit and the power generation
potential associated with them is only a fraction—perhaps 20%—of the estimated U.S.
hydrothermal potential [5, 12].
6.2.3 Hot Dry Rock Resource Assessment
Hot dry rock (HDR) resources are still in the experimental stage in the United States. The
resource is characterized by man-made reservoirs of hot water created by fracturing geothermally
heated crystalline rock formations at depths of 2,000 to 10,000 meters [36]. Surface water is then
pumped into the hot fractures and most of that water is recovered through production wells, as in
a natural hydrothermal power system. The superheated water transfers its heat to a secondary
fluid or working fluid and is then recirculated and pumped down an injection well. There are
currently no HDR systems generating electricity on a commercial scale anywhere in the world,
but an enormous amount of potential geothermal energy is available from HDR. By some
estimates, just 1% of the heat contained in the uppermost 10 km of the Earth’s crust equals 500
times the energy contained in all oil and gas resources.
In 1979, the U.S. Geological Survey (USGS) estimated that the known accessible hydrothermal
resource in the United States is about 25,000 MW of capacity for 30 years and the as-yet-
undiscovered accessible hydrothermal resource is approximately 95,000 MW to 150,000 MW for
30 years [4, 5, 6]. These USGS estimates did not consider deeper geothermal resources below |
3 km at the time, which advanced drilling technology for HDR will potentially achieve.
Figure 6-3 is a map produced by DOE’s Office of Energy Efficiency and Renewable Energy
(EERE) indicating geothermal resource temperatures at a depth of 6 km and a 2004 assessment
of heat flow in the United States by Southern Methodist University (SMU).
6-7
10581090
Geothermal Energy
Figure 6-3
U.S. Geothermal Resource Potential Map [Estimated Temperatures (°C) at 6 km Depth]
(top), 2004 SMU Heat Flow Assessment (Bottom)
6-8
10581090
Geothermal Energy
A 2007 study from the Massachusetts Institute of Technology (MIT) estimated that 100,000 MW
of electricity could be installed by 2050 using EGS technology. This projection is based in part
on an abundance of available geothermal resources and readiness of technology elements to
capture the resource. However, a DOE evaluation of the MIT study indicates that significant
advances are needed to achieve the goals of large-scale (100,000 MW) use of cost-competitive
geothermal energy. Needed advances include site characterization, reservoir creation, well field
development and completion, and system operation, as well as improvements in drilling and
power conversion technologies [37]. These technology improvements will also support ongoing
development and expansion of the hydrothermal industry.
Figure 6-4 shows the geographic distribution of estimated Engineered Geothermal Systems
(EGS) resources in the United States.
Wyoming Washington
0.59% 1.25%
Arizona Utah
10.56% 9.11%
California Oregon
9.30% 12.05%
Colorado
10.15% New Mexico
10.75%
Idaho
13.11%
Montana Nevada
3.27% 19.86%
Figure 6-4
Estimated EGS Resources in the United States (source: USGS 2008)
A couple of EGS projects are currently producing power while several others are expected to
produce power in the near term. These include:
• Soultz (France): 1.5-MW plant (in operation)
• Landau (Germany): 2.5-MW plant (in operation)
• Paralana (Australia): 7- to 30-MW plant (in drilling stages)
• Cooper Basin (Australia): 1-MW showcase plant by Geodynamics is scheduled to be
commissioned in 2012, and a 25- to 500-MW plant (in drilling stages) expected to begin as
the world’s first 25-MW EGS plant (originally planned to be operational by 2011-2012, now
delayed until at least 2013)
6-9
10581090
Geothermal Energy
• Desert Peak (Nevada): Plant expansion of an existing natural geothermal field (in planning
stages)
• In October 2008, the DOE selected four new cooperative projects for EGS system
demonstrations in the United States which it hopes will lead to technology readiness by 2015.
6.2.4 Installed Geothermal Capacity
In 2007, the World Geothermal Congress (WGC2005) estimated that the total geothermal
capacity in the world was 9,929 MW. As of March 2009, total installed capacity in the United
States was 3,040 MW according to the Geothermal Energy Association [35].
Table 6-2 presents data from countries that currently generate geothermal electricity for 2000,
2005, and 2007, as well as incremental capacity installed since 2005. The forecast for 2010
considered the existing projects in advanced stage of development. Figure 6-5 shows the
installed world capacity in 2007 [34].
Figure 6-5
Total Installed Capacity in 2007 (GHC Bulletin, September 2007)
6-10
10581090
Geothermal Energy
Table 6-2
Installed Geothermal Power Worldwide [28, 34]
Installed Installed Installed Running

COUNTRY Capacity Capacity Capacity Capacity Increment Increment 2010 Forecast
In 2000 (MW) In 2005 (MW) In 2007 (MW) In 2007 (MW) (MW) (%) (MW)
United States 2228 2564 2884 1935 123 5% 2817
Philippines 1909 1930 1969.7 1855.6 40 2% 1991
Indonesia 589.5 797 992 991.8 195 24% 1192
Mexico 755 953 953 953 1178
Italy 785 791 810.5 711 20 2% 910
Japan 546.9 535 535.2 530.2 535
New Zealand 437 435 471.6 373.1 37 8% 590
Iceland 170 202 421.2 420.9 219 109% 580
El Salvador 161 151 204.2 189 53 35% 204
Costa Rica 142.5 163 162.5 162.5 197
Kenya 45 129 128.8 128.8 164
Nicaragua 70 77 87.4 52.5 10 14% 143
Russia 23 79 79 79 185
Papua New Guinea 0 6 56 56 50 833% 56
Guatemala 33.4 33 53 49 20 61% 53
Turkey 20.4 20 38 29.5 18 90% 83
China 29.2 27.8 27.8 18.9 28
Portugal 16 16 23 23 7 44% 35
France 4.2 14.7 14.7 14.7 35
Germany 0 0.2 8.4 8.4 8 8
Ethiopia 7.3 7.3 7.3 7.3 7
Austria 0 1.1 1.1 0.7% 1
Thailand 0.3 0.3 0.3 0.3 0.3
Australia 0.2 0.2 0.2 0.1 0.2
Total 7973 8933 9929 8590 800 10,993
6-11
10581090
Geothermal Energy
Note that geothermal plants can typically achieve annual capacity factors of 85%-90% or higher
[3, 10, 13]. In fact, dry steam resource such as The Geysers typically run at a lower capacity
factor than possible due to different business arrangements (i.e., there is no requirement that the
utility buy all the electricity that can be generated).
Table 6-3 provides data from the Geothermal Energy Association detailing total U.S. installed
capacity as of March 2009. Figure 6-6 plots the trajectory of U.S. geothermal development.
Table 6-3
US Installed Capacity as of March 2009 [35]
Alaska California Hawaii Idaho Nevada New Utah Wyoming TOTAL

Mexico
0.68MW 2605.3MW 35MW 15.8MW 333MW 0.24MW 50MW 0.25MW 3,040MW
Total Installed Capacity
3100
3050
3000
2950
2900
2850
2800
2750
2700
Mar-06 Nov-06 Mar-07 Jan-08 Aug-08 Mar-09
Figure 6-6
Total U.S. Installed Geothermal Capacity as of March 2009 [35]
6.3 Technology Description
As noted, hydrothermal power generation involves well-established systems and technologies

with decades of practical experience. Most of this discussion will therefore focus on
hydrothermal technologies.
6-12
10581090
Geothermal Energy
6.3.1 Hydrothermal
Hydrothermal power is generated from steam and hot water drawn from geologic structures
relatively close to the Earth’s surface. A geothermal hydrothermal system consists of a
geothermal reservoir, wells, and a power plant (Figure 6-7). The set of producing wells are often
referred to as the field, or the well field. Sites using direct steam for power generation are located
in the United States, Italy, and Japan. Geothermal hot water or brine is used for generation in the
United States, Philippines, New Zealand, Mexico, Japan, El Salvador, Nicaragua, Kenya,
Turkey, Iceland, and the other countries listed in Table 6-2, above [2, 10].
Figure 6-7
Hydrothermal Power Plant at The Geysers, California
In general, geothermal fluids are tapped through wells, also referred to as “bores” or “bore
holes,” and are similar to oil and gas wells which use the same drilling technology. Well depths
typically range from 200 to 3,500 meters. The fluids surging out of the wells are piped out to the
power plant. One or more wells may be connected to feed steam to the turbine. Geothermal hot
water, steam, or vaporized secondary working fluids then impel a turbine, which in turn transfers
the energy by spinning a generator to create electricity. Waste heat from the process, mainly due
to heat rejection from the steam cycle, is then recovered by a heat exchanger or directly ejected
to the atmosphere through condensers and cooling towers. Remnant geothermal liquids,
including any excess condensate, are generally pumped back into the reservoir through injection
wells. In some systems outside the United States, cooled liquid leaving the plant is disposed of
on the surface or in streams. This “cooled” liquid is, of course, “hot” relative to ground surface,
stream, or river temperatures.
If present, non-condensable gases are removed from the system by gas ejection equipment and
released to the atmosphere after treatment mandated by emission regulations. Some emission
control systems may produce sludge or solids that are disposed of either in reclaim fields or in
landfills.
6-13
10581090
Geothermal Energy
The nominal plant size used in most system characterizations and industry comparisons is 50
MWe. Real-world systems range in size from 0.28 MWe to 180 MWe [3, 10]. The technology
driver affecting design, performance, and cost of these systems is directly linked to the reservoir
temperature. In general, higher temperatures facilitate higher thermodynamic cycle efficiencies,
and result in a lower cost of the electricity. This is because higher-temperature fluid has more
energy per unit mass and because the second law of thermodynamics states that higher
temperatures are converted into useful work more efficiently than lower temperatures. To allow
for these variations, this discussion addresses high-temperature reservoirs (flashed-steam
systems) and moderate-temperature reservoirs (binary systems).
In high-temperature geothermal systems with temperatures greater than 302°F (150°C), the
reservoir fluid consists of hot water and/or steam. These higher-temperature systems typically
flash the hot water into steam to drive steam turbines and are typically the better candidates for
power plant electricity generation.
Intermediate-temperature systems have resources between 194°F and 302°F (90°C and 150°C)
and usually require the use of a secondary fluid for power production. These binary power plants
require a closed-loop heat exchanger technology that allows transfer of the heat from the
geothermal fluid to a secondary fluid that vaporizes at a lower temperature, which in turn drives
the turbine.
The 1996 DOE/EPRI Next Generation Geothermal Power Plant (NGGPP) study [13] provides
substantial performance data and a robust set of cost data for a wide range of reservoir conditions
and power plant technologies, and is still used today. Additional general background information
on geothermal electric technologies and resources is available in references [9], [15], [16], and
[27].
6.3.1.1 Major Common System Components and Features

• Reservoir: A geothermal hydrothermal reservoir consisting of hot rock with substantial
permeability, and aqueous fluid in situ. The temperature of the fluid ranges from 212°F to
752°F (100°C to 400°C). The fluid may contain substantial amounts of dissolved solids and
non-condensable gases (particularly carbon dioxide and hydrogen sulfide).
• Exploration/Confirmation: An exploration and reservoir confirmation process to identify
and characterize the reservoir. This process is usually risky, complex and can add substantial
front-end cost to a hydrothermal project. For every successful reservoir, there are three failed
exploration/confirmation efforts. Such costs are usually borne out of developer’s equity and
can be a large barrier to exploration projects. Those costs are accounted for in this discussion
but are not represented in the system schematics. In the long run, the costs of exploration and
confirmation must be included in payments made to developers of geothermal fields or to
producers of geothermal fluids or power.
• Wells: Wells for production and injection of geothermal fluids. These range in total depth
from 200 to 3,500 meters at existing producing U.S. hydrothermal reservoirs. The wells are
drilled and completed using technology for deep wells that has been incrementally adapted
from oil and gas well technology since the 1960s, and are typically larger diameter than oil
and gas wells. The produced fluids range from totally liquid to liquid-vapor mixtures (with
6-14
10581090
Geothermal Energy
two-phase flow at the wellhead). In a dry steam field, the wells produce only steam (and
perhaps very small amounts of liquid condensate). Injection wells at some geothermal fields
return “spent” (heat removed) geothermal fluid to the underground reservoir.
• Reservoir Design and Management: The goal of a reservoir design and management
process is to optimize the harvesting of heat, and thus the production of electricity from the
reservoir, at least cost over the life of the plant. These costs are accounted for in this
discussion but are not represented in the system schematics.
• Surface Piping System: Insulated surface piping that transports fluid between the wells and
the power plant equipment.
• Power Conversion System (Power Plant): A powerhouse that converts heat (and other
energy) from the geothermal fluid into electricity. Power plants consist of a) one or more
turbines connected to one or more electric generators; b) a condenser to convert the vapor
exiting from the turbine (water or other working fluid) to a liquid; c) a heat rejection
subsystem to transfer waste heat from the condenser to the atmosphere. Most systems use
cooling towers (wet or dry), and others use cooling ponds; d) electrical controls and
conditioning equipment, including a step-up transformer to match transmission line voltage;
e) an injection pump that pressurizes the spent geothermal liquid from the power plant to
return it to the geothermal reservoir through the injection wells. Representative power-
conversion (power plant) technologies are described below.
• Operation and Maintenance: Activities and costs related to the operation and maintenance
(O&M) of the system over a typical 30-year useful life of an individual power plant and a 40-
to 100-year production life of a reservoir as a whole. Relative to fossil-fuel-fired power
generation, simple-cycle hydrothermal power plants are much less complex, operate at
relatively low pressure and temperature, and have fewer auxiliaries. Nonetheless, geothermal
plants incur a significant penalty with respect to the severe equipment operating environment
created by saturated steam containing relatively high levels of impurities of non condensable
gases, dissolved and suspended solids. Geothermal maintenance costs are about twice that of
fossil power plants, due principally to corrosion of well casings, surface pipelines, etc.;
deposition of geothermal mineral phases and corrosion products inside production and
injection wells, steam field piping and turbines; and erosion of surface piping, valves and
turbine blades.
6.3.1.2 Direct Steam
Dry steam (also called “direct steam” or “vapor-dominated”) power plants draw from
underground reservoirs of steam. Conventional steam turbines are used with hydrothermal fluids
that are wholly or primarily steam. The plant gathers steam from wells and routes it directly to a
turbine, which drives an electric generator. It rejects waste heat to the atmosphere through a
condenser and cooling tower, and returns remaining geothermal liquids to an underground
reservoir via injection wells. Dry steam resources are rare. The Geysers geothermal field in
northern California (Figure 6-3) is an important example of such a reservoir and accounts for
approximately half of existing U.S. geothermal capacity. However, geothermal developers do not
expect to find any more such fields in the United States. The only other major dry steam
geothermal field in the world is the Larderello field in Italy, site of the world’s first geothermal
power plants, originally installed in about 1904.
6-15
10581090
Geothermal Energy
In some situations, “vapor-dominated” reservoirs can produce two-phase water/steam mixtures.

The vapor-dominated term applies to the governing pressure and flow driver in the reservoir.
Sometimes the transition from nearly all-liquid production to increasingly higher fractions of
steam entering along with the liquid water (or “brine”) can indicate that the reservoir is reaching
an advanced stage of depletion. The higher steam fraction may increase the power generating
capacity for a brief time, but the well is near depletion and a replacement well will soon be
needed. EPRI supported development of a method of assessing changes in reservoir conditions
and predicting changes in the steam/water ratio [26]. The technique is based on the higher
mobility of radon gas in the steam phase than in the liquid phase.
6.3.1.3 Flash Steam
Facilities that use hydrothermal fluids above 360°F (180°C), primarily water, usually use flash
steam technology. This method uses very hot water that flows through wells under pressure.
Flash steam systems include one or two (for single-flash or double-flash, respectively) large
vessels called flash tanks, in which part of the geothermal fluid vaporizes into steam at pressures
below that of the reservoir. These vessels can be located at the power plant or at the individual
wells (at the “wellhead”) or at both, depending on whether single-phase or two-phase
transmission pipelines move the fluids from wells to power plant. The steam flow is typically
18% to 25% of the mass of the fluid from the reservoir (for double-flash plants) and is sent to the
high-pressure and low-pressure inlets of a conventional turbine-generator. Steam volume
depends on conditions in the reservoir and the designs of the production wells and power plant.
The remaining liquid (“brine”) from the second flash tank, containing 75% to 82% of the original
mass, is injected through a well back into the geothermal reservoir.
Three-stage flash power plants that were built and operated in New Zealand in the 1970s were
shut down in the 1980s due to resource depletion. Three stages of flash enable heat to be
converted to electricity with higher efficiency than the double-flash approach, which is, in turn,
more efficient than the single-flash approach. The added costs of each additional stage of flash,
dealing with lower-pressure/higher-volume steam flows, must be traded off against efficiency
and environmental considerations. One figure of merit that improves as a design goes to more
stages of flash and lower temperatures of spent fluid is the “hot-water rate” or “brine rate,”
measured in mass (tons) of geothermal fluid extracted per unit of electricity generated (MWh).
Lower rates mean that less geothermal fluid is required to generate a unit of electricity. In
general, the less fluid produced, the lower the total cost of the production system and the less its
environmental impact. Equipment present in all or most flashed-steam systems includes:
• One or two large flash tanks in which part of the geothermal fluid vaporizes (“flashes”) into
steam at pressures less than the pressure in the reservoir. The turbine in the dual-flash system
shown in Figure 6-8 has dual inlets to admit high-pressure steam from the first flash tank and
low-pressure steam from the second flash tank. The flash tanks can be located at the
wellhead, at the power plant, or at both locations—e.g., the first stage at the wellhead(s) and
second stage at the power plant. At Cerro Prieto in Mexico, just south of the U.S./Mexico
border near Mexicali, the original single-flash system uses flash separators at the wellheads.
A subsequent modification generates power from a second-stage flash and features a central
low-pressure flash steam facility and adjacent power plant. The liquid from the first
separation (wellhead separators) is flashed to generate the low-pressure steam that drives the
added turbine/generator.
6-16
10581090
Geothermal Energy
Figure 6-8
Dual-Flash (Two-Flash) Geothermal Power Plant
• Special features related to minimizing the deposition of silicate scale. The plant depicted in
the system diagram is similar to some but not all U.S. flash plants, and geothermal brine
contains substantial amounts of dissolved silica, which tends to precipitate upon equipment
walls as hard scale if not treated. Mitigation measures include:
− Elevating the conversion cycle’s brine exit temperature above that optimal for maximum
power production, which tends to keep some of the silica in solution (This is the method
of choice when silica presents a small to moderate problem.)
− Using a crystallizer-clarifier system, which consists of a brine solids clarifier and a
return line and injects silica seeds into the first flash tank to prevent the precipitation of
amorphous silica on the walls of the vessel and connecting pipes
− Using a pH modification system, which injects small quantities of acid upstream of the
first flash tank to reduce the pH of the geothermal fluid.
• Environmental control measures. At reservoirs where the concentration of non-condensable
gases (e.g., CO2) is high, substantial gas ejection equipment is attached to the condenser. Steam
or electricity drives the ejectors. If hydrogen sulfide in the gases requires abatement, H2S control
equipment is added downstream of the ejectors. H2S removal is a significant environmental
control system when such removal is necessary to prevent odor at plant sites. Other
environmental control measures can include settling and cooling ponds, water and solid waste
containment and disposal system, and other measures if appropriate or required for local
situations.
6-17
10581090
Geothermal Energy
Descriptions of flashed steam geothermal power plants are presented in the Renewable Energy
Technology Characterizations (RETC) [10]. The NGGPP report [13] provides a range of process
and cost information.
6.3.1.4 Binary Power Plants
For hydrothermal fluids with temperatures below 360°F (180°C), binary cycle technology is
generally most cost-effective. Binary cycles are also applied together with a flash cycle in
“hybrid” or “bottoming cycle” configurations, where the binary cycle generates power from the
lower-temperature liquid that remains after one or two flash steps upstream have exploited the
higher-temperature heat. In this application, the binary cycle is an alternative to the second or
third stage of flash. Thus, the binary cycle is another way to use more of the heat and increase
the megawatt-hours of electricity generated per ton of geothermal fluid brought to the power
plant from the well field.
In a binary cycle power system, the natural geothermal fluid flows in one of two loops that make
up the “binary” system. Figure 6-9 is a schematic diagram of a binary cycle geothermal power
plant.
Figure 6-9
Binary Cycle Geothermal Power Plant (Air-Cooled Design)
Hydrothermal fluid passes through a heat exchanger where it vaporizes a secondary fluid, usually
an organic compound with a low boiling point such as isopentane, which drives a specially-
designed turbine. Having given up its heat, the water is returned to the geothermal reservoir
using a deep injection well. Low-pressure vapor from the turbine is liquefied in the condenser
and re-pressurized by the hydrocarbon pump. Waste heat is ejected to the atmosphere through a
condenser and cooling tower. The water and working fluid are confined to separate closed loops
during the whole process, so there are few or no air emissions.
6-18
10581090
Geothermal Energy
Makeup water is required for the heat rejection system if wet cooling towers are used, but not if
dry cooling towers are used. The binary system characterized here is one with dry cooling (“air-
cooled”). However, wet cooling will usually be less expensive than dry cooling, if cooling water
is available. Due to the lower operating temperature of a binary geothermal power plant, more
water will be needed for the heat rejection system than for an equivalent sized fossil fuel plant.
Technical descriptions of binary-cycle or Organic Rankine Cycle (ORC) power systems and
other systems proposed for moderate-temperature reservoirs can be found in the NGGPP report
[13]. The RETC report [10] references others such as binary systems, vacuum-flash, and
ammonia-based cycles.
Equipment present in most binary systems includes:

• Downhole production pumps in the production wells. These keep geothermal fluid from
vaporizing in the wells or in the power plant, and enhance the production well flow rate.
• A working fluid pump, the “main cycle pump,” that pressurizes the low-boiling-temperature
liquid working fluid to drive it around the power-conversion loop.
• A primary heat exchanger, where heat is extracted from the geothermal fluid and transferred
to vaporize the binary working fluid (isobutane in the case shown in Figure 6-9).
• A turbine that converts energy in the high-temperature high-pressure working fluid vapor to
shaft energy, driving a generator.
• A condenser and cooling system, receiving the low-temperature low-pressure isobutane
vapor exhausted from the turbine and condensing it to a liquid for re-pressurization by the
working fluid pump. The heat extracted from the working fluid in the condenser is rejected to
the atmosphere via a wet or dry cooling system (the “cooling tower”).
Most geothermal binary plants consist of a number of smaller modules, each having a capacity of
1 to 12 MWe net. EPRI has extensively documented the largest binary power system ever built,
the nominal 48-MWe (net) Heber Binary Project, built by DOE and San Diego Gas and Electric
Company as a demonstration project in the 1980s. [17].
The trend in binary-cycle technology since the mid-1980s has been toward binary cycles of
relatively small size, despite the lower cost of electricity that would be expected to result from
economies of scale at larger sizes. The reasons for this are that binary cycles lend themselves
well to smaller sizes and to modular construction techniques, and the economics of financing
favors smaller sizes. For instance, in New Zealand, the large unit size that launched Wairakei at
over 100 MWe more than 40 years ago is no longer seen as the preferred path, because field
development in smaller increments of 20 MWe or less enables a well developer to establish a
revenue stream sooner, without the cost and risk of proving a 50- or 100-MWe resource in the
ground. The 100-MWe unit size was developed in New Zealand when there was a national
government-owned power generation and distribution corporation. Now, with smaller, private,
competing power generation companies in charge of generating electricity, smaller increments of
capital and less time from first exploration to power sales take priority. The new priorities have
led to the choice of binary cycles at 20- to 24-MWe or smaller unit size as the preferred
technology during the past decade.
6-19
10581090
Geothermal Energy
6.3.1.5 Low Enthalpy/Reverse Air Conditioning Cycles
For low-enthalpy geothermal resources, a cycle based on mass-produced air conditioning

components (Carrier Chiller) can be utilized. The cycle uses a single-stage centrifugal
compressor which runs in reverse as a radial inflow turbine, a heat exchanger originally designed
for large chiller applications, and low-cost R-134a fluid. A plant based upon low-enthalpy cycles
has been in operation in Alaska since 2006 and has demonstrated the ability to produce
electricity with high availability (98%) from a 75°C (165°F) geothermal resource and a 5°C
(40°F) river water heat sink at competitive cost. Furthermore, modules can be added to expand
the power of the geothermal plant's low-temperature resources.
In general, this low-cost technology can expand the minimum temperature range from 105°C
(225°F) to around 80°C (175°F) for producing power from lower-temperature, shallow
geothermal hot springs systems. As a partial result, technologies that harness reverse air
conditioning cycles may promote future short-term small-scale geothermal distributed co-
generation
6.3.2 Geo-pressured
Geo-pressured resources are natural deposits of hot brine found in conjunction with oil and gas
deposits. Methane gas dissolved in the hot brine, plus the thermal energy of the brine itself, give
these deposits a role as potential energy and power resources. As mentioned above, the extent of
this resource amounts to only about 20% of the hydrothermal resource and appears to be
expensive to exploit, except for a niche involving wells drilled for other purposes and used to
provide geothermal heat and dissolved methane as an alternative to being plugged and
abandoned [12]. Geo-pressured resources are found for the most part in the Texas-Louisiana
coastal region near the Gulf of Mexico [5, 12]. Use of geo-pressured resources depends upon the
economics of natural gas. This is because the low temperature of the hot brine produced and the
great depth and pressure (and, hence, expense) of the producing wells drilled into geo-pressured
zones make the production of electricity alone uneconomic.
6.3.3 Hot Dry Rock (Enhanced Geothermal Systems)
Hot dry rock (HDR) resources are relatively deep masses of rock that contain little or no steam
or water and are not very permeable. They exist where geothermal gradients—the vertical profile
of changing temperature—are well above average (>50°C/km). Rock temperature reaches
commercial usefulness at depths of about 3 km or more. Traditional hydrothermal systems rarely
require drilling deeper than 3 km, while the technical limit for current drilling technology is to
depths greater than 10 km [36]. The potential energy to be derived from HDR is enormous. The
U.S. Geological Survey estimates that 3 to 18 million EJ of accessible HDR energy exists in the
United States—enough to satisfy current U.S. energy consumption for tens of thousands of years
[4]. Figure 6-10 illustrates HDR geothermal production process.
6-20
10581090
Geothermal Energy
Figure 6-10
HDR Geothermal Production Process
To exploit HDR resources, site developers first create a permeable reservoir by hydraulic
fracturing in a “construction” phase [3]. Then, in the “operating” phase, water from the
surface must be pumped through the fractures to extract heat from the rock [10]. At sufficient
temperature, the water can be flashed to steam and used to generate power. Alternatively, the
injected water can be held at sufficient pressure to maintain it in liquid form for use in a binary
or a flash/binary hybrid cycle. The technical challenges lie in the sub-surface elements of the
system, not the power cycle, notably related to deep drilling and in situ fracturing [9, 15].
Flash steam or binary cycle technologies could both be used with HDR resources depending on
the temperature of geothermal fluid produced. However, because of constraints imposed by high
well costs, most of the accessible HDR resource would likely produce wellhead fluids in the
moderate temperature range best suited for binary cycle technology [9].
To date, HDR resources have not been developed commercially in the United States. Well costs
increase exponentially with depth; because HDR resources are much deeper than hydrothermal
resources, they are much more expensive to develop. Also, although the technical feasibility of
creating HDR reservoirs has been demonstrated at experimental sites in the United States,
Europe, and Japan, operational uncertainties regarding the resistance of the reservoir to flow,
thermal drawdown over time, and water loss have so far made commercial development risky.
Lower-cost resource assessment and drilling technologies will help bring these HDR systems
into commercial use [3, 4, 9]. Significant progress has been achieved in recent tests at Soultz,
France under European Union (EU) sponsorship, and in Australia under largely private
sponsorship. However, HDR systems lack a demonstration of their capability at the present time.
DOE has aptly named its effort to investigate the potential for increasing the amount of
electricity generated from geothermal sources as “enhanced geothermal systems” (EGS). As the
name suggests, EGS technology could be used to enhance and extend hydrothermal geothermal
production zones. The same technology that fractures HDR formations and allows them to heat
water that can flow to the surface to a geothermal power plant could also be applied to extend the
6-21
10581090
Geothermal Energy
life, flow rate, and size of the production zone of a hydrothermal geothermal field. Such
technology could enhance or restore the permeability of rock formations and bring in new
sources of water, such as the treated wastewater from the fringes of The Geysers field in northern
California. It could also help maintain production at sites where natural geothermal steam and
water are depleted or insufficient natural permeability is encountered.
6.3.3.1 Supercritical Cycles
Supercritical fluids reside at a temperature and pressure that can allow the fluid to diffuse
through solids. A supercritical fluid such as carbon dioxide (SCCO2) can be pumped into an
underground formation to fracture the rock, thus creating a reservoir for geothermal energy
production and heat transport. The supercritical fluid used to form the reservoir can also be used
as a geofluid to heat-up, expand, and be pumped out of the reservoir to transfer heat to a surface
power plant or other application.
Utilizing SCCO2 for geothermal power production in an EGS system has promise. Although the
heat capacity of SCCO2 is 40% that of water, it has a much lower viscosity and large density
differential between “hot” and “cold”—i.e., extraction and reinjection points that make this
technology worth further exploration. The large density differential has the ability to form a
thermal siphon in which the heavier “cold” SCCO2 at the injection point tends to sink while the
lighter “hot” SCCO2 at the extraction point tends to rise. This combined with the low viscosity of
SCCO2 could greatly reducing pumping needs, a critical consideration given that EGS systems
are envisioned residing several to tens of kilometers beneath the surface. This advantage of
SCCO2 would have an offsetting effect on its relatively low heat capacity compared to water
vapor. Another consideration is that subsurface minerals are insoluble in SCCO2, thus reducing
mineral scaling issues on well lining and surface equipment [40].
Supercritical cycle technology could use SCCO2 as the geofluid (for geothermal heat exchange in
the reservoir) and remove the need for a binary plant heat exchanger by directly expanding the
SCCO2 through a specifically designed turbine, increasing efficiency while lowering capital
investment. However, the produced SCCO2 stream needs to be dried before entering the turbines
to avoid condensation of liquid water during decompression and cooling. Brown [40] has
postulated that the SCCO2 would eventually dissolve all in situ water in the EGS reservoir,
resulting in pure SCCO2 at the extraction point.
Access to large quantities of CO2 is essential for supercritical plants, first to start the reservoir
and then possibly to make-up for underground capture. The cost of such access could be
defrayed by the CO2 sequestration benefits. Overall, the concept of SCCO2 as a heat transmission
fluid for EGS looks promising for future applications and deserves more study to develop the
scientific basis for a field demonstration.
6.3.4 Hot Sedimentary Aquifer (HSA)
Hot Sedimentary Aquifers are reservoirs in which rain water that has been absorbed into the
ground is heated at temperatures that increase with depth or by contact with hot rocks. The water
collects in porous rocks between two impermeable sedimentary layers, creating an aquifer from
6-22
10581090
Geothermal Energy
which hot fluid can be extracted, usually by drilling. HSA typically requires a binary cycle for
electricity production due to the temperature of the brine. The focus of HSA research is to find
shallow systems that reduce development costs and allow the use of proven hydrothermal
systems and supporting technology. Secondary reservoir stimulation techniques, known as
Secondary Enhancement of Sedimentary Aquifer Play (SESAP), have also been researched as a
way to increase permeability and production rates of HSA.
Sedimentary basins in Australia (Otway, Murray) and in the United States (Northern Gulf of
Mexico areas of Texas and Louisiana) are areas with high potential HSA resource. Figure 6-11
illustrates the HSA production process.
Figure 6-11
Hot Sedimentary Aquifer Production Process
6.3.5 Down-Hole Closed-Loop Systems
Down-hole closed-loop systems place a heat exchanger deep in the earth within the geothermal
resource for extracting geothermal energy. The working fluid cycles through the heat exchanger
in the ground and then returns to the surface for use in the power block. Because this system is a
closed loop, it removes the risk of contaminating the aquifer and uses limited water and space,
produces zero emissions and has low visual impact. Other advantages include no risk of induced
seismic instability and chronic pipe corrosion. The system is targeted for use in new drilled or
existing depleted or abandoned oil and gas wells, which are numerous in the United States, and
may be near existing T&D infrastructure sites.
6-23
10581090
Geothermal Energy
6.3.6 Geothermal Hybrid Plants
From an energy efficiency viewpoint, energy production requires the use of appropriate heat
sources optimized for the temperature/pressure requirements of the turbines. The hybrid
approach, in which more than one heating source is employed for a given application, offers
perhaps the best way to exploit geothermal technologies.
6.3.6.1 Geothermal-Solar Hybrid Cycles
Solar-geothermal hybrid plants combine a concentrated solar thermal field with a geothermal
plant. The arrangement can include either binary or flash steam plants. In either design, heat is
collected in the solar field and transferred to a heat transfer fluid (HTF). In a flash steam plant,
the HTF passes through a heat exchanger with geothermally heated water coming from a
production well. The HTF further heats the geothermal water before it enters the separator,
where it is flashed and then expanded through the turbine. In a binary system, the working fluid
is first heated by geothermal water before it is further heated in a HTF heat exchanger and
expanded through the turbine.
Solar-geothermal hybrid plants (Figures 6-12 and 6-13) offer improved performance over pure
geothermal systems due to the additional heat generated from a solar field. Additionally, they can
offset the risk of premature resource depletion, provide operating flexibility, and take advantage
of high peak summer electricity prices by generating additional electricity during hot summer
days when the solar resource is strongest.
Figure 6-12
Geothermal-Solar Hybrid Production Process for a Geothermal Flash Steam Plant
6-24
10581090
Geothermal Energy
Figure 6-13
Geothermal-Solar Hybrid Production Process for a Hot Water/Brine non-Flash Resource
6.3.6.2 Geothermal + Gas Peaker Cycles
Geothermal plants paired with gas turbine peaker plants offer potential value opportunities. The
reason is that gas peakers can be dispatched by the utility when market prices are high while the
geothermal plants are operated at base load. Moreover, when the gas turbines are on-line, power
produced by the geothermal plants can be enhanced by the heat exhaust from the gas turbines,
thus increasing efficiency. Figure 6-14 is a schematic of a geothermal/gas peaker arrangement.
Figure 6-14
Geothermal-Gas Peaker Production Process
6-25
10581090
Geothermal Energy
6.3.6.3 Geothermal-Fossil Hybrid Cycles
Hybrid steam power plants with geothermal feed-water heating enable the utilization of
geothermal energy in areas with moderate enthalpy resources. This scheme may be applicable to
hydrothermal and EGS-Hot Dry Rock geothermal technologies. The process of geothermal feed-
water preheating as a means of improving performance through an improved plant heat rate
forms both an alternative and an extension to the existing electricity generation methods based
on renewable energy.
6.3.6.4 Geothermal-Biomass Hybrid Cycles for Moderate-Low Enthalpy Sources
Figure 6-15 shows a possible geo-biomass hybrid cycle with energy transfer accomplished by
means of three heat exchangers: a pre-heater, a steam generator, and a super-heater. The energy
source for the first two heat exchangers is geothermal; the energy source for the third one may be
geothermal, or could alternatively be biomass, natural gas, or other available fuels.
In addition to steam cycles, organic cycles—which have operating points that may be more
adequate for geothermal production—can be used as working fluids. The organic working fluid
may generate more power than water due to increased heat transfer in the steam generator, but it
may also have a slightly lower efficiency and probably decrease the geothermal percentage of
power generation. The application of the optimal working fluid needs to be carefully planned and
adjusted for individual sites.
Figure 6-15
Geo-Biomass Hybrid Cycle
6-26
10581090
Geothermal Energy
6.3.7 Non-Electric (“Direct Use”) Geothermal
In addition to the electric power generation technologies addressed in this report, hot geothermal
fluids can be used directly in aquaculture, greenhouses, space heating, wood drying, and
vegetable drying. In the United States, non-electric energy uses are very small relative to electric
power applications. District-heating projects exist in Idaho and other states, and there are a few
new projects in the proposal stage. If the price of energy, especially natural gas, continues to rise,
similar projects will evolve.
Geothermal heat pumps (GHPs) are one of the fastest growing applications of renewable energy
in the world. This form of direct use of geothermal energy is based on the relatively constant
ground or groundwater temperature in the range of 4°C to 30°C (39°F to 86°F). GHPs provide
space heating, cooling, and domestic hot water for homes, schools, factories, public, and
commercial buildings.
Until recently, almost all ground source heat pump installations occurred in Europe and North
America. However, installations expanded to more than 40 countries in 2007, up from 26
countries in 2000. China is the most significant newcomer in the application of heat pumps to
space heating.
The potential of GHP is very large, as space heating and water heating are significant parts of the
energy budget in large parts of the world. In industrialized countries, 35% to 40% of the total
primary energy consumption is used in buildings. Future development predictions show
exponential worldwide growth in the heat pump sector: up to 200,000 MWth for 2020 to about
750,000 MWth by 2050. In contrast, estimated capacity worldwide in 2008 for geothermal heat
pumps was about 30,000 MWth.
6.4 Technology Status
While geothermal power production (hydrothermal, not HDR) is generally considered a

commercial technology, there are many opportunities for improvement through research and
development. Geothermal projects are very capital intensive. Continued research and
development will be useful on many fronts, particularly in developing improved geophysical
methods to detect geothermal reservoirs and resources, decreasing the cost of drilling geothermal
wells, decreasing the cost of power production by reducing maintenance cost, increasing the
conversion effectiveness of power plants sited at low-temperature reservoirs, and decreasing
O&M costs of wells, field equipment, and power plant equipment performance.
Technologies for exploration, drilling, and reservoir analysis and management are essentially the
same for all types of geothermal systems. There is substantial room for improvement in most
aspects of these technologies, including both the fluid-production (exploration, wells, and
reservoir management) and electricity-conversion (power plant) components. Improvement in
drill bit technology is expected to help reduce the cost of drilling deep geothermal wells by about
20% over the next decade. The cost of conversion technologies should decrease substantially
over the next decade for lower-temperature systems such as binary-cycle plants, but may not
decrease much for higher-temperature flash systems. The main thrusts for reducing power plant
costs involve [10, 13]:
6-27
10581090
Geothermal Energy
• Substantial changes in the basic conversion cycle designs used in the plants, including the
addition of “topping” and “bottoming” cycles, improved working fluids, and the used of
various hybrid cycles. An example of a topping cycle is the Rotary Separator Turbine (RST)
developed by Douglas Energy and previously by Biphase Energy Systems, with major testing
support from EPRI during 1978–1984 and from Utah Power & Light in 1981–1987.
Subsequent testing has been done at Cerro Prieto, the geothermal field in Mexico just south
of the California border, with support from the California Energy Commission and DOE. The
RST extracts added electric power at the high-temperature side of a flash process by having
the flash and separation take place in a rotary separator designed to capture the kinetic energy
that the flashing steam imparts to the liquid phase. Examples of bottoming cycles include the
third stage of flash in use in New Zealand, a binary cycle added at the low-temperature side
of a flash cycle, and a special low-temperature water-ammonia binary cycle (the Kalina
Cycle™ of Exergy Inc.) added below a flash or a binary cycle.
• Efforts on the part of system operators to reduce O&M costs, especially by reducing the
number of staff employed at each system and site. This is being accomplished through
increased automation of controls and monitoring. An example is the operation and
maintenance of the Wairakei and Ohaaki-Broadlands fields by Contact Energy in New
Zealand. The central control system at the Wairakei plant also controls the Ohaaki plant 15
km away, and a roving operator performs monitoring and maintenance tasks for both plants.
• Gradual reduction in complex instrumentation and controls as operations are streamlined and
engineers learn what components are safe to omit.
The consensus in the industry is that characterizing the commercial potential of identified
geothermal reservoirs is a high priority [3]. Techniques such as fracture mapping, more accurate
thermal gradient wells, and other untested methods should be evaluated and refined if
appropriate. In the near term, such work will focus on:
• Techniques for improving the cost-effective definition of resources as discussed above
• Reducing the costs of drilling production wells
• Improving the efficiency of geothermal facilities
• Understanding the results of wastewater injection and improving this process
• Investigating environmental impacts of geothermal energy developments.
6.4.1 Recent U.S. Developments
Geothermal electric power plants in the United States are located in eight states: Alaska,
California, Hawaii, Idaho, Nevada, New Mexico, Utah, and Wyoming. The total installed
capacity was 3,040 MW as of March 2009, with 12 states with geothermal projects currently
under consideration or development. A Geothermal Energy Association survey identified up to
3,860 MW of new geothermal power plant capacity under development in the United States,
which includes projects in the initial development phase [35].
6-28
10581090
Geothermal Energy
6.4.1.1 Alaska
The first geothermal power plant in Alaska was installed in 2006 at Chena Hot Springs. The
plant pairs the direct use of geothermal thermal energy with on-site geothermal electricity
generation, and uses a binary cycle plant to produce 200 kW gross from the coldest geothermal
resource worldwide, just 74°C. Two second-generation units were added (200 kW and 280 kW,
respectively), increasing the state's combined installed capacity to 680 kW. Alaska has 60 to 100
MW of additional geothermal power in early project development.
6.4.1.2 California
In California—site of the vast majority of U.S. geothermal power production—geothermal

development continues at a high level of activity, with 50 MW under construction at North
Brawley 1 by Ormat Technologies, Inc. An additional 367.4 MW of projects were in the process
of securing power purchase agreements and permits in early 2008. Early project development is
underway for another 900–1,000 MW, though not all of that capacity is likely to come on line.
Geothermal power plants operating in California include:

• The Geysers: 22 dry steam units, totaling 1,531 MW of installed capacity (but only 932 MW
operating due to a steam shortfall)
• Imperial Valley-East Mesa: 70 units totaling 100 MW of installed capacity
• Imperial Valley-Heber: 14 units totaling 94 MW of installed capacity
• Imperial Valley-Heber South: one unit totaling 14 MW of installed capacity
• Imperial Valley-Salton Sea: 10 units totaling 327 MW of installed capacity
• COSO: nine units totaling 270 MW
• North Brawley one unit of 50 MW
• Others: 29 units.
The total installed geothermal capacity in the state as of March 2009 was 2,605.3 MW.
Geothermal power supplied about 4.5% of California’s electric generation, producing a net total
of more than 14,000 GWh/year. With an aggressive renewable portfolio standard, California
expects the state to at least double its geothermal capacity in the next decade.
6.4.1.3 Hawaii
Currently, one geothermal power plant operates on Hawaii. The existing 10 flash + binary units
provide 35 MW installed capacity (30 MW running, after rehabilitation and work over) and no
additions have been made since 2005. This power plant supplies approximately 20% of the total
electricity need of the Big Island and its 160,000 inhabitants. An 8-MW expansion to the plant is
planned, with permitting in place to add an additional 22 MW thereafter.
6-29
10581090
Geothermal Energy
6.4.1.4 Idaho
In early 2008, Idaho’s first geothermal power plant came on line. The Raft River binary plant has
a nameplate rating of 15.8 MW, though currently its net electrical power output is between 10.5
and 11.5 MW. The plant’s owner, U.S. Geothermal, is in the process of expanding the project to
39 MW. There are several other projects in the early development phases that could potentially
add 250–325 MW of electricity in Idaho.
6.4.1.5 Nevada
In 2008, 18 geothermal plants operated in Nevada, with a collective nameplate capacity of 333
MW. Between confirmed and unconfirmed projects, a total of 60 projects are under development
in Nevada, which could eventually supply between 1,767 and 3,297 MW of electricity.
6.4.1.6 New Mexico
Raser Technologies’ 240-kW pilot installation project at Fallon went on line in July 2008,
making it New Mexico’s first geothermal power plant. The full project, named Lightning Dock,
is a binary plant designed to produce 10 MW of electrical power and is expected to go on line by
the end of 2008. A direct-use aquaculture facility is also operating in New Mexico, with the
potential for a 1-MW plant to be built at that site in the future.
6.4.1.7 Utah
Utah had three geothermal power plants in operation as of March 2009. The Blundell Plant has
two units for a total of 36 MW. Utah’s second geothermal power plant, Thermo Hot Springs,
came on line in December 2008 and has a gross capacity of 14 MW.
In April 2007, Enel North America, Inc. acquired the decommissioned Cove Fort plant,
which was being redeveloped to supply 36.6 MW in 2009 and a total of 65 MW by 2011. Raser
Technologies is developing a 20-MW project in central Utah. Including other projects in early
development phases, there is a potential for an additional 194 MW of geothermal plants in Utah.
6.4.1.8 Wyoming
The first geothermal power plant in Wyoming came on line in September 2008. The co-
production demonstration project consists of a 250-kW organic Rankine cycle power unit.
6.4.1.9 Other States
Recent interest in geothermal energy in particular and renewable energy in general has
precipitated expansion of the geothermal electricity market, with a number of states seeing
developers working on projects. Altogether, there are a total of 126 confirmed geothermal
projects in the United States in various stages of development [35].
6-30
10581090
Geothermal Energy
In addition to the projects identified above, developers have plans for geothermal power plants in
Arizona, Colorado, Florida, Oregon, Washington, and Wyoming. Other states such as Montana
and Texas are eager to explore their potential geothermal power. The Energy Information
Administration’s Annual Energy Outlook 2007 (AEO2007) report [7] forecasts increased
geothermal electricity generation, particularly in the Western United States, where geothermal
output could increase from 15 billion kWh in 2005 to 21 billion kWh in 2025.
6.4.2 Hot Dry Rock Technology Status
Hot Dry Rock (HDR) field investigations began in the early 1970s at Fenton Hill, New Mexico,
near the Los Alamos National Laboratory. This project was the first attempt in the world to make
a deep, full-scale engineered reservoir. Experiments from 1973 to 1996 aimed to develop
methods to economically extract energy from HDR systems located in basement rock at suitably
high temperatures. Support for the project declined during the later phases of work and by 2000,
all experiments were terminated and the site was decommissioned.
Building on the experience of the Fenton Hill project, experiments were performed at two sites
in Japan during the 1980s and 1990s. These experiments included work similar to that of the
DOE EGS program, namely enhancing a poor or declining conventional hydrothermal hot-water-
producing field through targeted rock fracturing in zones where more permeability was needed.
The most significant experiment in the world since 1994 has been the European project at Soultz,
France, where two fracture zones have been completed, one deeper and hotter than the other. The
deep zone could be the source of sufficiently high-temperature hot water to drive an electric
power generation station. Circulation tests have been performed, from injection well to hot rock
fractured by high-pressure injection to the production well and back to the surface. The wells
were stimulated by adding salt and nearby lake water to provide initial brine. During the 2001 to
2004 period, a pilot plant was built using the previously drilled and stimulated three-well system.
During the second phase, which was to last from 2005 to 2008, an initial 1.5-MW power plant
was constructed for operation using the medium circulation loop of 26 gal/s [100 l/s]. The plant
went online in June 2008.
In mid-2003, Geodynamics Ltd. of Australia began drilling to build the world’s first commercial
HDR plant. The well will extend approximately three miles (5 km) under South Australia’s
Cooper Basin to tap a promising geothermal region in which rocks achieve temperatures of about
2
550°F (290°C). HDR exploration in South Australia alone covers an area of 991 km and has
2
potential for future capacities in the thousands of megawatts. The nearly 1000 km HDR resource
is equivalent to 50 billion barrels of oil or 10.3 billion tonnes of coal—20 times larger than all
the known Australian oil reserves and equivalent to 40 years of the nation’s current black coal
production.
Geodynamics has secured the rights to another HDR field in the Cooper Basin and two others
in the Hunter Valley in New South Wales. After developing an initial demonstration HDR plant
in the 10- to 15-MWe range, the company plans to build a commercial-scale power plant capable
of generating hundreds of megawatts.
6-31
10581090
Geothermal Energy
Currently, Habanero 1 & 2, Geodynamics’ first two geothermal wells, have produced good
results. The company has successfully completed the riskiest task of the project, hydraulic
stimulation of the underground wells and heat exchanger. In 2005, the firm reported that using a
larger than expected heat exchanger had exceeded expectations. An initial water circulation test
in spring 2005 lasted 40 hours and reached temperatures of 389°F (198.5°C). Habanero #2
produced about 10 MW of thermal power during diagnostic flow tests. The flow test in April
2005 found some potential for power production, but concluded that a dropped plug was
restricting the connection to the geothermal reservoir. The company’s attempt to drill another
sidetrack encountered problems and was concluded in June 2006 when the drill pipe became
stuck. A new well, Habanero #3, is currently progressing in the last segment of drilling from
4,011m (13,158 ft) to the total well depth of 4,252 m (13,950 ft).
HDR development to date indicates that there is a very good potential for economic energy
extraction in the future. This potential has been considerably enhanced by the discovery of
overpressures in the granite fracture network, which could add a large convective heat
component into the original design. Overall, EGS technology has the potential to provide a
sizeable addition to current geothermal production.
6.4.3 Geothermal Technologies for Electricity Production Deployment Curve

(Grubb Curve)
Figure 6-16 illustrates the Grubb curve, which communicates the maturity of different
geothermal technologies. The curve can be used to understand the cost required to complete
research from concept to finished product or to complete a particular stage of technology
development. While early-stage development does not require as much capital relative to
development at middle stages, technologies in the last stages of development may require
significant capital to purchase major components and conduct full-scale test over many months
or years.
Technical risk can also be assessed using the Grubb curve. In the first development stages little is
known about the core technology with regard to the size, details, and costs of its final form. As a
result, the technical risk is at its highest in the early stages of development. As the technology
matures, the technical risk lessens but the overall risk to the development project can remain high
when considering the likelihood of failure.
6-32
10581090
Geothermal Energy
Figure 6-16
The Grubb Curve
6.4.4 Geophysical Methods in Geothermal Exploration
This section explains the diverse geophysical methods available for geothermal exploration as
well as their respective limitations; advocates the need to integrate the different geophysical
methods to minimize drilling risk and development costs; and emphasizes the importance of
complementing geophysical information with geological information to produce reliable results.
Geophysical methods in geothermal exploration can be divided into four main groups depending
on the physical parameters measured:
• Potential methods based on the density and magnetic properties of rocks as well as two of the
Earth's potential fields—magnetism and gravity
• Electric and electromagnetic methods, based on the electromagnetic properties of rocks
(conductivity, permittivity) and the Maxwell equations
• Seismic methods, based on the elastic properties of rocks and the equations of wave
propagation in continuous media
• Radiometric methods, based on the radioactive emission of rocks and atomic physics
equations. (These methods are most commonly used in well logging.)
Each method has a specific application, depending on the physical properties of the target and
how precisely these properties can be detected by the technology available.
6-33
10581090
Geothermal Energy
Gravimetric methods are comparatively easy to use and fairly economical. They provide a good
estimate of the extent of bodies with certain density. The resolution and quality of data, however,
decrease considerably with depth. Gravimetric studies therefore provide a useful tool for shallow
reservoirs.
Similarly, magnetic methods have been very popular during the last 30 years because of the
rapidity with which the measurements can be made and the low cost of the operation.
Restrictions are the depth of resolution, the complexity of the interpretation which compromises
reliability for structures with complicated geometric shapes, and insensitivity to the actual
presence of water.
Methods for measuring the electrical resistance of the sub-surface can be divided in two general
groups:
• Those that measure the difference in electrical potential;
• Those that measure an electromagnetic field, naturally or artificially created.
Electrical potential has been used mainly for shallow depths; for example, for groundwater
aquifers or very shallow geothermal reservoirs.
The most common geophysical methods used today for geothermal exploration are
electromagnetic. They can be induced actively via the Transient Electro-Magnetic (TEM)
method, which is now routinely applied to depths of 2000 m. For greater depths, the Magneto
Telluric (MT) method, which measures the Earth’s impedance to naturally occurring magnetic
waves, has become the standard approach.
Seismic methods use the propagation of elastic waves, which are either generated artificially by
an explosive source or occur naturally due to earthquake activity. Active seismic methods are the
standard tool for hydrocarbon prospecting, as they can be used to supply a detailed image of the
sub-surface structure in the sedimentary environment of most oil and gas reservoirs. Passive
seismology, if recorded appropriately, can be used to help understand the structural context or to
give an outline of the actual fluid/geothermal reservoir.
Electrical and seismic methods are also used for down-hole tools. Some specific tools were
specifically developed for well logging, such as to perform radiometric measurements of the rock
units accessed by the well via neutron and gamma ray measurement.
The principal methods used for geothermal regional studies are MT and seismic surveys, while
gravity and aeromagnetic approaches are often applied to increase the reliability of conclusions
based on the former two methods. Seismic tomography of the Earth’s crust, which is quite
expensive, reveals geometrical boundaries. Without geological information, none of the
geophysical methods can produce reliable results upon which to base decisions on the existence
of a geothermal reservoir and its boundaries. So, seismic and electromagnetic methods are often
supplemented by gravity and/or aeromagnetic surveys of a different scale that depends on the
required resolution. Magnetic anomalies correlate quite well with horizontal zonation of the
structure, while gravity maps are used to locate deep faults and delineate fracture zones
characterized by low density.
6-34
10581090
Geothermal Energy
Since the mining cost (exploration and drilling) to access geothermal resources represents over
60% of a project's total investment, a reduction in mining cost would increase the
competitiveness of geothermal energy significantly. This goal can be achieved by developing
innovative detection methods prior to drilling fluid bearing zones in naturally and/or artificially
fractured geothermal reservoirs.
There is substantial RD&D associated with geophysical methods improvement and application:
• Before and during production: Identify prospective reservoirs without drilling, define
boundaries (lateral and vertical), and improve methods to identify drilling targets
(productivity).
• During and after production: Continuously characterize the reservoir, predict fluid
circulation during stimulation, track injected fluids and predict reservoir
performance/lifetime (effectiveness and sustainability).
6.4.5 Geothermal Drilling Technology
Well costs are a significant economic component of any geothermal development project. Insight
into geothermal well costs is gained by examining trends from experience in the oil and gas
industry. The similarity between oil and gas wells and geothermal wells makes it possible to
normalize any geothermal well cost from the past three decades to present current values, so that
the well cost can be compared on a common dollar basis.
Wells are classified in three categories:

• Shallow wells (1,500-3,000 m)
• Mid range wells (4,000-5,000 m)
• Deep wells (6,000-10,000 m).
Shallow wells supply the majority of current hydrothermal plants. Unfortunately, however, a
scarcity of geothermal well cost data makes it impossible to estimate statistically meaningful
well costs via a comparison to oil and gas wells costs. There is simply too large a degree of
uncertainty; an average completed geothermal well may cost up to 30% more than oil and gas
well for the same depth (up to 3,000 m). Well design concepts and predictions for the deeper
categories—6,000 m, 7,500 m, and 10,000m—would be even more speculative as there have
been only two or three wells drilled close to depths of 10,000 m in the United States.
Emerging technologies, which have yet to be demonstrated in geothermal applications and are
still going through development and commercialization, have the potential to significantly reduce
well costs, especially those at 4,000 m and deeper.
Technologies with the potential to reduce well costs include:

• Expandable tubular casing;
• Drilling with casing that may permit longer casing intervals;
6-35
10581090
Geothermal Energy
• Well design changes, particularly involving the use of smaller increments in casing diameters
with depth;
• Under-reaming, a key enabling technology for almost all of the EGS drilling applications.
Three revolutionary drilling technologies that may have a particularly large impact on lowering
EGS drilling costs are hydrothermal flame spallation and fusion drilling, chemically enhanced
drilling, and metal shot abrasive-assisted drilling. Each of these methods augments or avoids the
traditional methods of penetration based on crushing and grinding rock with a hardened material
in the drill bit itself, thereby reducing the tendency of the system to wear or fail.
In addition, rate of penetration (ROP) issues can significantly affect drilling costs in crystalline
formations by increasing costs by as much as 20% above those for more easily drilled basin and
range formations.
Geothermal power systems combine fuel supply and power conversion systems into one system.
The geothermal fluid serves as the equivalent of fuel. Unlike fuel-fired power systems, in a
geothermal plant the fuel supply (the geothermal resource) and electricity generation (the power
plant) are integrated and physically connected. As a result, the power plant cost is even more
site-specific than for other power generation technologies. The cost is affected not only by the
size and design of the power plant, but also the geothermal resource temperature and pressure,
steam, impurity and salt content, and well depth.
This close and direct link is emphasized in the cost breakdowns provided here by assuming a fuel
cost of zero, and including the capital and operating costs of the geothermal field as part of the
costs of the power plant itself. In this way, geothermal is treated with the same type of cost
breakdown as solar and wind energy. In contrast, biomass is treated as having a fuel supply that
is separate from the power plant system, analogous to a coal-, gas-, or oil-powered plant, which
results in a fuel cost line item in the cost breakdowns provided.
Geothermal power plants are able to generate energy at all times, except during planned and
unplanned maintenance and repair episodes. Experience has led to longer periods between major
overhauls, with many operators achieving five-year time spans. Because of the strong economic
incentive to stay on line, unplanned maintenance outages are kept to a minimum through the
application of highly reliable equipment and adequate spare parts inventories. Some facilities are
achieving capacity factors of 85% to 95%. Plant operators prefer to run their plants as baseload
units because the fixed costs of operation far exceed the variable costs. An exception occurs
when the fuel supply (steam) is less than that needed to satisfy the power plant’s capacity. In
such a case, the plant will operate at or near full capacity during peak-use periods (highest price
for electricity) and then back down during off-peak periods.
Experiments have shown that some geothermal power plants, such as the dry steam plants
at the Geysers facility in northern California, can be cycled to follow system load in the
intermediate-baseloaded area of the utility time-demand curve, thereby increasing their value in
certain applications. It is likely that load following would be more difficult for flash and binary-
6-36
10581090
Geothermal Energy
cycle plants than for dry steam plants. Current contract capacity factors are on the order of 60%
to 80%, but capacity factors for many operating plants are close to 100%. A capacity factor
greater than 100% is possible, because some plants can operate above their design and nameplate
rating nearly all the time. Recent cost reductions in flash power plants may enhance their
competitiveness in the United States. In the international market, geothermal development in
regions such as Indonesia and the Philippines is returning to the strong growth that began in the
early 1990s [2].
During the 1990s, competition from low-cost electricity generated from natural gas hurt the
economic competitiveness of geothermal electric systems, slowing their development and
deployment. During the early 1990s, geothermal power developers expected to compete
successfully against 6 to 7 cent/kWh power in the Western United States. By the end of the
decade, they instead found themselves competing against 2.5 to 3.5 cent/kWh electricity. This
situation may be improving because of the rise in natural gas prices since 2002 and changing
public and private preferences for more diverse sources of electricity supply, particularly from
renewable and fossil carbon-free sources. Future renewable portfolio standard (RPS) demands in
the Western states will likely drive geothermal power development, and correspondingly
geothermal may significantly contribute to the renewable energy power supply.
The two factors most frequently cited for limited geothermal development are permitting issues
and the high cost of resource exploration and development. Several proposed projects have been
severely delayed in the permitting process, and market prices have just recently been sufficient to
encourage more substantial project developments [3, 27]. Another lesser factor is the limited
number of economically feasible resource sites located near transmission grids. While market
participants do not generally cite transmission constraints as a restriction on development, the
U.S. Forest Service’s denial of permits for transmission lines and the high costs of alternate
routing have delayed several planned projects in California [3].
The current state of many aspects of geothermal technology was fairly well documented in the
1990s. The 1996 Next Generation Geothermal Power Plants (NGGPP) study, prepared for DOE,
EPRI, BPA, and SCE by The Ben Holt Company, characterizes conventional flash and binary
technology and evaluates new technologies proposed for the next generation of geothermal
power plants [13]. Prior to that study, it had been difficult to obtain current cost and performance
data for geothermal power plants because of the proprietary nature of such information. Newer
cost models still reference this document for their base case comparisons [1, 24, 30]. For HDR,
the cost estimates provided for creating a HDR reservoir, as well as its performance, were based
on assessments made by scientists at Los Alamos National Laboratory, where HDR had been
studied in previous years [11]. Table 6-4 provides an updated cost breakdown, performed by
EPRI, of four different 50-MW geothermal plants in 2009 U.S. dollars.
6-37
10581090
Geothermal Energy
Table 6-4
Estimated Costs for a 50-MW Geothermal Plant (Dec-2009 $)
Reverse
Flash Steam Binary EGS
Refrigeration Cycle
Capital Cost
Plant: $1230–$1540 Plant: $154–$1850 Plant: $1850–$2050 Plant: $1230–$2460
($/kW)
Well field: $970– Well field: $1850–

Well field: $770–$1540 Well field: $770–$1540
$2050 $6660
Breakeven
Generation 3.6 ¢–5.6 ¢ 4.6 ¢–9.2 ¢ 6.2 ¢–12.3 ¢ 7.7 ¢–16.4 ¢ +
(¢/kWh)
6.5.1 Example Geothermal Project Costs
Data specific to California and Nevada from a 2005 assessment by the California Energy
Commission [32], suggest that the capital cost of incremental generation capacity averaged about
$3,100/kW installed (see Table 6-5). For California sites alone, the average capital cost of
incremental generation capacity was somewhat lower, about $2,950/kW installed. These cost
estimates include the following components:
• Exploration (up to siting of the first deep, commercial-diameter well)
• Confirmation drilling (up to achieving 25% of required capacity at the wellhead)
• Development drilling (up to achieving 105% of required capacity at the wellhead)
• Construction of the power plant (including ancillary site facilities)
• Transmission-line costs.
Table 6-5
Estimated Geothermal Development Costs (Dec-2009 $)
Most Likely Average

Region Incremental Capacity Development Cost
(Gross MW) Per kW Installed
California 3,000 $3,024
Western Nevada 1,300 $3,536
Total 4,300 $3,178
6-38
10581090
Geothermal Energy
6.5.2 Factors that Influence Geothermal Power Plant Costs
Many factors influence the cost of a geothermal power plant. In general, geothermal power
plants are affected by the cost of steel, other metals, and labor—factors universal to the power
industry. However, drilling cost may vary as well. Geothermal projects are site-specific, thus the
cost to connect to the electrical grid varies from project to project. Also, whether the project is
the first in a particular area or reservoir will impact both risks and costs. The acquisition and
leasing of land also varies, because to fully explore a geothermal resource a developer is required
to lease the rights to 2,000 acres or more. Challenges to leasing and permitting vary from project
to project; especially on federal lands. These factors include:
• Size of the plant
• Power plant technology
• Knowledge of resource
• Temperature of the resource
• Chemistry of the geothermal water
• Resource depth and permeability
• Environmental policies
• Tax incentives and subsidies
• Markets
• Financing options and costs time delays.
6.5.3 Engineering and Economic Evaluation
Table 6-6 presents performance and cost estimates for 50- and 100-MW flash steam and 50-MW
binary-cycle geothermal power plants. The data are from the 2010 EPRI report, Engineering and
Economic Evaluation of Geothermal Power Plants [41].
Three geothermal power plant cases are addressed: 50-MW and 40-MW single-flash steam and
50-MW single-pressure binary-cycle systems. Table 6-7 summarizes the design assumptions for
the three cases.
6-39
10581090
Geothermal Energy
Table 6-6
Design Assumptions for Geothermal Plants
Design Assumption Flash Steam Binary Cycle Comments

Single is more common.
Dual flash systems have
Flash stages Single N/A
slightly higher
efficiencies and outputs.
Bottom-exhaust, dual-
50 MWn 50 MWn
Rated Capacity and flow turbine.
Configuration Top-exhaust, dual-flow
40 MWn N/A
turbine.
6 bara
Turbine inlet pressure Midrange for many new
(flash) (1 bara = 1 flash plants.
bar absolute)
Two-stage
NCG System None
hybrid system
Single pressure, Not
recuperated or considered/compared:
Type of binary cycle N/A unrecuperated; working Mixed working fluids;
fluid optimized for combined cycles; or
conditions; 50 MWn mixed cycles
Working fluids Preferred for higher
N/A Iso-Pentane
considered temperature applications
Working fluid No impurities or NCGs
N/A Pure
composition considered.
Wet Cooling
Cooling Air-Cooled Condenser
Tower
Flash based on use of
Condensate condensate and cooling
Makeup water None
from steam towers. Binary based on
air-cooled condensers.
Assumes combustible
mixture of H2S, methane,
hydrogen that might be
present in non-
condensable gases, to
H2S abatement Incinerator None
allow use of an
incinerator. Other
abatement options
would likely be more
costly.
6-40
10581090
Geothermal Energy
Table 6-7
Performance and Total Capital Requirement Estimates for Geothermal Power Plants
(December 2010 $, Source: EPRI [41])
Flash Steam 50 Flash Steam, 40 Binary Cycle 50

MWn MWn MWn
Rated Capacity
Net Capacity, kWn 50,000 40,000 50,000
Gross Capacity, kWg 54,560 43,880 57,580
Brine Flow Rate, kg/s 380 307 458
Resource Temperature, °C 280 280 280
Physical Plant
Plant Life, Year 30 30 30
2
Land Required, m 30,000 25,000 40,000
Capital Costs, $/kWg
Month/Year Dollars December, 2010 December, 2010 December, 2010
Identification Phase $15 $15 $15
Well Field $1,558 $1,595 $1,737
Gathering System $501 $504 $573
Total Plant Direct Construction Cost $1,598 $1,628 $1,650
Total Plant Indirect Construction
$786 $868 $665
Cost
Project and Process Contingency - - -
Total Plant Cost (TPC) $4,458 $4,565 $4,639
AFUDC (Interest during
$293 $300 $305
construction)
Total Plant Investment (inc.
$4,752 $4,865 $4,945
AFUDC)
Owner’s Costs, $/kWn
Other owners costs and
$238 $243 $247
contingency
Total Capital Requirement (TCR), $4,989 $5,109 $5,192
O&M Costs
Fixed, $/kW-yr 76.4 76.8 80.6
†
Nominal Variable, $/MWh 6 6 4
Makeup Drilling, $/MWh 3.6 4.5 3.6
Total Variable, $/MWh 9.6 10.5 7.6
Exergetic Efficiency, % 37.1 36.8 30.8
Capacity Factor, % 96 96 96
Design & Cost Estimate Rating Preliminary Preliminary Preliminary
†Note that this is the nominal annual VOM cost, which excludes makeup drilling.
6-41
10581090
Geothermal Energy
Table 6-8
Levelized Cost of Electricity Estimates for Geothermal Power Plants (30-year Project Life,
Constant December 2010 $, Source: EPRI [41])
Levelized Cost of Electricity ($/MWh) Flash Steam Flash Steam, Binary Cycle
50 MWn 40 MWn 50 MWn
Capital Charge @ 8.26%/yr based on
$47.7 $49.1 $52.4
Total Plant Cost
Fixed O&M $9.1 $9.1 $9.6
Variable O&M $9.6 $10.5 $7.6
Total $66.4 $68.7 $69.6
While geothermal energy is a renewable energy technology, it is not free of environmental

concerns. These concerns include land use issues in addition to emissions and chemicals that
may be entrained in the steam and must be captured and processed. Also, geothermal energy
generation has resulted in land impacts such as subsidence and induced seismicity.
6.6.1 Land Use
Land use issues are similar to those experienced by conventional power plants. Geothermal
power plants are sized according to the volume, temperature, and chemistry of the geothermal
reservoir underlying the project.
Land use patterns for geothermal power plants vary with site design and type of installation.
While a well field may cover 100 to 200 acres, heat is tapped by drilling slant wells from a single
aboveground location. A typical 50-MW plant and related well sites may occupy one or two
acres of land. In some cases, a single plant can require multiple wellhead locations. In that case,
aboveground lines carry the resource to the power plant.
A table developed for the EPRI report Greenhouse Gas Reduction with Renewables [8] and
based on land use figures in the DOE/EPRI RETC report [10], compares and quantifies the land
use requirements of the various renewable energy resources and technologies, as shown in Table
6-9. Note that geothermal energy and solar energy are more land-area-effective, measured in
kilowatts per acre, than wind and biomass. The area needed for a geothermal power plant itself is
very small, about two acres for a 50-MW plant, but the value most relevant to the comparison is
the area of the entire geothermal field, not just the site of the power plant. When doing such
comparisons of land use, two points should be kept in mind:
1. Both wind and geothermal energy production are compatible with other simultaneous uses of
much of the same land. This is especially true of wind, as shown in photographs of
agricultural land with wind turbines on “islands” within fields of crops or grazing cattle. In
the geothermal case, the pipelines to move geothermal fluids from wellhead to power plant
6-42
10581090
Geothermal Energy
and the wellhead sites themselves disrupt other land uses. In New Zealand geothermal fields,
cattle or sheep can be seen grazing on land between the well sites and pipelines (see Figure
6-17).
2. In the United States. and in many other countries as well, land is not the limiting factor in
most settings. This is especially important in considering biomass energy, where the land
comparison should be made not to the land use of other energy sources but rather to the land
used for agriculture and forestry.
Table 6-9
Land Area Requirements for Current Renewable Technologies
Area to Power Per Land Area for 100 GWe

Unit Size
Support Unit Unit of Area Area Fraction
(MWe)
Size (Acres)* (kW/acre) (106 Acres) of U.S.**
Solar Thermal 75 408 180 0.54 0.028%
PV, Residential 0.0026 0*** NA 0 0
PV, Utility-Scale 2.4 24 100 1 0.053%
Wind 50 3,707 14 7.4 0.39%
Geothermal 50 420 120 0.84 0.044%
Biomass 50 39,400 1.3 78.7 4.1%
* To convert acres to hectares, multiply by 0.4.
** Fraction is of the 1.9 billion acres of the continental U.S.
*** No land is required for residential PV systems, which are typically installed on existing structures.
6.6.2 Subsidence
Land above or adjacent to a subsurface geothermal reservoir may sink or “subside” due to the
extraction of geothermal fluid from the reservoir. Subsidence is not always easy to detect and
may be of no consequence.
Several areas of subsidence have been detected in the New Zealand fields. At Wairakei, the
oldest geothermal field in New Zealand and one that has operated at close to 170-MWe capacity
for approximately 40 years, one area of subsidence has created a pond at a low spot in a stream.
2
Another has been detected over an area of approximately 1 km and could eventually lead to
rerouting a section of roadway. A series of elevation surveys, conducted every several years
throughout the region of the Wairakei field and environs, made it possible to detect this rather
inconspicuous subsidence.
In anticipation of geothermal development expansion, biennial surveys of elevations in and near

the Imperial Valley (California) geothermal fields were performed in the 1970s. Because of a
gravity flow “ditch irrigation” system in a low desert valley and the agriculture-based economy,
the surveys were needed to ensure that no harm was caused by geothermal production and to
alert and correct the situation if adverse trends developed. Several hundred megawatts of
generating capacity have been installed and operated in the Imperial Valley over the past two
decades, and there has been no significant subsidence.
6-43
10581090
Geothermal Energy
(Evan Hughes for EPRI)
Figure 6-17
Steam Pipelines for the Ohaaki, New Zealand Plant Run Through Land Used for
Grazing and Agroforestry
6.6.3 Emissions
The environmental impacts of generating electricity from geothermal resources are benign
relative to conventional power generation options. Geothermal power generation does not
produce the federally regulated air contaminants commonly associated with other power
generation options, including sulfur dioxide, particulates, carbon monoxide, hydrocarbons, and
photochemical oxidants. Some hydrothermal fluids contain hydrogen sulfide and/or high levels
of dissolved solids such as sodium chloride (salt). Hydrogen sulfide emissions are abated, when
necessary, with environmental control technology. Emission factors are typically in the range of
45 kg/MWh for carbon dioxide and 0.0145 kg/MWh for hydrogen sulfide. Hydrogen sulfide is
now routinely abated at geothermal power plants, resulting in the conversion of over 99.9% of
the hydrogen sulfide from geothermal non-condensable gases into elemental sulfur, which can be
converted to a non-hazardous fertilizer feedstock. Since 1976, hydrogen sulfide emissions have
declined from 1,900 lbs/hr to 200 lbs/hr or less, while at the same time the geothermal power
production has increased from 500 MW to over 2,500 MW. Another hazardous element found in
a few geothermal plants is mercury. Although mercury is not present in every geothermal
resource, where it is present mercury capture equipment typically reduces emissions by 90% or
more. The relatively “highest” mercury emitters are two facilities at The Geysers in California.
They release mercury at levels that do not trigger any health risk analyses under California
regulations.
6-44
10581090
Geothermal Energy
The visible plumes seen rising from some geothermal power plants are in fact water vapor
(steam). A case study by GEA [27] showed that a coal plant updated with scrubbers and other
emissions control technologies emits 24 times more carbon dioxide, 10,837 times more sulfur
dioxide, and 3,865 times more nitrous oxides per megawatt-hour than a geothermal plant.
Averages of four significant pollutants emitted by geothermal and coal facilities are presented in
Table 6-10 below. Following the table is a brief discussion of other emissions that have
sometimes been associated with geothermal development. Issues revolving around groundwater
contamination are avoided through protective well completion practices. The geothermal
industry has developed various methods to mitigate chemical emission concerns.
Table 6-10
Averages of Four Significant Pollutants, as Emitted from Geothermal and Coal Facilities,
are Listed in the Table Below (Source: GEA 2005 [27])
Nitrogen Sulfur Dioxide Particulate Carbon

Emission
Oxides (NOX) (SO2)* Matter (PM) Dioxide (CO2)
Geothermal Emissions (lb/MWh) 0 0 – 0.35 0 0 – 60
Coal Emissions (lb/MWh) 4.31 10.39 2.23 2191
Emissions Offset by Geothermal

32,000 tons 78,000 tons 17,000 tons 16 million tons
Use (per yr)
* While geothermal plants do not emit sulfur dioxide directly, once hydrogen sulfide is released as a gas into the
atmosphere, it eventually converts to sulfur dioxide and sulfuric acid. Therefore, any sulfur dioxide emissions
associated with geothermal energy derive from hydrogen sulfide emissions.
Generally, there is less likelihood of adverse environmental impacts from binary-cycle

generation than from flash steam generation because the hotter geothermal fluids (steam and
water) used in flash plants tend to contain greater concentrations of chemical contaminants than
do the cooler geothermal fluids (water) typically used in binary plants. In general, hotter water
will dissolve more mineral from a rock formation than will cooler water. Also, in binary plants
that employ dry rather than wet cooling systems, the geothermal fluid remains in a closed system
and is never exposed to the atmosphere before it is injected back into the reservoir. The potential
environmental impact of a HDR binary system is likely to be even lower. This is because the
water used in such a system originates from a shallow groundwater well or other source with low
levels of dissolved solids and no hydrogen sulfide to begin with, and because the limited contact
time of the injected surface water with the hot rock will in most cases limit the level of dissolved
mineral (including dissolved gas) to below that of chemical equilibrium.
In some situations, geothermal operations offer novel benefits to the environment. For example,
operators of The Geysers complex in northern California are injecting treated municipal
wastewater into underground steam-generating reservoirs whose natural stores are being
depleted. This technique helps meet the generation facility’s resource challenges in addition
to mitigating the waste disposal problems of nearby communities.
6-45
10581090
Geothermal Energy
6.6.4 Thermal Discharge
Geothermal power plants are intrinsically less efficient than fossil-fuel-fired power plants. This
is because high-temperature steam can be converted into work at higher thermal efficiency than
low-temperature steam. Temperatures of steam produced in fossil fuel boilers are in the range of
800–1000°F (400–500°C), much higher than the 300–600°F (150–300°C) of geothermal steam
and hot water. Less heat converted to work and electricity means more heat “wasted,” much of
which is discharged into the environment. Fossil fuel steam plants have thermal efficiencies in
the 32% to 38% range, nuclear plants close to 30% to 32%, and geothermal plants as low as 5%
to 17%.
For this reason, geothermal power plants have very large cooling systems relative to other power
generation options. The Ohaaki Power Station in New Zealand installed a natural-draft cooling
tower, one that appears comparable to those seen at nuclear power plants and at some fossil fuel
power plants located on rivers (Figure 6-6). The reason for the cooling tower at Ohaaki is the
same as that at U.S. nuclear or fossil plants: to avoid the discharge of relatively large amounts of
waste heat into a stream, river, or estuary of limited capacity and the resulting increased
temperature of water returned to the source.
In many geothermal locations, water is scarce and air cooling is the only option. Air cooling can
be done by using fans or natural-draft cooling towers. The fan-driven cooling system can be built
quickly and at relative low cost, but its operating costs are high due to maintenance costs and
parasitic losses associated with the fans. Innovative cooling concepts are being developed to
improve the efficiency of heat exchangers and reduce the capital cost of natural cooling towers
by using hybrid air condensers and storing cool capacity (as either sensible or latent storage) to
extend night time cooling through the day.
6.6.5 Induced Seismicity
Seismicity has been correlated with underground fluids extraction and injection. Seismic activity
has steadily increased as production has increased at the Geysers geothermal field in northern
California. In most geothermal fields, fluids are being withdrawn and injected in the same
region, thermal and chemical effects are also occurring, and all of these effects can combine to
produce a variety of mechanisms for inducing seismicity.
Water injection into geothermal systems, which has become a nearly universal and often required
strategy for extended and sustained production of geothermal resources, seems to be a common
cause of induced seismicity. However, the data available do not support the hypothesis that
induced seismicity in geothermal regions is caused by injection.
In many geothermal fields and potential geothermal fields in the United States, the potential for
induced or naturally occurring seismicity (less than grade 5) should not cause significant
damage. This is the case for two reasons:
1. Faults are not close enough to the geothermal plants to create a large damaging event
2. If faults are nearby, geothermal production and injection activities are occurring at depths
less than 5 km, which is shallower than most seismic activity (which tends to occur at depths
of between 5 and 10 km).
6-46
10581090
Geothermal Energy
Overall, the impact of induced seismicity on the implementation of various different enhanced
geothermal activities will depend on the risk associated with the activity and the cost-benefit
ratio. All experience to date has shown that the risk, while not zero, has been either minimal or
can be handled in a cost-effective manner.
6.7 Design, Deployment, and O&M Issues
Many geothermal developers acquire the rights to a geothermal resource and rely on service
companies for exploration, drilling, well testing, reservoir engineering, and power plant design
and construction. However, most developers handle their own power plant operation.
For domestic power projects, coated corrosion-resistant pipe and heat exchangers are typically
obtained from U.S. manufacturers. However, virtually all steam turbines are purchased from
Japanese manufacturers. A few U.S. companies manufacture their own small binary and hybrid
power plant equipment. One important developer, Ormat, is the U.S. arm of an Israeli company
that uses the Israeli Ormat assembled, skid-mounted, and modular equipment.
Most of the necessary drilling and well-testing equipment is adapted from the oil and gas
industry. However, some of this equipment has been specifically refined for the hotter and more
corrosive environments of geothermal drilling. While DOE’s geothermal research and
development program is working on geothermal-specific improvements to some power plant
equipment, in general the pipe, heat exchangers, turbines, and cooling towers come from the
fossil fuel power plant industry.
Technology is expected to improve incrementally over the coming years. Industry is working
with DOE’s geothermal R&D program to improve exploration, drilling, reservoir engineering,
and energy conversion technology. R&D success in fracture detection and fracture permeability
studies could give the industry important new tools to improve geothermal exploration and
reservoir engineering effectiveness, and also drive down project costs. As discussed above under
EGS, these fracture and permeability enhancements of natural hydrothermal hot water reservoirs
could be the start of commercial activity in the hot dry rock field. Better materials and coatings
for power plant equipment are being researched to reduce high maintenance costs. On the energy
conversion front, R&D also continues on binary geothermal systems that use a secondary
working fluid to produce energy from the more abundant moderate- to low-temperature
geothermal resources.
6.7.1 Industry Trends
Geothermal project activity is escalating significantly after relatively slow growth over the past
two decades. The global geothermal pipeline exceeds 9 GW for projects under development,
equivalent to approximately 80% of total global installed capacity built over the past three
decades. The recent renaissance is a reflection of geothermal's successful track record and
performance and cost attributes. Even following the significant downturn in global energy prices
during the end of 2008 and 2009, there remain great expectations for geothermal's long-term cost
competitiveness as global demand for energy escalates. Indeed, the global installed geothermal
capacity is forecasted to more than triple its current installed base to over 31 GW by 2020.
6-47
10581090
Geothermal Energy
Today, there are over 215 commercial geothermal projects operating in 24 countries with a
cumulative installed capacity of approximately 10.5 GW worldwide. The United States is
currently the world’s top geothermal market, with more than 3 GW of installed capacity. With a
pipeline of more than 4.4GW of confirmed projects, the United States is poised to more than
double existing capacity over the next five years. U.S. geothermal activity will be primarily
concentrated in a few Western states that are home to vast hydrothermal resources. Developers
are expected to focus on "low-hanging fruit" opportunities located in California and Nevada.
Internationally, Indonesia and the Philippines are poised to ramp up their geothermal capacity in
the near term. Meanwhile, market activity in Japan, Iceland, Italy, New Zealand, and Mexico,
which account for over 65% of the remaining global installed geothermal capacity, should also
increase. Although these markets have seen relatively limited growth over the past few years,
urgency to advance low-carbon baseload power generation is helping to re-ignite new capacity
growth in these markets.
Separately, attention is turning to new markets. Chile, Turkey, Russia, Nicaragua, Germany,
Australia, and East Africa are all poised to initiate projects as greater effort is made to integrate
renewables, diversify energy portfolios, and hedge against natural gas price volatility.
6.7.2 Research Needs and Recommendations
While geothermal is considered a mature technology, there are many opportunities for
improvement through research and development. Geothermal projects are very capital intensive.
Discussions with industry leaders suggest that that future geothermal R&D programs should
focus on reducing the cost of finding and developing new geothermal resources.
The industry consensus is that characterizing the commercial potential of identified geothermal
reservoirs is a high priority. Techniques such as fracture mapping, more accurate thermal-
gradient wells, and other, untested methods should be evaluated and refined, if appropriate. The
objective is to be able to measure the temperature, fluid characteristics, and permeability of the
resource prior to committing to expensive production wells and generation equipment.
Industry comments suggest that structured, integrated tests and experiments by industry in
collaboration with the California Energy Commission and DOE are essential to ensure further
development of commercial geothermal power. In the near term, this work should focus on the
following areas:
• Reducing life-cycle costs of electrical power generation
• Improving geothermal facility efficiencies
• Finding techniques to more cost-effectively define resources
• Supporting funding to reduce production drilling costs and develop direct-use demonstration
and commercial projects
• Improving technology and technical support for geothermal direct-use projects
• Reducing geothermal heat pump (GHP) initial costs
6-48
10581090
Geothermal Energy
• Improving professional training and consumer education campaigns for GHP

• Supporting financing programs for residential applications for GHP
• Understanding and improving wastewater injection
• Improving public awareness about the benefits of geothermal power, direct-use, and GHP.
Enhanced geothermal systems are still widely experimental. A 2008 DOE survey identified
three critical assumptions for EGS technology that require evaluation and testing before the
economic viability of EGS can be confirmed:
• Demonstration of commercial-scale reservoirs: Stimulation and maintenance of a large
volume of rock is required in order to minimize temperature decline in the reservoir. Actual
stimulated volumes have not been reliably quantified in previous work.
• Sustained reservoir production: Recent studies conclude that 200°C fluid flowing at 80
kg/sec (equivalent to about 5 MWe) is needed for economic viability. No EGS project to date
has attained flow rates in excess of approximately 25 kg/sec.
• Replication of EGS reservoir performance: EGS technology has not been proven to work at
commercial scales over a range of sites with different geologic characteristics.
These assumptions can be tested with multiple EGS reservoir demonstrations using current
technologies. In the long term, significant reduction in drilling costs will be necessary to access
deeper resources, and the cost of converting geothermal energy into electricity must be reduced.
These improvements will move EGS technology forward as an economically viable means of
tapping the nation’s geothermal resources.
6.8 Key Research Participants
The following includes brief descriptions of government and non-governmental organizations

involved in geothermal energy research. Note that while the websites for these organizations
contain information and contacts that may be of interest, EPRI cannot vouch for the accuracy or
quality of all their content.
DOE Geothermal Energy Program

The DOE administers its geothermal energy efforts through its Energy Efficiency and Renewable
Energy Network.
www.eere.energy.gov/geothermal.
NREL Geothermal Technologies Program

This National Renewable Energy Laboratory program supports activities of DOE’s Geothermal
Energy Program and works with DOE and industry partners in geothermal R&D efforts.
www.nrel.gov/geothermal/.
6-49
10581090
Geothermal Energy
Sandia Geothermal Research Department

Sandia National Laboratory hosts a Geothermal Research Department focused on developing
well construction technologies to reduce the cost of geothermal wells. Hard rock drill bit
technology, high-temperature instrumentation, rig instrumentation, and data management are
among the projects it pursues.
www.sandia.gov/geothermal/.
Idaho National Engineering and Environmental Laboratory

This laboratory’s Geothermal Energy website serves as a repository for some DOE technical data
related to geothermal energy, including maps and graphics, information about U.S. laws and
standards, annual research program summaries, proposal solicitations, and other information.
http://geothermal.id.doe.gov/.
Stanford University Geothermal Program

Stanford’s geothermal program has conducted geothermal reservoir engineering research for
more than 20 years, completing some 150 projects. Study areas have included well test analysis
of fractured and multiphase reservoirs, design and interpretation of tracer tests in fractured
reservoirs, experimental measurements of fluid flow parameters, and optimization of production
and reinjection strategies.
http://ekofisk.stanford.edu/geotherm.html.
Southern Methodist University

SMU has the most up-to-date data currently available for resource assessment based on the 1978
USGS data and other information collected since then.
www.smu.edu/geothermal/.
Geocrack
This website provides information on the computer modeling of fluid flow in geothermal
reservoirs, including technical papers, progress reports, and downloadable software.
www.mne.ksu.edu/~geocrack/.
European Deep Geothermal Energy Programme

This website describes a hot dry rock research program funded by a consortium of European
government, research, and utility organizations in Soultz-sous-Forêts, France. The multiyear
project aims to develop a scientific pilot plant by 2006 and a subsequent commercial plant that
could produce approximately 25 MW.
www.soultz.net.
Geothermal Energy Association

GEA is a trade association composed of U.S. companies that support the expanded use of
geothermal energy and are developing geothermal resources worldwide for both electric power
generation and direct-heat uses.
www.geo-energy.org.
6-50
10581090
Geothermal Energy
Geothermal Resources Council

The GRC is a nonprofit educational association formed in 1970 to encourage geothermal
research and development, and to serve as an informational conduit among researchers, industry,
and the public.
www.geothermal.org.
International Geothermal Association

Based in Italy with members in 63 nations, IGA’s objective is to encourage research,
development, and utilization of geothermal resources worldwide by compiling, publishing, and
disseminating scientific and technical information.
http://iga.igg.cnr.it/index.php.
Center for Renewable Energy and Sustainable Technology

The Center for Renewable and Sustainable Technology (CREST) provides excellent background
information and extensive links to other websites. CREST has been a part of the Renewable
Energy Policy Project since 1999.
http://www.repp.org/geothermal/index.html.
Calpine Corp.
Founded in 1984, Calpine operates The Geysers in northern California and is the largest
producer of geothermal electricity in the world.
www.calpine.com.
Unocal Geothermal
Unocal operates three of the world’s largest geothermal electricity projects, Tiwi and Mak-Ban
in the Philippines and Gunung Salek in Indonesia, which together have a combined installed
generating capacity of 1,100 MW. Unocal merged with Chevron Corporation in 2005.
http://www.chevron.com/deliveringenergy/geothermal/.
The World Bank

The World Bank provides a source of financial and technical cost data for geothermal
development in countries around the world.
http://econ.worldbank.org.
GeothermalBiz.com
Geothermal Biz website provide general information on state government agencies, for example
state energy offices, relevant legislation, e.g., for renewable portfolio standards (RPS) and public
benefits funds. Also included is information on state, local, and utility incentives that promote
geothermal energy such as tax incentives, grants, loans, rebates, and utility green pricing
programs.
http://www.geothermal-biz.com.
6-51
10581090
Geothermal Energy
6.9 References
1. Renewable Energy Annual 2003. Contains data for 2002 and is also available on the web at
http://www.eia.doe.gov/cneaf/solar.renewables/page/rea_data/rea_sum.html, EIA,
Washington, D.C.: December 2004. DOE/EIA-0603(99).
2. Huttrer, G., “The Status of World Geothermal Power 1995-2000.” A paper prepared for the
World Geothermal Conference and the DOE Geothermal Power Program, as an update of
the 1995 status report on geothermal power deployment, based on the “Country Update”
papers submitted to the World Geothermal Conference held in Japan, May-June 2000.
Paper prepared by Gerald W. Huttrer of Geothermal Management Co., Frisco, Colorado:
May 2000.
3. Market Assessment of Renewable Energy in California, EPRI report to the California Energy
Commission. EPRI, Palo Alto, California: July 2001. 1001193.
4. Duffield, W.A., Sass, J.H., and Sorey, M.L., “Tapping the Earth’s Natural Heat,” U.S.
Geological Survey Circular 1125, 1994.
5. Assessment of Geothermal Resources of the U.S. – 1978. Editor: L.J.P. Muffler, USGS
Circular 790, 1978.
6. Geothermal Energy in the Western U.S. and Hawaii: Resources and Projected Electricity
Generation Supplies, EIA: September 1991. DOE/EIA-0544.
7. Annual Energy Outlook 2007, EIA: February 2007.
8. Greenhouse Gas Reduction with Renewables. EPRI, Palo Alto, California: December 2000.
TR-113785.
9. Economic Predictions for Heat Mining: A Review and Analysis of Hot Dry Rock (HDR)
Geothermal Energy Technology, by Tester, J.W., and H.J. Herzog, Massachusetts Institute
of Technology: July 1990.
10. Renewable Energy Technology Characterizations, EPRI and DOE. EPRI, Palo Alto,
California: December 1997. TR-109496.
11. Brown, D. and Duchane, D., Los Alamos National Laboratory, personal communication
to Lynn McLarty (one of the authors of the RETC [10] above), November 5, 1996.
12. Assessment of Geopressured Energy Resources on the Gulf Coast of the U.S. Report to
EPRI by Southwest Research Inst. (San Antonio, TX). EPRI, Palo Alto, California: 1983.
13. Brugman, J., Hattar, M., Nichols, K., and Y. Esaki, Next Generation Geothermal Power
Plants. EPRI: February 1996. TR-106223.
14. Proceedings of the 10th EPRI Geothermal Conference and Workshop, San Diego, June 1986.
EPRI, Palo Alto, California: December 1986. AP-5059-SR.
15. Armstead, H.C.H. and Tester, J.W., Heat Mining: A New Source of Energy, E.&F.N.
Spon Ltd. University Press, London: 1987.
16. Mason, T.R., “Improving the Competitive Position of Geothermal Energy,” Proceedings of
Geothermal Program Review XIV, DOE: April 1996. Report DOE/EE-0106.
6-52
10581090
Geothermal Energy
17. Heber Binary Cycle Geothermal Demonstration Power Plant, EPRI, Palo Alto, California:
August 1998. AP-5787-SR. (Note: This was the last in a series of several reports published
by EPRI during the development and implementation of the Heber Project between 1978
and 1988.)
18. Los Alamos National Laboratory. “Proceedings of the 3rd International Hot Dry Rock
Energy Forum,” Santa Fe, New Mexico: May 13-16, 1996.
19. EPRI Technical Assessment Guide Electricity Supply Volume 1, Rev. 7. EPRI, Palo Alto,
California: June 1993. TR-102275-V1R7.
20. Annual Energy Outlook 1996, EIA: August 1996. DOE/EIA-0603(96).
21. Entingh, D.J., Technology Evolution Rationale for Technology Characterizations of U.S.
Geothermal Hydrothermal Electric Systems, BNF Technologies Inc., Alexandria, Virginia,
Draft Report: March 1993.
22. Renewable Power Industry Status Overview. EPRI, Palo Alto, California: December 1998.
TR-111893.
23. Annual Energy Outlook 1998 with Projections Through 2020, EIA, Office of Integrated
Analysis and Forecasting, Washington, D.C. 20585: December 1997. DOE/EIA-0383 (98),
Distribution Category UC-950.
24. Annual Energy Outlook 2002 with Projections Through 2020, EIA, Office of Integrated
Analysis and Forecasting, Washington, D.C. 20585: December 2001.
25. Hunt, Trevor, geophysical and geothermal consultant in Taupo, New Zealand. Personal
communication to Evan Hughes, October 2003.
26. Radon as an In Situ Tracer in Geothermal Reservoirs, Project 1992-1, Final Report to
EPRI by Professor Paul Kruger of Stanford University: August 1987. EPRI AP-5315.
27. A Guide to Geothermal Energy and the Environment, Geothermal Energy Association
by Alyssa Kagel, Diana Bates, and Karl Gawell, Earth Day 2005.
28. World Geothermal Generation 2001-2005 State of the Art by R. Bertani, 2005. Prepared for
World Geothermal Congress 2005 Turkey.
29. Classification of Geothermal Systems – A Possible Scheme by Subir K. Sanyal, 2005.
Prepared for Thirtieth Workshop on Geothermal Reservoir Engineering, Stanford University,
Stanford, California: January 31-February 2, 2005. SGP-TR-176.
30. Cost of Geothermal Power and Factors that Affect It by Subir K. Sanyal, 2005.Prepared for
Thirtieth Workshop on Geothermal Reservoir Engineering, Stanford University, Stanford,
California: January 26-January 28, 2005. SGP-TR-175.
31. Geothermal Bulletin Volume 33 Number 3, June 2004. Geothermal Resource Council.
32. New Study Highlights Geothermal Resources Available for Development in California and
Nevada by Jim Lovekin, December 2004. GRC Bulletin p. 242.
33. New Geothermal Site Identification and Qualification: April 2004. CEC Publication No.
P500-04-051.
34. Bertani, Ruggero “World Geothermal Generation in 2007,” GHG Bulletin, September 2007.
6-53
10581090
Geothermal Energy
35. U.S. Geothermal Power Production and Development Update, March 2009, Geothermal
Energy Association.
36. The Future of Geothermal Energy: Impact of Enhanced Geothermal Systems (EGS) on the
st
United States in the 21 Century, Massachusetts Institute of Technology, 2006.
37. An Evaluation of Enhanced Geothermal Systems Technology, Geothermal Technologies
Program, Energy Efficiency and Renewable Energy, DOE, 2008.
38. S.K. Sanyal et al. "Future of Geothermal Energy," Geothermal Energy Conference, Canadian
Geothermal Energy Association, April 2009.
39. Bertani, Ruggero "Long Term Projections of Geothermal-Electric Development in the
World," GeoTHERM Conference – Offenburg, March 2009.
40. “Brown, Donald W., “A Hot Dry Rock Geothermal Energy Concept Utilizing Supercritical
CO2 Instead of Water” PROCEEDINGS, Twenty-Fifth Workshop on Geothermal Reservoir
Engineering, Stanford University, Stanford, California, January 24-26, 2000.
41. Engineering and Economic Evaluation of Geothermal Power Plants. EPRI, Palo Alto, CA:
2010 1019761.
6-54
10581090
7
SOLAR THERMAL
Trough Technology
Installed Capacity (est.) • 950 MW worldwide (as of September 2010)
• 421 MW in U.S.
Technology Readiness • Commercial
• Moderate confidence in cost estimates and technology projections
Environmental Impact • Minimal during manufacturing and operation
• Significant land usage may impact species
• Water usage concern in some desert locations; plants may use dry
cooling
• Synthetic oil heat transfer fluid is a volatile organic compound (VOC)
Economic Status • Competitive in some markets with high government subsidies, aided in
United States by DOE loan guarantee.
• Resurgence in new system construction after two-decade layoff; lost
manufacturing capability and experience being regained
Policy Status • Along with other solar technologies, encouraged by RPS and energy
policies in most jurisdictions
Trends to Watch • Hybrid operation and thermal storage
• New solar trough projects in Arizona, Nevada, California, Spain, Algeria,
Egypt, Italy, Morocco, and elsewhere, with several nearing completion in
2010–2011.
• DOE loan guarantees backed by ARRA stimulus funding are providing
significant incentive in the United States.
• New materials and techniques to reduce manufacturing and O&M costs
• Lower-cost mirror support structures
• Evident interest in non-glass mirrors and reflectors even while glass
manufacturers are emerging to support the solar thermal market.
• Interest in direct steam generation remains high.
• Research continues on advanced, alternative heat transfer fluid
materials, including molten salts that remain liquid over a wider range of
temperatures.
7-1
10581090
Solar Thermal
Central Receiver Technology

Installed Capacity (est.) • 45 MW worldwide (as of September 2010)
Technology Readiness • Demonstration
Moderate-low confidence in cost estimates and technology projections
Environmental Impact • Minimal during manufacturing and operation
• Most modern designs use steam or molten salt for heat transfer and
storage, which are easily contained if spilled
• Water usage concern in some desert locations; plants may use dry
cooling
Economic Status • Potentially high financing costs for first utility-scale systems, mitigated in
the United States by DOE loan guarantees.
• Potential for lower cost with higher operating temperatures.
• Long-term performance remains to be demonstrated, with significant
uncertainty about component and O&M costs
Trends to Watch • Large projects planned in the U.S. and Europe
• Thermal storage to allow dispatchable solar power; value of storage
depends on whether local time-of-use rates and other pricing and policy
circumstances make it worth the investment.
• Molten salt and direct steam projects in Spain, Nevada, and California.
Dish/Engine Technology
Installed Capacity (est.) • 2.5 MW worldwide
Technology Readiness • Early demonstration
• Low confidence in cost estimates and development projections
Environmental Impact • Minimal during manufacturing
• None in operation
Economic Status • Very dependent on the success of demonstration projects. May become
competitive with success of SCE/SDG&E projects.
• High evident risk may be mitigated in United States by DOE loan
guarantees.
• Long-term performance remains to be demonstrated, with component
and O&M costs largely uncertain.
policies in most jurisdictions.
Trends to Watch • Future large-scale demonstration projects following successful
deployment of 1.5-MW system in 2010, including large power purchase
agreements with SCE and SDG&E.
7-2
10581090
Solar Thermal
• Increasing field experience leading to better understanding of long-term

performance and O&M costs.
• R&D to adapt thermal energy storage to dish/engine system.
• New materials and techniques to reduce manufacturing and O&M costs.
Compact Linear Fresnel Reflector Technology

Installed Capacity (est.) • 9.4 MW worldwide (as of September 2010)
Technology Readiness • Early engineering and demonstration
• Low confidence in cost estimates and development projections
Environmental Impact • Minimal during manufacturing
• None in operation
• Water usage concern in some desert locations
Economic Status • Successful early demonstrations suggest economic competitiveness in
some markets and applications.
• The 2009 acquisition of Ausra by large French company Areva, and
Areva’s stated intent to make CLFR the centerpiece of its solar energy
business, may bode well for future investment and prospects. In addition,
other corporations are expressing quiet interest in developing CLFR
projects.
• Long-term performance and reliability remains to be demonstrated, with
great uncertainty about component and O&M costs.
• DOE loan guarantees may encourage new project proposals.
Trends to Watch • New demonstration projects and smaller scale commercial plants.
• Advanced steam storage technology
• SkyFuel plans for molten salt as heat transfer fluid and storage media,
supported by DOE research grant.
7.1 Introduction
Solar thermal electric power technology has received substantial development attention since the
early 1970s. Government energy research programs in the developed countries have invested
several billions of dollars (U.S. equivalent) over the past three decades aimed at its advancement
and commercial introduction. U.S. development efforts were particularly strong during the
1980s, and resulted in deployment of a significant amount of experimental and commercial
hardware during that decade.
Attention subsided in the 1990s primarily because conventional energy prices seemed to stabilize
at affordable levels. In addition, the primary interest of the solar power community shifted from
large-scale wholesale power to smaller-scale retail power, due in part to the emerging electric
7-3
10581090
Solar Thermal
sector restructuring movement. Key elements of that movement included increased emphasis on
dispersed, distributed generation and retail competition for electricity supply. In addition,
development issues had surfaced for the solar thermal options, and the resources to resolve those
issues were not forthcoming because of the market’s characteristics at the time. Nonetheless,
several major field deployment and advancement activities were carried out in the 1990s,
resulting in important new contributions that enhanced the prospects for these technologies.
Driven by looming capacity shortfalls, the new century found the restructuring movement
slowing and industry interest in central-station power plants increasing. Growing public interest
in renewable energy, fueled by pending carbon emissions constraints and renewed concerns over
energy prices and security, has put solar thermal electric generation back into serious
consideration. In the past few years, there have been indications that the time has come for
significant new investments in solar thermal. New plants are now operating in the United States,
Spain and elsewhere, with hundreds of more megawatts of solar thermal capacity on the drawing
boards. Real-world operating experience is bolstering confidence that solar thermal plants can
run reliably, and will further understanding of their engineering challenges, operating issues, and
O&M costs as such data are made public. Anecdotal evidence says that industry experience with
plants that have opened has been generally favorable.
7.1.1 Solar Thermal Technologies
There are three common types of solar thermal power systems: parabolic trough, central receiver
or power tower, and dish/engine. In addition, there is a relatively new “compact linear Fresnel
reflector” technology, which is being tested at a 5-MW demonstration that began in late 2008.
Another technology, “solar chimney,” has been proposed but has not yet been demonstrated for
electricity production at large scales. Except for the solar chimney, all of these technologies
involve a heat-driven engine and most can be readily hybridized with fossil fuel. Some can be
adapted to use thermal storage. The primary advantage of hybridization and thermal storage is
that the technologies can provide dispatchable power and operate during periods when solar
energy is not sufficient. Thus, hybridization and thermal storage can enhance the economic value
of the electricity produced.
Parabolic trough systems use banks of trough-shaped mirrors with a parabolic cross-section to
focus sunlight on highly absorbing receiver tubes that contain a heat-transfer fluid (HTF). This
fluid, typically a synthetic oil, is heated and pumped through a series of heat exchangers to
produce superheated steam that powers a conventional turbine generator to produce electricity.
Nine trough systems, built in 1984 to 1990, are generating 354 MW in southern California.
These 14- to 80-MW systems are hybridized to derive up to 25% of their output from natural-gas
firing and provide dispatchable power independent of the solar energy available. In addition, a
1-MW trough system, commissioned in 2006, is operating in Arizona, and a 64-MW trough
facility in Nevada became operational in early 2007. The first of several Spanish 50-MW trough
21
plants was completed in late 2008 and nine such plants are now operational. The Spanish plants
are especially notable because several include 7.5 hours or more of molten-salt thermal energy
storage.
21
Solnova 1, 3 and 4; Andasol 1 and 2; Ibersol; Alvarado 1; Extresol 1; and LaFlorida.
7-4
10581090
Solar Thermal
Central receiver systems, also referred to as power towers, use a field array of large mirrors
called “heliostats” that track the sun and focus its light onto a central receiver mounted on top of
a tower. The first central receiver in the United States, Solar One, was installed in southern
California and operated in the mid-1980s. It used water as the HTF and generated steam directly
to power a 10-MW steam turbine. In 1992, a consortium of U.S. utilities, DOE, and EPRI formed
to retrofit Solar One and demonstrate a system, aptly named Solar Two, which incorporated
molten salt as the HTF and energy storage medium. In this system, molten salt was pumped from
a “cold” tank and cycled through the receiver, where it was heated and returned to a “hot” tank.
The hot salt could then be used to generate electricity when needed. Current designs allow
storage times ranging from 3 to 13 hours. Such thermal storage capability makes central
receivers (and troughs, when so equipped) the most flexible of solar technologies, promising
dispatchable power with load factors of up to 65%—in principle, even 100%. Solar Two
operated from April 1996 to April 1999. The EU’s first commercial 11-MW central receiver
project, PS10, became operational in Spain in August 2007, and the second, PS20, began
operation adjacent to PS10 in April 2009. Also in Spain, the 17-MW Gemasolar central receiver
project is scheduled to begin operation in April 2011. Meanwhile, a number of new U.S. and EU
power tower projects are being proposed for construction over the next five to seven years.
Dish/engine systems use an array of mirrors made from glass facets to form a parabolic dish that
focuses solar energy onto a receiver located at the focal point of the dish. A HTF, typically
helium or hydrogen, is heated in the receiver tube and used to generate electricity in a small
engine attached directly to the receiver. Current designs employ a Stirling engine, but future
designs could use Brayton-cycle (turbine) engines or dense arrays of high-efficiency
photovoltaic cells. Stirling Energy Systems (SES), which is working with Sandia National
Laboratories on a six-dish 150-kW demonstration near Albuquerque, New Mexico, signed
separate power purchase agreements (PPAs) in 2005 with Southern California Edison and San
Diego Gas & Electric for up to 1,750 MW. Delivery of the first units is scheduled for 2012.
Meanwhile, the 60-dish, 1.5-MW Maricopa dish/engine plant in Arizona began providing
electricity to Salt River Project in January 2010.
Long-term cost projections for trough technology are higher than those for power towers and
dish/engine systems due in large part to the lower solar concentration involved, which results in
lower temperatures and efficiency. However, with more than 20 years of operating experience,
continued technology improvements, and O&M cost reductions, troughs are likely to be the least
expensive, most reliable solar thermal electric technology for near-term deployment.
The more novel compact linear Fresnel reflector (CLFR) and solar chimney designs are not
yet as well understood as the first three technologies above. Therefore, they are only briefly
described in this section, as no significant economic details are yet known.
7.1.2 Hybridization
All solar thermal technologies have the potential to operate as hybrids, with collected solar
energy introduced into a gas turbine or coal boiler system in concert with conventional fossil
fuels or other sources of heat. For example, a parabolic-trough solar field can be integrated with
a conventional combined-cycle plant to make up an integrated solar combined-cycle (ISCC)
system. Such hybridization can effectively increase the dispatchability of solar electricity and
7-5
10581090
Solar Thermal
thereby minimize grid integration issues. For retrofit applications, this approach is limited to
receiving about 10%–20% of the total heat input from sunlight due to the limitation of the
existing equipment to accommodate larger amounts of solar. Central-receiver solar fields are
better suited to hybridization with natural gas combined-cycle and Rankine-cycle plants, and can
potentially accommodate higher levels of solar input due to the higher steam temperatures that
more closely match the existing steam turbine conditions. Solar thermal hybridization using
CLFR technology for feedwater heating at a coal plant has been demonstrated at only small scale
to date, but over a dozen hybrid plants, mostly ISCC, are planned or under construction.
As originally proposed by Luz Solar International, the ISCC configuration consists of a

parabolic-trough field, a combined-cycle power plant fueled by natural gas or other fossil fuel,
and a solar steam generator. During solar-powered operation, water from the cooler end of the
steam cycle is diverted and heated into saturated steam by the solar steam generator. The
saturated steam is then returned to the heat recovery steam generator (HRSG), where the
combined fossil- and solar-heated steam flow is then superheated by fossil-fueled heating. Other
integration options may be feasible depending on the steam conditions available from the solar
field. When sunlight is unavailable, an integrated plant can operate as a conventional fossil-
fueled combined-cycle facility.
Advantages of solar thermal hybrid plants include the high efficiency with which they convert
solar energy to electrical energy compared to stand-alone solar plants, the relatively low
incremental unit cost for the larger steam turbine in an integrated plant compared to a solar-only
plant, and reduced thermal inefficiencies associated with the daily start up and shut down of the
steam turbine. Ten parabolic-trough projects up to 75 MW in size in the United States, Algeria,
Egypt, Mexico, Spain and Morocco plan to use the ISCC cycle. An additional project in Spain
plans to use CLFR technology. Other projects in the 20- to 50-MW range are being studied in
Australia, Italy, China, India, South Africa, Greece and Algeria.
A novel hybridization option well-suited to central-receiver plants is one in which a pressurized

volumetric air receiver contributes heat energy to the compressed air of a gas turbine before it
enters the combustor. Additional heat input from conventional combustion is also required. A
receiver module, consisting of a secondary concentrator and a volumetric receiver unit, was tested
at Spain’s Plataforma Solar de Almeria, where air temperatures up to 1000°C and receiver
efficiencies of more than 70% were achieved. This European Commission-funded project’s
participants included ORMAT in Israel, CIEMAT in Spain, DLR in Germany, Solucar in Spain,
and others. It was conducted in two stages, termed REFOS and SOLGATE22. This concept
involves very high operating temperatures and the need to use window materials such as quartz
to contain pressurized air while admitting solar flux. This may limit the practical size of such
units to a few tens of megawatts.
22
See www.ec.europa.eu/research/energy/pdf/solgate_en.pdf for the 2005 European Commission report “Solgate:
Solar Hybrid Gas Turbine Electric Power System,” which summarizes recent work in this area.
7-6
10581090
Solar Thermal
7.2 Resources
This section of the Renewable Energy Technology Guide addresses solar resource assessment,
technology overview and status for utility-scale solar thermal electric applications. It also
discusses likely development pathways, performance and cost history and projections, operating
and maintenance labor requirements, environmental issues, and potential for greenhouse gas
reduction.
The solar energy resource at a given location is characterized by the solar radiation per unit area
(or “insolation”) expressed in units of kilowatt-hours or megajoules per square meter per year
(kWh/m2/yr or MJ/m2/yr). The insolation reaching the Earth’s surface varies with latitude,
altitude, time of day and season, as well as with local weather and atmospheric conditions arising
from natural particulates or air pollution.
As sunlight passes through the atmosphere, some is reflected, some is absorbed, and some is
scattered. Because of these losses, the amount of energy that actually reaches the Earth never
exceeds about 70% of that present outside our atmosphere. Lower-latitude regions in the
southern United States, and especially those with dry climates in the Southwest, typically exhibit
the highest average insolation in the country—indeed, among the highest in the world.
Insolation reaches a solar collector either directly (“direct-normal radiation”), after being
scattered (“diffuse radiation”), or after being reflected from the ground. These are important
distinctions because all solar thermal designs employ optical concentration and can use only
direct-normal solar radiation.
Figure 7-1 shows the distribution of direct-normal insolation in the world and United States. The
areas with the highest insolation include parts of the United States, Australia, Africa, and the
Middle East. In the United States, direct-normal insolation is most abundant in the Southwest,
and the largest regions of highest-quality resource are in Nevada, Arizona, California, and New
Mexico. Average insolation values for a given location, such as those shown in Figure 7-1, are
adequate for rough calculations, but to accurately predict the output and, therefore, the energy
cost of a solar thermal system, measurements should be taken over the course of at least one year
at a potential power plant location.
To assess the potential solar resource overall, it is important to know the geographic regions
where insolation is generally high, but the truly available resource is also constrained by both
land topology and prior claims on it for other purposes. Solar thermal plants cannot be easily
installed in the mountains and they would be unwelcome in many places such as national parks,
military reservations, and urban areas. An NREL study used GIS screening techniques to identify
the land comprising the premium sites suitable for solar thermal power plant deployment in the
U.S. Southwest [11]. The study began with all land having an average insolation greater than
6.75 kWh/m2 and filtered out unsuitable areas using the following criteria: military bases with a
one-mile buffer; national wilderness areas with a five-mile buffer; Fish and Wildlife Service land
with a one-mile buffer; National Park Service land with a five-mile buffer; National Forest
Service land; cropland; major highways with a half-mile buffer; navigable waterways with a
half-mile buffer; lakes with a two-mile buffer; major urbanized areas with a four-mile buffer;
railroads with a 500-foot buffer; and locations 9,000 feet above sea level with a 4.5-mile buffer
7-7
10581090
Solar Thermal
around each plant. The study concluded that using only 3% of the land area that remained after
filtering could provide over 1 million GWh of solar thermal-generated electricity annually,
representing about 25% of net U.S. electricity generation in 2008. A recently commissioned
study by the DOE intends to outline the feasibility of achieving 10% and 20% of power
generation from solar (PV and solar thermal) by 2030.
Figure 7-1
Distribution of Direct-Normal Insolation Worldwide (Top, kWh/m2/yr) and U.S.
(Bottom, Wh/m2/day) (Sources: U.S. DOE and ISET)
7-8
10581090
Solar Thermal
7.3 Technology Description

This section describes the general function and operating characteristics of the most significant
solar thermal technologies: parabolic trough, central receiver, dish/engine, compact linear
Fresnel reflector, and solar chimney. These descriptions focus on the history and engineering
technologies themselves. More information on current and future projects is provided in the
following section, Technology Status.
7.3.1 Parabolic Trough
Parabolic trough technology is currently the most proven solar thermal electric technology. To
date, the parabolic trough is the only solar thermal technology that has achieved commercial
operation at significant scale. Some 354 MW of trough solar-electric systems, hybridized to use
natural-gas firing for up to 25% of the necessary heat input, have been operating in southern
California since 1990. An additional 1-MW trough system was commissioned in Arizona in
2006, a 64-MW Nevada facility began operating in 2007, and nine 50-MW units have begun
operating in Spain as of September 2010. Overall, approximately 950 MW of parabolic trough
capacity exists worldwide.
A parabolic trough collector field consists of several thousand single-axis tracking solar
collectors, each with an area of 100 m2 to 500 m2 (see Figure 7-2). The solar field is highly
modular, comprising many parallel rows of solar collectors aligned on a north-south axis. Each
collector has a linear parabolic-shaped reflector that focuses the sun’s direct-beam radiation onto
a linear receiver located at the focus of the parabola. The collectors track the sun from east to
west during the day to ensure that sunlight is continuously focused on the receiver.
The sun heats the HTF from about 560°F (290°C) to 735°F (390°C) as it circulates through the
receiver tube. The HTF then returns to a series of heat exchangers in the power block, where it
is used to generate high-pressure superheated steam at 1,450 psia (10 MPa) and 700°F (370°C).
The superheated steam is then fed to a conventional reheat steam turbine/generator to produce
electricity. Spent steam from the turbine is condensed in a standard condenser and returned
to the heat exchangers via condensate and feedwater pumps to be transformed back into steam.
Condenser cooling is typically provided by mechanical draft wet cooling towers, although dry
cooling is an option. After passing through the HTF side of the solar heat exchangers, the cooled
HTF is recirculated through the solar field.
Historically, parabolic trough plants use solar energy as the primary energy source to produce
electricity. Given sufficient solar input, such plants can operate at full rated power using solar
energy alone. During summer months, the plants typically operate for 10 to 12 hours a day at full
rated electric output. However, all plants built to date have been solar/fossil hybrids, meaning
that they have a backup fossil-fired capability that can be used to supplement solar output during
periods of low solar radiation. In light of recently high natural gas prices and the relatively high
heat rates of the smaller turbines at these plants, their actual gas usage has been lower than the
historical and contractually allowed 25% rate. Thermal storage is another strategy that can
provide dispatchability without a fossil-fuel backup system.
7-9
10581090
Solar Thermal
Figure 7-2
Solar Trough Collector Field at Kramer Junction, California. Note: The Receiver Tube at
the Reflectors’ Focal Point that Transports Heat-Transfer Fluid
Figure 7-3
Andasol 1 and 2 Parabolic Trough Plants, Spain (Source: Solar Millennium)
In the United States, the solar collector field typically provides maximum solar output in late
spring or summer. The gas-fired heater is typically used to extend operation to late afternoon and
early evening hours during the summer peak period and to augment the solar collector field
during cloudy periods. Although the gas-fired heater is physically capable of more extended use,
power plants operated as qualifying facilities (QF) under PURPA are limited to generating 25%
of total energy from gas. Utility-owned plants are not subject to the same limitation on gas use at
the present time. The California plants’ principals formerly included Sunray Energy, the Kramer
Junction Co., KJC Operating Co., and FPL Energy, but now FPL Energy is the principal owner.
The plants benefit from federal and state investment incentives, solar property tax exclusion, and
accelerated depreciation intended to encourage solar projects.
7-10
10581090
Solar Thermal
7.3.2 Central Receiver
Central receiver plants use hundreds or thousands of sun-tracking mirrors called heliostats to
focus concentrated solar radiation on a tower-mounted heat exchanger (receiver). Because they
perform better at larger sizes, these plants are best suited for utility-scale applications of many
tens to a couple hundred of megawatts.
The heliostat field surrounding the tower is laid out to optimize either the annual or seasonal
performance of the plant. In a typical installation, solar energy collection occurs at a rate that
exceeds the maximum required to provide steam to the turbine. Consequently, a thermal storage
system can be charged at the same time that the plant is producing power at full capacity.
Without energy storage or fossil-fuel hybridization, solar technologies are limited to annual
capacity factors of less than approximately 30%. However, with sufficient storage capability, a
central receiver could potentially operate all of the time without a backup fuel source. The Solar
Two project actually demonstrated such 24-hour operation.
Although central receivers are commercially less mature than parabolic trough systems, several
component and experimental systems have been field tested around the world in the last 25 years
to demonstrate the engineering feasibility and economic potential of the technology. These
experimental facilities were built to show that central receivers can produce electricity and to test
and improve individual system components.
The first large-scale test of the central receiver system was the 10-MWe Solar One project at
Daggett, California, which operated from 1982 to 1988 and was then the world’s largest central
receiver plant. It proved that large-scale central receiver power production was feasible. In that
plant, water was converted to steam in the receiver and used directly to power a conventional
Rankine-cycle steam turbine. The thermal storage system stored heat from solar-produced steam
in a thermocline tank filled with rocks and sand, using oil as the heat-transfer fluid. The storage
system extended the plant’s power-generation capability into the night and provided heat for
generating low-grade steam for keeping parts of the plant warm during off-hours and for
morning startup. Unfortunately, the storage system was complex and thermodynamically
inefficient. While Solar One successfully demonstrated central receiver technology, it also
revealed the disadvantages of a water/steam system, such as intermittent turbine operation due
to short-term cloud cover and lack of effective thermal storage.
7-11
10581090
Solar Thermal
Table 7-1
Central Receiver Demonstration Projects
Power
Heat Transfer Operation
Project Country Output Storage Medium
Fluid Began
(MWe)
SSPS Spain 0.5 Liquid Sodium Sodium 1981

EURELIOS Italy 1 Steam Nitrate Salt/Water 1981
SUNSHINE Japan 1 Steam Nitrate Salt/Water 1981
Solar One USA 10 Steam Oil/Rock 1982
CESA-1 Spain 1 Steam Nitrate Salt 1983

MSEE/Cat B USA 1 Molten Nitrate Nitrate Salt 1984
THEMIS France 2.5 Hi-Tec Salt Hi-Tec Salt 1984

SPP-5 Russia 5 Steam Water/Steam 1986
TSA Spain 1 Air Ceramic 1993
Solar Two USA 10 Molten Nitrate Salt Nitrate Salt 1996

PS10 Spain 11 Steam Water/Steam 2007
Solar Energy Israel 6 Steam na 2008

Development Center
PS20 Spain 20 Steam Water/Steam 2009
Eureka Spain 2 Steam Water/Steam 2009

Jülich Germany 1.5 Air Air 2009
AZ20 Spain 20 Steam Water/Steam na
Gemasolar Spain 17 Molten Nitrate Salt Nitrate Salt 2011

(aka Solar Tres)
During the operation of Solar One, research began on a more advanced molten-salt central
receiver design that would store energy more efficiently. This led to the Solar Two project
(Figure 7-4). In the Solar Two power tower, liquid salt at 550°F (290°C) was pumped from a
“cold” storage tank through the receiver, where it was heated to 1,050°F (570°C), then on to a
“hot” tank for storage. When power was needed from the plant, hot salt was pumped to a steam
generating system that produced superheated steam for a conventional Rankine-cycle
turbine/generator system. From the steam generator, the salt was returned to the cold tank, where
it was stored and eventually reheated in the receiver. Determining the optimum storage size to
meet power-dispatch requirements was an important part of the system design process.
7-12
10581090
Solar Thermal
Figure 7-4
Solar Two Central Receiver in Operation. The two tanks to the left of the tower base are the
storage tanks for the hot and cold molten salt solutions.
The Solar Two storage medium was a 60/40 mixture of sodium nitrate and potassium nitrate,
which melts at 428°F (220°C). It was maintained in a molten state at 554°F (290°C) in the “cold”
storage tank. Molten salt can be difficult to contain and transport because it has a low viscosity
(similar to water) and it wets metal surfaces extremely well. An important consideration in
successfully implementing this technology is identifying pumps, valves, valve packing, and
gasket materials that will work with molten salt. Accordingly, Solar Two was designed with a
minimum number of gasketed flanges, while most instrument transducers, valves, and fittings
were welded in place. Designers specified stainless steel for all pipes, valves, and vessels
because of its corrosion resistance in the molten-salt environment. In addition, the design must
allow the receiver to rapidly change temperature during cloud passage, for example, without
being damaged. The Solar Two receiver could safely change from 554°F to 1058°F (290°C to
570°C) in less than one minute.
The developers of Solar Two envisioned the project primarily as a demonstration of a complete
molten salt central receiver power plant, albeit at a less than commercial scale. In reality, the
program was more a development effort than a demonstration effort. Operators used many of the
program’s resources to improve system components and devise solutions for problems that arose.
The result is that Solar Two was quite successful as a development program. Much was learned
and documented that should substantially reduce the technical risks associated with future molten
salt central receiver plants.
7-13
10581090
Solar Thermal
Solar Two’s major contributions toward resolving key issues included:

• It clearly demonstrated that a solar central receiver can operate as a dispatchable power plant
as part of a utility generating mix. On several occasions the plant operated for many
consecutive days, providing power to the grid during peak load periods even after sundown,
and in some cases even through the night. During one such period, the plant operated
continuously for 154 hours, although at reduced power because of limited thermal storage
capacity.
• Although trace heating was a major unresolved concern at the conclusion of earlier central
receiver projects, substantial progress toward resolving this issue was made at Solar Two.
Trace heating is needed to prevent the molten salt from solidifying while the plant is not
operating. After the total rework of the original trace heating system during 1996, the system
operated with good reliability through the end of the project. Effective trace heating in the
vicinity of bends and unusual shapes such as those present in the receiver piping is still an
issue of some concern, but reasonable solutions have been proposed that can be employed in
future plants.
• Except for the heliostat field, which had known deficiencies, Solar Two actually achieved
design-point efficiencies for all its major subsystems. Of particular importance were the
receiver and thermal storage performance parameters. The receiver demonstrated efficiencies
in the range of 86% to 88%, and the effective efficiency of the storage subsystem was 99%
after taking into account that some of the collected thermal energy is transported quickly to
the steam generator. Even if all of the energy were to reside in thermal storage for a 24-hour
period before transport to the steam generator, efficiency would have been about 97%.
• Storage performance is highly significant. That subsystem worked under normal operating
conditions for essentially three years, and was particularly well characterized. It clearly
demonstrated the ability to store energy for several days with minimal losses. In addition,
post-experiment examination of the tanks (carbon-steel for the low-temperature tank and
stainless steel for the high-temperature tank) showed little evidence of corrosion, which
suggests not only excellent performance but also long service life for molten-salt storage
systems. No other renewable energy generation option can offer storage with performance
approaching this level. In addition, the incremental cost of this storage is relatively low, by
some estimates, approximately 10% of the total plant cost assuming five hours of storage.
• Project funding limitations, coupled with concerns about the condition of the heliostat field,
prevented the operation of Solar Two for another year or two. This is unfortunate, because a
number of equipment and operating strategy issues had been resolved during the project, and
it may have been possible to obtain operating experience under conditions characteristic of
commercial operation. Data on annual operating efficiencies obtained under such conditions
could have been adjusted to account for known deficiencies in the plant such as turbine
generator limitations and heliostat field condition.
7-14
10581090
Solar Thermal
In addition to annual energy production potential, several other key technical issues remain
unresolved and cause risk levels to remain high. These include:
• Long thermal time constants still result in limitations on useful energy collection. Morning
warm-up periods for the thermal collection and electricity generation subsystems are still
measured in hours. Warm-up periods might be reduced by improving warming strategies
using stored thermal energy and by increasing plant capacity factors by increasing field size
and storage capacity. The effectiveness of such measures remains to be demonstrated in
future plants.
• Lifetime and maintenance requirements for molten-salt valves are still uncertain. The Solar
Two evaluation team recommended redesigning the salt loop to minimize the use of valves.
Another recommendation for future plants was to use pumps with long shafts so that they can
be placed on top of the salt tanks.
• Acceptable parasitic power consumption needs to be demonstrated. Substantial related
progress was made with Solar Two, but long-term power production is needed to resolve this
concern.
• System startup in high winds is still an issue. As with earlier central receiver field tests, the
inability to operate during high winds was a significant cause of lost energy at Solar Two.
• Heliostat field availability at Solar Two was less than expected. In contrast, availability for
Solar One was very high. While inattention to the entire plant facility between the two
projects was likely a major cause for the drop in availability, the issue of long-term heliostat
life is still open. It also appears that earlier high hopes for potentially low-cost stretched
membrane heliostats have diminished and that, for the foreseeable future, plants will rely on
glass-metal designs. This has implications for commercial plant costs because the glass-metal
designs are more expensive. On the other hand, experience with trough systems demonstrates
that glass reflectors perform very well under real-world operating conditions.
The 11-MW PS10 plant near Seville, Spain was the world’s largest operating central receiver
plant in 2008. The system operates at saturated steam conditions of 482°F (250°C). After
roughly a year and a half of commercial operation, the plant owners reported that the
performance had improved by 10% due to optimization of the temperature gradient across the
receiver surface and other operational improvements. The 20-MW PS20 plant sited next to PS10
came online in 2009. The larger tower was designed with lower auxiliary loads to boost
performance. Eureka, a 2-MW high-temperature demonstration tower, is also operational at the
same site to validate superheated steam operation.
7.3.3 Dish/Engine
Dish/engine systems employ a parabolic reflector to focus concentrated sunlight onto a receiver
located at the focal point of the dish (Figure 7-5). The sunlight heats a working fluid in the
receiver tube, which transfers heat to a small engine used to generate electricity. Many
thermodynamic cycles and working fluids have been considered for dish/engine systems, but the
Stirling and open Brayton (gas turbine) cycles are generally favored. Conventional automotive
Otto and Diesel engine cycles are not feasible in this application because of the difficulties in
integrating them with concentrated solar energy.
7-15
10581090
Solar Thermal
Figure 7-5
25-kW SAIC Dish/Engine System at the DOE Mesa Top Thermal Test Facility
(Source: NREL)
The key components that affect the cost and performance of dish/engine systems are the
concentrator, the receiver, and the engine itself. To achieve the high temperatures required to
efficiently convert heat to work, the dish must precisely track the sun on two axes.
Until recently, the only dish/engine systems in operation were scattered demonstration units and
several installed at Sandia National Laboratories’ Solar Thermal Test Facility. However, in
January 2010, SES and Tessera Solar commissioned the 1.5-MW Maricopa Solar Plant,
composed of 60 25-kW SunCatcher units. The plant—built next to the existing Agua Fria
Generating Station on land owned by Salt River Project (SRP), which will also purchase the
generated solar power—will be operated by Tessera under a 10-year agreement. Construction
was completed in just four months. One purpose of the project is to demonstrate the technology’s
readiness for SES’s 709-MW Imperial Valley Solar PPA with SDG&E and its 850-MW Calico
Solar PPA with SCE.
7.3.3.1 Concentrators
The size of a dish/engine system’s solar concentrator is determined by the heat input
requirements of the engine. At a nominal maximum direct-normal solar insolation of 1,000
W/m2, a 25-kW dish/Stirling system has a concentrator diameter of approximately 10 meters.
Concentrators use a reflective surface of aluminum or silver deposited on glass or plastic. The
most durable reflective surfaces have been silver/glass mirrors. Because dish concentrators have
short focal lengths, relatively thin glass mirrors approximately 1 mm are required to
accommodate the required curvatures. In addition, for rear-surface-reflector mirrors, glass with
7-16
10581090
Solar Thermal
low iron content helps to reduce absorbance losses in the glass. Depending on the glass thickness
and iron content, silvered mirrors have solar reflectance values in the range of 90% to 94%.
Although glass mirrors achieve high performance and the technology is well-understood, they
drive up the cost of the concentrator. For this reason, ongoing work has centered on developing
low-cost reflective elements, such as those based on polymer films and stainless-steel
membranes.
Typically, concentrators approximate a paraboloid of rotation by using an array made up of

multiple mirrors supported by a truss structure. One innovative solar-concentrator design used
stretched membranes, in which a thin reflective membrane is stretched across a rim or hoop. A
second membrane is used to close off the space behind. A partial vacuum is drawn in this space,
bringing the reflective membrane into an approximately spherical shape. This design proved
problematic in operation, however, and is not being currently pursued.
A concentrator’s optical design and accuracy determine the concentration ratio, defined as the
average solar flux through the receiver aperture divided by the ambient direct-normal solar
insolation. Typical concentration ratios are over 2,000. The intercept fraction, defined as the
fraction of the reflected solar flux that passes through the receiver aperture, is usually greater
than 95%.
The receiver absorbs energy reflected by the concentrator and transfers it to the engine’s working
fluid. The absorbing surface is usually placed behind the focus of the concentrator to reduce the
flux intensity incident on it. Stirling and Brayton engines each have their own interface issues.
Stirling engine receivers must efficiently transfer concentrated solar energy to a high-pressure
oscillating gas, usually helium or hydrogen. In Brayton receivers, the flow must be steady and at
relatively low pressure.
There are two general types of Stirling receivers: direct-illumination receivers (DIRs) and
indirect receivers which use an intermediate heat-transfer fluid. DIRs adapt the heater tubes of
the Stirling engine to absorb the concentrated solar flux. Because of the high heat-transfer
capability of high-velocity, high-pressure helium or hydrogen, DIRs are capable of absorbing
2
high levels of solar flux (approximately 75 W/cm ). However, balancing the temperatures and
heat addition between the cylinders of a multiple-cylinder Stirling engine remains an integration
issue.
Liquid-metal heat-pipe solar receivers may help to solve this problem. In a heat-pipe receiver,
liquid sodium metal is vaporized on the absorber surface of the receiver and condensed on the
Stirling engine’s heater tubes. This results in a uniform heater-tube temperature, thereby enabling
a higher engine working temperature for a given material, and therefore higher engine efficiency.
Although not yet demonstrated, longer-life receivers and engine heater heads are also
theoretically possible through the use of a heat pipe. Stirling receivers are typically about 90%
efficient in transferring energy delivered by the concentrator to the engine.
Solar receivers for dish/Brayton systems are less developed. In addition, the poor heat-transfer
coefficients of relatively low-pressure air, combined with the need to minimize pressure drops in
the receiver, make receiver design challenging.
7-17
10581090
Solar Thermal
7.3.3.2 Engines
The engine in a dish/engine system converts heat to mechanical power in a manner similar to
conventional engines by compressing a working fluid when it is cold, heating the compressed
working fluid, and then expanding it through a turbine or with a piston to produce work. The
mechanical power is converted to electric power by an electric generator or alternator. Electric
output in current dish/engine prototypes using a Stirling engine is about 25 kWe.
The Stirling engine is a leading candidate for dish/engine systems because of its high efficiency
and because its characteristic external heating makes it adaptable to concentrated solar flux. The
Stirling engines used in solar dish/Stirling systems use hydrogen or helium working gas and
operate at high temperatures and pressures, up to 700°C (1,300°F) and 20 MPa (2,900 psia),
respectively. In the Stirling cycle, a working gas is alternately heated and cooled by constant-
temperature and constant-volume processes. Stirling engines usually incorporate an efficiency-
enhancing regenerator that captures heat during constant-volume cooling and replaces it when
the gas is heated at constant volume. There are a number of mechanical configurations that
implement these constant-temperature and constant-volume processes. Most involve the use of
pistons and cylinders. Some use a displacer (a piston that displaces the working gas without
changing its volume) to shuttle the working gas back and forth from the hot region to the cold
region of the engine. For most engine designs, power is extracted by a rotating crankshaft. An
exception is the free-piston configuration, where the pistons are not constrained by crankshafts or
other mechanisms. They bounce back and forth on gas springs and the power is extracted from
the power piston by a linear alternator or pump. The best Stirling engines today achieve thermal-
to-electric conversion efficiencies of about 40%.
The Brayton engine, also called the jet engine, combustion turbine, or gas turbine, is an internal-
combustion engine similar to an Otto-or Diesel-cycle engine in that air is compressed, fuel is
added, and the mixture is burned. In the Brayton cycle, however, the hot exhaust gases expand
through a turbine, rather than against a piston. In a dish/Brayton system, solar heat is used to
replace (or supplement) the fuel. As in the Stirling engine, recuperation of waste heat is
a key to achieving high efficiency. Therefore, waste heat exhausted from the turbine is used to
preheat air from the compressor.
7.3.4 Compact Linear Fresnel Reflector
Compact linear Fresnel reflector (CLFR) technology is designed to reduce capital costs
compared to parabolic trough and central receiver systems. The unresolved question is whether
the capital cost is sufficiently low to compensate for the lesser performance of CLFR systems.
CLFR technology is conceptually similar to the parabolic trough, except instead of using curved
mirrors it uses a field of nearly flat mirrors individually tilted and turned on their axes to reflect
sunlight to the receiver (Figure 7-6). It has some similarities to central receiver technology as
well, in that its reflectors are separately mounted from a stationary receiver.
A complete CLFR system uses arrays of mirror strips in a large field with several receivers; each
individual mirrored reflector has the option of directing reflected solar radiation to at least two
different receivers. This minimizes shading losses, allows arrays to be much more densely
7-18
10581090
Solar Thermal
packed, and permits the receiver tubes to be lower than would otherwise be possible. The
“Fresnel” in CLFR refers to the optical arrangement of reflectors in rows. While a conventional
parabolic trough solar thermal system has one curved reflector for each receiver line, the CLFR
system typically has 10.
Figure 7-6
CLFR Array at Liddell Power Station, Australia
(Source: SolarPACES)
The tower-mounted linear receivers consist of a series of steel tubes surrounded by a trapezoidal
reflective surface. Nominal sun concentration is less than 80 suns. Within the receiver,
pressurized water is converted to saturated conditions at approximately 270°C, and the steam
from this process drives conventional Rankine cycle steam turbines and generators.
7.3.5 Solar Chimney
A solar chimney generates electricity when air heated by the sun rises through a very tall
convection tower, thereby driving wind turbines (Figure 7-7). The air is heated in a large
greenhouse-like structure surrounding the base of the chimney. Turbines could be installed in a
ring around the base of the tower or within the chimney itself.
The generating capacity of a solar chimney depends on the size of the collector area and the
chimney height. Solar chimneys have been proposed that are as tall as a kilometer (0.6 mi) with
7-km (4.4-mi) collector diameter and would generate up to 200 MW.
The German government funded a research prototype chimney in Spain in 1982 that was 195
meters (640 feet) high, 10 meters (33 feet) in diameter. The plant produced a maximum electrical
output of about 50 kW and operated successfully until its decommissioning in 1989.
7-19
10581090
Solar Thermal
Figure 7-7
Cross-Section Diagram of Solar Chimney. Sunlight heats trapped air that then rises
through the cylindrical tower. Water tubes inside the greenhouse-like structure
surrounding the chimney would help retain heat. (Source: C. Pietschiny)
Architect Christian Pietschiny has suggested a novel strategy for gaining broader support and
economic benefit from a solar chimney: make it the world’s largest work of art. Pietschiny
proposes that a solar chimney presented as an enormous public expression of aesthetics, ecology,
and economy could attract attention, investors, and even tourists who would not otherwise take
an interest in an electricity generation project. Figure 7-8 is a painting showing one approach to
this vision.
Figure 7-8
Solar Chimney Sculpture (Source: C. Pietschiny)
7-20
10581090
Solar Thermal
7.4 Technology Status
After decades during which solar thermal technologies were hindered by an underdeveloped
industrial base, opportunities in the field are attracting investment and growth. Policies and
public funding to stimulate solar energy have successfully drawn the attention of industrial
players such as Siemens, Alcoa, and Rio Glass.
Progress in the 1980s, most evident with parabolic trough systems, was largely fueled by public
policies that encouraged deployment of solar thermal and other renewable technologies. After
many fallow years, similar strong legislative backing has returned. In the United States, ARRA
commitments are funding solar R&D through the DOE and its national laboratories and backing
the DOE loan guarantee program for clean energy projects. In Spain and Germany, interest
among industrial players has grown in recent years. China and India have made politically
backed overtures into solar technologies, notably the Indian Solar Mission and programs to
subsidize solar power in China. Assuming that this trend continues, the extensive existing
industrial infrastructure can quickly become engaged in solar thermal technology and business
development, as most of the equipment in solar thermal plants consists of glass and metal formed
by routine industrial processes.
Table 7-2 is a technology monitoring guide of the leading developers and technical issues in
solar thermal power generation. Table 7-3 is a technology process development “map” for solar
thermal power generation.
7-21
10581090
Solar Thermal
Table 7-2
Technology Monitoring Guide Solar Thermal Power Plants
Leading Developers of the Science or Technology

R&D Government Nonprofit Developers and Changes to Unresolved
Technologies Major Trends
Intensity Organizations Organizations Vendors Watch for Issues
Parabolic Moderate DOE (Sandia, DLR FlabegSolar Lower-cost Direct steam Steam or gas
Trough NREL), (Germany) Acciona Solar mirror support generation, molten- flow control,
CIEMAT Siemens (Solel) structures, salt or gas HTFs, cost reduction
(Spain) Solar Millennium thermal storage, non-glass reflective potential of
Abengoa/Solucar improved- film collectors, reflective film
SENER performance single-vessel collectors, freeze
Cobra receiver tubes, storage, lower-cost protection of
Schott Solar larger plants, storage strategies, molten-salt HTF
Skyfuel hybrid plants improved receiver in collector field,
Iberdrola tube designs operation of
Flagsol thermal-gradient
TSK storage tank
Central Moderate DOE (Sandia, DLR Nexant (Bechtel) Molten-salt Thermal storage, Energy
Receiver NREL), (Germany) Solar Reserve/USRG receiver, decoupling energy production,
CIEMAT (UTC)/Rocketdyne superheated- collection from high temperature
(Spain) Abengoa/Solucar steam receiver, electricity operation, cost
Sener air receiver, generation,
BrightSource smaller/cheaper multiple distributed
eSolar heliostat towers vs. single
designs tower,
supercritical-CO2
Brayton engines
Parabolic Low DOE (Sandia), Stirling Energy Lower-cost dish Durable free-piston Engine
Dish/Stirling CIEMAT Systems (SES) and mirror engine, phase- availability, O&M
Engine (Spain) Schlaich-Bergermann structures change thermal costs, cycling
and Partner (SBP) storage impacts
Infinia
Compact Low DOE Areva (Ausra) Early demos, Scale-up to Cost, efficiency,
Linear CIEMAT TSK hybrid megawatt scale, proven
Fresnel operation, proof stand-alone performance,
Reflector of economics operation long-term O&M
cost.
7-22
10581090
Solar Thermal
Table 7-3
Technology Process Development Map: Solar Thermal Power Plants
Parabolic Parabolic Molten Salt Direct Steam

Dish Stirling
Trough/Gas Trough with Central Central
Engine
Hybrid Storage Receiver Receiver
Features and Advantages
Technology Parabolic trough Parabolic trough Heliostat field Heliostat field Parabolic dish
Characterization
Heat transfer Heat transfer Heat transfer Direct steam Heated gas
from oil to steam from oil to molten from molten salt cycle (saturated (helium/hydrogen)
cycle salt to steam to steam cycle or superheated) powers Stirling
cycle engine
1
Efficiency (net) 15-17% 13-14% 15-20% 15-22% 16-30%
Resources Flat terrain with Flat terrain with Flat terrain with Relatively flat High annual direct
high annual high annual direct high annual terrain with high insolation
direct insolation insolation and direct insolation annual direct
and some some natural gas and some insolation and
natural gas availability (for natural gas some natural
availability freeze protection availability (for gas availability
(depending on or backup) freeze protection (for backup)
amount of or backup)
hybridization)
Challenges and Disadvantages
Cost Capital cost Capital and O&M O&M cost Capital cost Capital and O&M
cost cost
Operation Optimum control High freezing Steam cycle Uncertain

of storage point of salt control, high maintenance costs
subsystem, high parasitic losses
freezing point of
salt
Technology Direct steam Direct steam Reliable, cost- Steam control Need for
Needs generation generation or effective trace system, energy highly reliable
or other high- other high- heating storage, reflectors,
temperature temperature HTF, capability parasitic loss trackers, and
HTF, inexpensive reduction engines
inexpensive reflectors with
reflectors with resistance to
resistance soiling, lower cost
to soiling storage
Costs
Relative Capital Medium High Medium Uncertain Uncertain
Relative O&M Medium High High Uncertain Uncertain

2
Busbar From 13¢/kWh From 13¢/kWh From 11¢/kWh Uncertain Uncertain
1
Expected values for new plants
2
Does not reflect potentially high financing costs for first of a kind systems nor subsidies
7-23
10581090
Solar Thermal

Technology Process Development Map: Solar Thermal Power Plants
Parabolic Parabolic Molten Salt Direct Steam

Dish Stirling
Trough/Gas Trough with Central Central
Engine
Hybrid Storage Receiver Receiver
Development Timeframe
Research Late 1970s Current 1980s Late 1970s 1970s-1980s
Development 1980s Current 1980s Early 1980s 1980s-1990s
Demonstration 1986-1987 Current 1993-1998 1984-1987 1983-1990s
Commercialization 1988 2009 2013 2009 2010
Markets
Key Issues Examples First-of-kind High first PS10, PS20, High first cost
operating (Andasol-1) cost hurts and various hurts market
successfully operating and market modular prospects
several others prospects concepts have
High first cost under stimulated Limited
hurts market construction in Limited interest competition
prospects Spain competition for Stirling
for molten- engine and
Dispatchability salt parabolic dish
helps market equipment; collectors
prospects strong
competition
for other key
components
Impact on Recent interest, Recent Interest PS10 and Surprising

Technology Outlook driven in part interest, driven initially PS20 designs announce-
by RPS in CA, in part by RPS lagged appear to ment of SES-
NV, AZ, NM, and Spanish trough and be limited SCE/SDG&E
CO, TX, and solar incentive direct steam to very high PPAs shows
Spanish solar may spur very tower due to feed-in tariff promise to
incentive may large-scale high cost market accelerate this
spur very large- development in and need for technology to
scale next 5-10 years relatively Brightsource large-scale
development in large system central tower plants.
next 5-10 years sizes design and
(>100 MW). eSolar Addition of
modular thermal
Increased design impact storage may
interest in is unclear be key
troughs and
dishes may
spur this
technology
7-24
10581090
Solar Thermal
In 1984, Luz International installed Solar Electric Generating System I (SEGS I) in Daggett,
California. It originally had an electric capacity of 13.8 MW, incorporated six hours of thermal
storage and used natural-gas-fueled superheaters to supplement the solar energy. The storage
system was taken out of service after a fire in 1999, but the rest of the plant is still operating. Luz
also constructed additional plants, SEGS II through VII, each with 30-MW capacity. In 1990,
Luz completed construction of SEGS VIII and IX in Harper Lake, each with 80-MW rated
capacity. The total generating capacity of these plants is 354 MW; individually and combined,
they are the largest solar power plants in the world. From the mid 1990s until about 2000, they
had generated more solar-electric kilowatt-hours than the total of all other worldwide sources.
As a result of regulatory and policy obstacles, as well as marginal economics without ongoing
tax credits, Luz International and four subsidiaries filed for bankruptcy on November 25, 1991.
Several companies assumed operation and maintenance of the facilities as independent power
generators. As of February 2005, FPL Energy held a significant minority ownership in all but
SEGS I and II, making it the largest holder of solar-electric generation facilities. In its later
years, Luz had begun developing direct steam generating solar collectors and integrating the
solar collector field into a gas-fired combined cycle plant. Other firms that continue involvement
in this technology include Acciona Solar Power, Solar Millennium AG, Abengoa/Solucar,
Siemens (formerly Solel), Schott Solar, SENER, Cobra, Iberdrola, SkyFuel, Flagsol (a joint
venture between Solar Millennium and Ferrostaal), TSK, and FlabegSolar. Specifically in the
field of mirrors, Rio Glass and Saint Gobain are very active. In December 2002, Sierra Pacific
Resources of Las Vegas announced plans for its two Nevada utility subsidiaries, Nevada Power
and Sierra Pacific Power, to buy the output of what later became a 64-MW parabolic trough
plant near Boulder City, Nevada. The project was built by Acciona Solar Power and is part of the
utility’s strategy to comply with the state’s renewable portfolio standard. Groundbreaking for the
plant’s construction occurred in February 2006 and the plant was commissioned in June 2007.
The Spanish government’s alternative energy program has been testing trough systems at its
solar test facility in Almería. The version of the Spanish solar feed-in tariff begun in 2007
provides two options for payments to solar plants up to 50-MW capacity: a flat €269/MWh or a
premium over market of €254/MWh, capped at €344/MWh total price. This offer is valid for up
to 500 MW of solar generation and would-be solar-plant builders have applied for permits
totaling perhaps twice that amount. The first to be finished was Andasol 1, a 50-MW plant with
over seven hours of storage. It began commercial operation in early 2009. Eight additional 50-
MW trough plants are now operational.
The World Bank’s Global Environment Facility (GEF) has pursued government and commercial
interests in building the next wave of solar thermal trough power plants. The builders envision
these installations as fossil-hybrid combined-cycle plants, with the solar contribution limited to
10-20%. Egypt is now building a hybrid solar fossil thermal power plant that will generate 140
MW, 20 MW of which will be provided by solar power. The Moroccan national utility, ONE,
plans to build an integrated solar combined cycle (ISCC) plant with a net output of 470 MW, 20-
MW of which will be solar trough. Separate from the GEF, Algeria plans a 150-MW total, 20-
MW solar plant and GEF studies for similar systems are underway in India and Mexico. There
are plans for two additional ISCC plants in Spain.
7-25
10581090
Solar Thermal
In the United States, there are plans for three ISCC plants. The first to receive regulatory
approval and begin construction was FPL’s $476 million Martin Solar Energy Center, which
adds 75-MW of solar capacity to an existing 1000-MW combined-cycle plant. It is due to be
completed by the end of 2010. Another project, the Palmdale (Calif.) Hybrid Plant, is conceived
as a city-owned 570-MW combined-cycle plant with about 50 MW of solar capacity. As of
February 2010, the project was being analyzed by California Energy Commission staff. A third
project, dubbed Victorville 2, would have also been 570-MW combined-cycle plant with
approximately 50 MW of solar capacity. In August 2010, the city of Victorville, Calif.,
announced that it was soliciting bids from companies interested in leasing or buying the land
purchased for the project and developing a stand-alone solar farm instead. Its status is uncertain.
Figure 7-9
Martin Next-Generation Solar Energy Center (Source: FPL)
During 2008–2009, nearly 4,000 MW of additional parabolic trough systems were announced as
scheduled to be on line by 2011–2014 in the United States, Spain, and Abu Dhabi.
Development efforts are underway to improve solar collector assembly technology. A

consortium in Europe has developed and tested an advanced collector, dubbed the EuroTrough,
that is intended to contain the best features of the second and third generations of Luz collectors
(LS-2 and LS-3). Several of the parties to the EuroTrough development have also continued to
evolve variations of their own that minimize support frame material and offer simpler, faster
installation features. In addition, Acciona in the United States developed a collector based on the
LS-2 design that offers advantages in manufacturing, installation, and resistance to high winds.
German companies Siemens (formerly Solel, the Israeli firm that owned the rights to the Luz
technology) and Schott Solar have each developed improved vacuum-insulated receiver tubes.
7-26
10581090
Solar Thermal
After operating for three years and successfully demonstrating the feasibility of molten-salt
central receiver technology, Solar Two shut down in 1999. Since the Solar Two project, the locus
of central receiver activity has moved to Europe, where interest in such concepts has continued
for more than two decades. A consortium formed by Boeing, Bechtel subsidiary Nexant, and
their Spanish partner Ghersa invested significantly to investigate design options, prepare cost
estimates, and negotiate contract opportunities for a new central receiver plant in Spain,
Gemasolar (formerly Solar Tres). After some delays and corporate realignment, Torresol Energy
is now developing the project.
The 17-MW Gemasolar project near Seville, Spain is under construction. It will make use of
several advanced technologies, including 2650 glass-metal heliostats (120 m² each) with higher
reflectivity glass and lower manufacturing costs, and a large molten-salt storage system that will
provide up to 15 hours of capacity at 565°C (1050°F) and would enable capacity factors of about
75%. The central-receiver tower will be 147 meters (480 feet) tall. Table 7-4 compares Solar
Two and Gemasolar.
Table 7-4
Comparison of the Solar Two and Gemasolar Projects
Solar Two Gemasolar
Steam Turbine Net Power (MW) 10 17

2
Annual DNI (MWh/m ) 2.7 2.1
Field Layout Surround Surround
Solar Multiple 1.1 ~2

2
Field Mirror Area (m ) 82,000 318,000
Receiver Size (MWt) 40 120

Receiver Configuration Cylindrical Cylindrical
Storage (hrs) 3 15
Mature annual capacity factor 20% ~75%

Mature annual efficiency 8% unknown
In January 2009, Torresol Energy—a joint venture of SENER and Masdar of Abu Dhabi—
invested €171 million to begin construction on the project. In November 2009, the European
Investment Bank (EIB) loaned an additional €80 million. The overall project is also backed with
funds provided under the terms of the Fifth European Community Framework Research Program
(Contract NNE5-2001-369). At last report in late 2010, construction on Gemasolar was
proceeding well and it was on track to be completed by April 2011.
7-27
10581090
Solar Thermal
Figure 7-10
Gemasolar Plant Under Construction, May 2010 (Source: SENER)
Figure 7-11
Gemasolar Molten Salt Storage Tanks Under Construction, June 2010 (Source: SENER)
The 5-MW Sierra SunTower built by eSolar began generating electricity in August 2009.
Constructed in less than one year, the plant features 24,000 heliostats, approximately 1-m2 each,
that concentrate sunlight onto a central tower. The Sierra SunTower facility achieves 5 MW by
combining two 2.5-MW modules. Power generated is being sold to Southern California Edison.
7-28
10581090
Solar Thermal
Figure 7-12
Sierra SunTower plant (Source: eSolar)
The project, on private land zoned for heavy industrial use, was fully financed and developed by
eSolar, which hopes to use it as a template for future plants. In February 2009, eSolar announced
an agreement with NRG Energy to develop three plants in California and New Mexico that
would generate up to 465 MW. In March 2009, the company licensed its technology to the
ACME Group to develop up to 1 GW of solar-thermal capacity in India, and in January 2010
signed a memorandum of understanding to develop up to 2 GW of solar power in China.
In September 2007, BrightSource filed an application with the California Energy Commission
(CEC) to build two 100-MW plants and one 200-MW plant on Federal land, to be called Ivanpah
1, Ivanpah 2, and Ivanpah 3. More recently, they won conditional approval for a DOE loan
guarantee, and expect to begin construction in time to take advantage of an offer of cash in lieu
of an investment tax credit. Electricity generated will be sold to PG&E. In 2009, BrightSource
announced an additional 910 MW of contracts with PG&E and a 1300-MW PPA with SCE.
BrightSource expects to produce electricity at 70% of the levelized cost of the SEGS plants with
20% solar-to-electric efficiency.
The Ivanpah plants will employ BrightSource’s proprietary “Luz Power Tower” LPT 550
technology, whose working fluid operates at 550°C (1020°F). The company is also developing a
next-generation central receiver called the “LPT 650” that would operate at 650°C (1200°F).
BrightSource expects each 100-MW unit to employ a single tower surrounded by 50,000
2
heliostats, each about 10-m in area. As suggested in Figure 7-13, a 200-MW unit (most distant
in the figure) could use multiple towers. These plants could in principle also incorporate thermal
energy storage to increase their flexibility and dispatchability; however, this has not been
demonstrated and it seems likely that the plants will include a natural-gas-fired boiler to simplify
morning and evening start-up and shut-down. The plants will use dry air-cooled condensers,
rather than wet cooling, to minimize water usage in the desert environment.
7-29
10581090
Solar Thermal
Figure 7-13
Artist’s conception of BrightSource Ivanpah Solar Power Complex. (Source: BrightSource
Energy)
BrightSource selected Bechtel, an equity investor in the project, to build the Ivanpah plant in
September 2009. In February 2010, BrightSource requested the CEC to consider a revised design
that has only three towers and covers about 12% less land to lessen the plant’s environmental
impact, specifically to reduce incursion on desert tortoise habitat. Ivanpah is one of several
projects that BLM is fast-tracking to expedite their approval to make them eligible for cash
refunds in lieu of investment tax credits under ARRA. If all goes well, construction could begin
in late 2010 and operations commence in 2013.
In late 2009, SolarReserve announced three large molten salt central receiver projects, one in
Spain and two in the United States. One of them, the 100-MW Crescent Dunes Solar Energy
Project with NV Energy to be located near Tonopah, Nevada, won approval from the Public
Utilities Commission of Nevada in July 2010. This is good news for the solar thermal
development community, as such plants represent an order-of-magnitude size increase over Solar
Two—and nearly as large a step beyond Gemasolar. As such, they could demonstrate actual
commercial-scale central-receiver-with-storage operation. However, such a large one-step scale
increase entails considerable risk, which SolarReserve addresses through a technology guarantee
from United Technology Corporation (UTC), a $50 billion company that licenses the technology
to SolarReserve.
7-30
10581090
Solar Thermal
South Africa’s major electric utility, Eskom, has also been actively evaluating solar thermal
options for use in that country. Unlike the solar energy resources in Spain, where the best
resources are still 10% to 20% below that of the best Southwestern U.S. sites, some South
African sites are among the best in the world. There have been reports that Eskom may evaluate
and build a 100-MWe central receiver plant.
7.4.3 Dish/Engine
The dish/engine designs of Stirling Energy Systems, Inc. (SES) stem from the McDonnell
Douglas work of the 1980s. SES holds the rights to the McDonnell Douglas concentrator
technology, and continues the relationship with Kockums to use its 4-95 25-kWe engine, which
SES has the exclusive right to market in North America. The company’s “SunCatcher” product
is a 25-kW dish/engine unit with radial curved glass mirror facets arranged in a paraboloid to
focus sunlight onto a four-cylinder reciprocating Stirling engine. In February 2008, a SunCatcher
system at Sandia National Laboratories set a world record for dish/engine solar-to-electric
efficiency of 31.25%. In 2004, SES began working with Sandia on a six-dish 150-kW
demonstration program to clarify and mitigate scale-up issues for the SES design. Research with
Sandia and others led to improvements in design, materials and weight, which should translate
into higher efficiency and lower cost.
SES is the leading producer of parabolic dish technology in the world today, although Infinia has
made great strides in recent years and has promising ideas for incorporating thermal energy
storage into its dish/engine designs. Planned SES projects include a 34,000-dish 850-MW power
purchase agreement (PPA) with Southern California Edison called the Calico Solar Project, a
709-MW, 28,360-dish PPA with San Diego Gas & Electric called the Imperial Valley Solar
Project, and the 27-MW, 1080-dish Western Ranch Project planned for Texas. Delivery of the
first units is scheduled for 2012.
In January 2010, SES and Tessera Solar commissioned the 1.5-MW Maricopa Solar Plant,
composed of 60 25-kW SunCatcher units. The plant—built next to the existing Agua Fria
Generating Station on land owned by Salt River Project (SRP), which will purchase the
generated solar power—will be operated by Tessera under a 10-year agreement. Construction
was completed in just four months. It is being studied as a demonstration and test-bed for the
larger projects to come.
One effective strategy SES is pursuing is to develop technologies with production partners from
the automotive industry that are already familiar with similar materials and requirements. For
example, in September 2009 SES and Tessera signed an agreement with Tower Automotive to
supply mirror facets, and chose Linamar, a vehicle engine designer and builder, to manufacture
its Stirling engine power conversion units.
7-31
10581090
Solar Thermal
Figure 7-14
SunCatcher™ Units Deployed at the Maricopa Solar Plant (Source: Tessera Solar)
Another notable company is Infinia, based in Kennewick, Washington. The company’s 3-kW
“PowerDish” production unit was demonstrated in Spain and the United States, and in mid-2010
was in pilot-line production. The company claims multiple years of maintenance-free operation
at a cost moderately lower than photovoltaic technology. They report a net peak efficiency of
24%. One unique aspect of the Infinia system is the possibility of thermal energy storage using
phase-change materials. The company received $9.4 million in 2008 and another $3 million in
2009 from DOE to develop a thermal energy storage system for its dish/Stirling unit. Infinia
expects its product to be commercially available in 2011-2012.
Figure 7-15
Thirty Infinia PowerDish™ Units Demonstrated in Spain (Source: DOE EERE)
7-32
10581090
Solar Thermal
A European program called EuroDish is a joint venture of several German and Spanish firms,
with government support from the European Community and the German and Spanish
governments. This program is headed by Schlaich-Bergermann and Partner (SBP), and uses the
SOLO Kleinmotoren V-161 engine. Twelve units have been built and tested, most on the order
of 10 kW in size. SBP has focused attention on achieving fully automatic operation. There
appears to have been little EuroDish activity since about 2007.
In addition, a number of small industrial firms have emerged to develop and/or commercialize
Stirling systems of various sizes ranging from tens of watts to tens of kilowatts. Many of these
have originated since 1995, apparently in response to growing interest in distributed generation.
Owing to their use of external, continuous combustion, Stirling engines are expected to have
environmental characteristics superior to those of internal combustion engines. In fact the SOLO
unit mentioned above is being developed primarily for on-site cogeneration.
Until it was acquired by Areva in February 2010, Ausra was the world’s leading proponent and
developer of CLFR technology. The first significant prototype demonstration of CLFR was the
Liddell Solar Thermal Station in Australia, where a 3-MW CLFR system was built to augment
the steam cycle of a conventional 2000-MW coal-fired steam turbine plant. In October 2008, the
5-MW Kimberlina pilot plant was commissioned in central California to become the first
commercial demonstration of CLFR technology. Areva’s stated intention is to build an
international solar steam-generating business around Ausra’s CLFR technology, which may
stimulate additional interest in the approach and investment by others.
Puerto Errado 1 (PE1) is a 1.4-MW CLFR plant commissioned in April 2009 near the town of
Calasparra, Spain. Built by Novatec Biosol and Prointec, the plant consists of two 800-meter-
long solar fields providing steam for a Rankine-cycle turbine-generator. It was considered a
prototype for larger plants to follow.
In February 2010, the international engineering firm TSK announced plans to build the 30-MW
Puerto Errado 2 (PE2) CLFR plant in Spain, consisting of two 15-MW GE steam turbines. Each
turbine will be served by a solar field consisting of 14 receivers 940 meters long, each flanked
with eight rows of flat mirrors with a total area of 300,000 m2 to concentrate sunlight on the
receivers. The plant’s solar field technology was developed by Novatec Biosol. Groundbreaking
took place in April 2010, with completion scheduled for April 2012.
The cost of electricity from solar thermal power plants depends on a multitude of factors,
including capital and O&M cost and system performance. It is important to note that the
technology cost and the eventual cost of electricity generated will be significantly influenced by
factors independent of the technology itself. For example, for parabolic troughs and central
receivers, small stand-alone projects would be very expensive. To reduce the technology costs to
be competitive with those of conventional fossil technologies, solar projects must be scaled up to
larger plant sizes. One strategy might be to create a solar power park, in which multiple projects
7-33
10581090
Solar Thermal
could be built at a single site over time. In addition, because solar thermal technologies are
capital-intensive, the cost of capital and taxation issues will have a significant impact on their
competitiveness.
Except for parabolic trough, none of the solar thermal technologies described in this report have
ever been commercially available. Although the PS10 and PS20 central receiver projects are now
providing electricity to the Spanish grid, they remain heavily dependent on generous feed-in
tariffs and have not released information about their capital expenses or O&M costs. Therefore,
real cost data based on recent commercial production and plant operation do not exist (except,
again, for parabolic troughs).
Due to the limited activity in building new parabolic trough plants until recently and the
demonstration status of dish Stirling technologies, EPRI notes that they stand to benefit
significantly from learning-by-doing cost reductions during commercial scale up, and this effect
cannot be represented accurately in the cost projections. EPRI therefore recommends that use of
these data be restricted to order-of-magnitude cost comparisons in far-future planning, because
actual costs will depend greatly upon the cumulative deployments achieved by a given year for
each of the technologies.
7.5.1 Engineering and Economic Evaluation
The performance and cost data presented in Tables 7-5 and 7-6 are from the 2010 EPRI Report,
Solar Thermal Technology Status and Performance and Cost Estimates—2010. [19] Four solar
thermal technologies are addressed, parabolic trough with no thermal storage, parabolic trough
with six-hour thermal storage, parbolic trough with 10-hour thermal storage, and dish-Stirling
with no thermal sotorage. The levelized cost of electiricity estimates in Table 7-6 are in constant
2010 dollars and are levelized over the 30-year project life.
Table 7-5
Performance and Cost Estimates for Solar Thermal Technologies (December-2010 $,
Source: EPRI [19])
Parabolic Parabolic Central

Trough- Trough- Receiver-
Dish Stirling
No Thermal 6-hr Thermal 10-hr Thermal
1 1 1
Storage Storage Storage
Rated Capacity
4,000 x 25 kW
Plant Size (no/of units x unit size, MW) 2 x 125 MW 2 x 125 MW 100 MW
(100 MW)
Physical Plant
Unit Life, Years 30 30 30 30
Land Required, acres/MW
Cooling Type Wet Wet Dry NA
7-34
10581090
Solar Thermal

Performance and Cost Estimates for Solar Thermal Technologies (December-2010 $,
Source: EPRI [19])

Trough- Trough- Receiver-
Dish Stirling
1 1 1
Storage Storage Storage
Scheduling
Hypothetical In-Service Year Jan 2011 Jan 2011 Jan 2011 Jan 2011
Preconst., License & Design Time, Months 18 18 18 18
Idealized Plant Construction Time, Months 24 24 24 18
Capital Costs
Month/Year Dollars EOY 2010 EOY 2010 EOY 2010 EOY 2010
Collectors, and Concentrators $2,168 $3,068 $3,410 $2,200
Thermal Storage $648 $470
Power Generation and Balance of Plant $852 $852 $950 $920
Project and Process Contingency $300 $456 $480 $620
$3,320 $5,024 $5,310 $3,740
Total Plant Cost
AFUDC (Interest during construction) $72 $112 $120 $40
Total Plant Investment (includes AFUDC) $3,392 $5,136 $5,430 $3,780
$680 $1,028 $1,090 $760
Owner’sCosts
Total Capital Requirement, Hypothetical In- $4,072 $6,164 $6,510 $4,540
Service Year (Includes AFUDC)
Operational and Maintenance Costs
Fixed O&M, $/kW-yr $64 $68 $66 $62
Equivalent Planned Outage Rate, %
Equivalent Unplanned Outage Rate, %
Equivalent Availability, % 96 96 96 98
Plant Annual Output MWh/yr 556,000 892,000 430,000 208,000
2
Assumed DNI, kWh/m -day 7.65 7.65 7.65 7.65
Early Early Early
Commercial Commercial Commercial
Design & Cost Estimate Rating Simplified Simplified Simplified Simplified
7-35
10581090
Solar Thermal

Constant-Dollar Levelized Cost of Electricity for Solar Thermal Power Plants (December
2010 $, Source: EPRI [19])

Trough- Trough- Receive-
Storage Storage Storage Dish Stirling
Plant Size (no. of units x unit size) 2 x 125 MW 2 x 125 MW 100 MW 4,000 x 25 kW
(100 MW)
Capacity Factor 25.4% 40.7% 49.1% 23.7%
LCOE with Investment Tax Credit, $/MWh 116 100 86 133
Fixed O&M portion of LCOE 35 23 19 36
Capital Charge Portion of LCOE 81 77 67 97
LCOE without Investment Tax Credit, $/MWh 196 175 152 228
Fixed O&M Portion of LCOE 35 23 19 36
*Constant-dollar evaluation assumes zero escalation for general inflation and fixed O&M inflation is excluded
from the debt interest and equity return rates.
The data presented in Table 7-7 below are derived from studies by EPRI and others [1, 7, 8, 9,
12, 19] and will be used to estimate upper and lower bounds for cost and performance of a
trough plant with 6 hours of storage.
The lower bound on annual solar efficiency can be inferred from the SEGS VI column of the
table. The 10.7% efficiency value shown is somewhat higher than the 9.8% value calculated for
the entire Kramer Junction park in 1998. This can be explained entirely by the SEGS VI
2
collector aperture area, which, at 188,000 m , is less than one-fifth or about 14% of the total
collector area of the park. Owing to higher steam-cycle efficiency, this plant outperformed the
earlier plants and proved that parabolic trough installations could achieve higher efficiency.
Contrary to earlier assumptions, it has been demonstrated that the later SEGS plants, including
SEGS VI, use little if any gas for plant warming. Thus, for the current evaluation, a value of 11%
will be used as the lower bound on this efficiency.
The column labeled “Advanced” assumes a good measure of optimism and includes assumptions
about cost reductions that are debatable and may prove unrealistic. For this evaluation, the data
presented for the near-term plant appear reasonable as a lower bound on cost for current
technology, and with experience and expanded production it seems reasonable to expect those
costs to drop by 10%. Hence, $490/m2 will be used as the lower bound cost estimate.
7-36
10581090
Solar Thermal
Table 7-6
Selected Data for Solar Trough Systems
SEGS VI Near Term Advanced
Plant Size (MWe) 30 125 200
Solar Field Area (m2) 188,000 1,157,560 2,000,000
Thermal Storage (hours) None 6 12

(fossil hybrid)
Capacity Factor (total) (%) 34 35.8 53

Solar Capacity Factor (%) 22.51 35.8 53
Insolation (DNI) (kWh/m2-yr) 2519 2519 2519
Annual Energy Production (GWh/yr) 89.4 427.2 928.6

(total) (gross)
Annual Solar Energy (GWh/yr) (gross) 59.0 427.2 928.6
O&M Costs ($/kW-yr) 107 65 281
O&M Costs (¢/kWh) 3.61 1.91 0.6
Collector Costs ($/kWe) 3048 2,710 14701
Collector Cost ($/m2) 4861 2931 147

Storage Cost ($/kWe) NA 754 324
Total Capital Cost ($/kW) 3972 5,028 2535
Annual Solar-Electric Efficiency (net) 10.7 13.5 16
Plant Cost1 ($/m2) 633.8 542.9 253.5

1
Data in italics are calculated values inferred from other entries or data discussed in references.
The near-term plant also provides the basis for estimating the upper bound on efficiency, again
because the apparent optimism of the advanced plant has been tempered. However, it seems
reasonable to expect that at least half of the 20% receiver efficiency improvement assumed in the
advanced plant will actually occur, which would increase plant efficiency by about 10% over
that of the near-term plant. Hence, 15% will be used for the current evaluation as the upper
bound.
For the upper bound on cost, the value achieved with SEGS VI provides a firm basis. It is
reasonable to assume that additional experience would reduce that cost somewhat, perhaps by
5% to approximately $600/m2. Because that value is given in 1997 dollars, it is adjusted from
December 1997 to December 2009 dollars. Assuming a 2.5%/yr escalation rate, the increase is
about 34% and the resulting cost is about $800/m2.
A 90%-confidence levelized cost of electricity (LCOE) range can now be calculated for trough
systems. This analysis assumes a 10.5% fixed charge rate, covering all annual charges related to
the plant’s capital investment, and annual insolation of 2,500 kWh/m2. This insolation is
7-37
10581090
Solar Thermal
somewhat less than that reported above for Kramer Junction, but is felt to be more representative
of typical locations for trough plants. For O&M costs, the upper bound assumes the value
demonstrated at the Kramer Junction 150-MWe park: 2.5¢/kWh. For the lower bound, it is
reasonable to assume that mature plants would approach a value between those shown for the
advanced and near-term plants. For this evaluation, 1.0¢/kWh will be used. The resulting upper
and lower values for levelized COE, not including any financial incentives, are:
Levelized COEU = (100)(0.105)(800)/((2500)(0.11)) + 2.5 = 33.0 ¢/kWh
Levelized COEL = (100)(0.105)(490)/((2500)(0.15)) + 1.0 = 14.7 ¢/kWh
Hence the LCOE for trough plants having modest technology advances is estimated to lie
between about 15 ¢/kWh and 33 ¢/kWh. These values are based on the assumed fixed charge
rate and insolation level, and can be easily adjusted using the above relationships to reflect
different values for those parameters if desired.
With respect to the cost of a central receiver project, there is little basis to make substantial
changes from the projections EPRI developed in 1989 [7]. Solar Two, the only major central
receiver hardware activity in the United States since 1989, provided no new cost information,
although it did provide solid, positive experience with molten salt storage equipment and
operation. It therefore provides a qualitative reduction in prudent contingency factors for some of
the major plant components, and it strengthens the basis for a portion of the lower-bound cost
estimate from the 1989 evaluation. However, the lower bound also assumed significant cost
reduction of heliostats, based on the use of stretched polymer membranes rather than glass
reflectors. In 1989, the expectation was that the unit-area cost of heliostats would eventually drop
to $75/m2 if polymer reflectance and lifetime were acceptable. However, experience with
polymer membranes over the past decade for heliostats, dish reflectors, and troughs has not been
encouraging. All current and foreseeable system development work is now based on glass
reflectors. Hence the heliostat portion of the lower-bound cost is now somewhat suspect.
Cost data from the PS10 and PS20 plants are currently not publicly available, and such saturated
steam systems are not likely to become economical commercial products. Generous Spanish
feed-in tariffs were required to overcome the economic disadvantages of saturated steam and
make PS10 and PS20 feasible in the context of their unique market. Until large-scale superheated
steam systems are developed, the economics of commercial direct steam towers will not be well-
understood.
Additional perspective on cost projections can be obtained from EPRI and DOE’s 1997
Renewable Energy Technology Characterization report [1] and EPRI’s 2008 New Mexico
Central Station Solar Power: Feasibility Study [12]. These studies offer projections for a 2010
central receiver plant with 13 hours of molten-salt thermal storage and current cost estimates for
a 2011 plant with three to nine hours of storage, respectively. Projections in the former report are
also presented for 2020 and 2030, which include modest improvements in cost and operating
efficiencies. The improvements in operating efficiencies are judged to be speculative. The results
of the study in New Mexico are considered the most realistic near-term cost and performance
7-38
10581090
Solar Thermal
figures. It should also be noted that $75/m2 in current dollars has been a long-term goal for
heliostat costs for more than two decades. Hence, the goal in constant-dollar terms has actually
been decreasing. This reduces the credibility of the goal, especially because experience suggests
that the hoped-for reductions in materials costs are becoming less likely.
The 2011 central receiver plant cost, expressed in dollars per unit area of heliostats, was
projected to be $474/m2 in Dec. 2009 dollars. In EPRI’s 1989 evaluation, the corresponding cost
2
was about $296/m in 2009 dollars. In light of the above discussion of heliostat costs and, in
particular, the speculative nature of the $75/m2 goal, this evaluation will retain the lower-bound
cost estimate of $296/m2. The upper-bound cost estimate will be changed from the value
calculated for the 1989 evaluation to 10% above the New Mexico estimate, or $520/m2.
The upper and lower bounds on the efficiency were assumed to be 10% above and below the
16% estimate in the New Mexico study. O&M was assumed to be similar to that of the trough.
Using the forgoing values in the same formula pair as the parabolic trough calculation yields an
LCOE range, expressed in December 2009 dollars, of approximately 8¢/kWh to 18¢/kWh.
7.5.4 Dish/Engine
Progress with dish/engine systems over the past 15 years has come primarily from testing
individual units fielded by SES, SAIC/STM, SBP, and Sandia. In aggregate, including the earlier
experience of the 1980s, these units logged nearly 80,000 operating hours through mid-2002.
The units at Sandia in Albuquerque have run unattended without incident for periods of several
weeks, which bodes well for increased reliability of system operation. However, there is as yet
no large body of operating experience on which to base projections for mature-system energy
costs that have substantially more validity than those offered in EPRI’s 1989 evaluation [7].
Similarly, because these programs are all heavily supported by government sources and
commercial offerings do not yet have public visibility, there is little basis for developing new
cost estimates with increased confidence. However, one factor is evident that would tend to
increase the lower bound cost estimate described in the 1989 evaluation: it now appears that all
firms pursuing dish systems are employing glass reflectors. As with central receiver heliostats,
experience with polymer membranes has been disappointing. This is expected to increase the
cost of the dish concentrator component by as much as 100%, which in turn would increase the
cost of the entire unit by about 30%. The impact on the lower-bound LCOE would be an increase
of about 30%, assuming no change in the upper-bound efficiency value.
Consequently, the LCOE range for dish systems today is estimated to be 9 to 35¢/kWh, in
2009 dollars.
Like solar PV technologies, solar thermal electric technologies are environmentally benign
relative to other forms of electricity generation. Without energy storage, all large-scale solar
plants have similar footprints, roughly 5 acres/MW (2 hectares/MW). Except for dish-Stirling
7-39
10581090
Solar Thermal
plants, solar thermal electric generators also have water requirements unless they use dry cooling
(which most new plants in the U.S. Southwest plan to use) and potentially damaging outcomes of
operational mishaps. Of course, if the plant is hybridized, operation of the fossil-fueled section
will be accompanied by the usual emissions and hazards of fossil-fuel use.
7.6.1 Land Use
As mentioned, solar thermal plants without storage require about 5 acres/MW of peak capacity in
good solar-resource locales (greater than 2,200 kWh/m2/yr). However, plants that incorporate
storage will require more land per peak megawatt. Large-scale solar thermal plants have
footprints in the range of 5 to 10 acres/MW (2 to 4 hectares/MW) depending on the number of
hours of storage capacity (10 acres corresponds to about 9 hours of thermal energy storage in
good solar-resource locales of over 2200 kWh/m2/yr). Parabolic trough and central receiver
plants with thermal energy storage systems have an oversized collector field to capture energy
for the storage system during the peak hours of the day. The generating units will typically not be
designed to use the entire peak thermal output of the collector field. Therefore, a more
meaningful metric of land use would be the area required per annual megawatt-hour of output,
-3 -3
which would range between 2.7–3.0 × 10 acres/MWh/yr (1.1–1.2 × 10 hectare/MWh/yr). Note
that, although conceptually distinct, these two quantities actually have the same fundamental
units of area per unit power output.
Dust suppression must be considered due to the large uncovered areas that are created by solar
thermal projects. Such consideration would include an economic evaluation of alternative control
methods for the collector field including compaction, dust suppressants, and permanent
stabilization.
In most cases, land appropriate for solar thermal projects consists of inhospitable, desert-like
terrain of little use for alternative development but which nevertheless raises concerns about
habitat disruption. In late 2009, potential conflicts between solar power and wildlife habitat were
spotlighted with Sen. Dianne Feinstein’s (D.-Calif.) introduction of the California Desert
Protection Act of 2010 (S.2921), which at this writing in late 2010 is still making its way through
Senate committees. The act sets aside approximately 2.5 million acres of the Mojave Desert for
conservation by expanding Death Valley National Park, Joshua Tree National Park, and the
Mojave National Preserve, as well as establishing the new Mojave Trails National Monument
and Sand-to-Snow National Monument. Several solar power projects—mostly solar thermal but
also photovoltaic—had been slated for siting on land that will be off limits to development if the
bill passes. Press reports claimed that as many as 13 solar power project developers cancelled
proposals rather than work around the restrictions. However, of possible benefit to solar projects,
Section 203 of the bill addresses renewable energy and requires the Bureau of Land
Management, Department of Defense, and U.S. Forest Service to undertake environmental
studies and identify federal land where such projects could be expedited. As of May 2010, the
Bureau of Land Management was considering more than 350,000 acres for potential renewable
energy development—much more than sufficient for any proposed use. In general,
environmental impacts on native plants and wildlife can be minimized with proper planning and
management.
7-40
10581090
Solar Thermal
7.6.2 Water Use
Water requirements of trough and tower solar thermal generating plants are similar to those of
other steam plants of equal nameplate capacity using wet cooling towers. Dry cooling is a viable
water-conserving alternative, but at a cost of up to 10% lower operating efficiency, depending on
ambient conditions. Increasingly, dry cooling is being required for thermal plants in desert
locations. In wet-cooled plants some 2 to 4 m3 of cooling water is needed for optimum efficiency
for every megawatt-hour generated. Dish/Stirling engines require no water. In all cases, a small
amount of water is needed for mirror cleaning.
7.6.3 Molten Salt for Central Receiver
No hazardous gaseous or liquid emissions are released during operation of a solar central
receiver plant. If a molten salt spill occurs, the salt will freeze before significant soil
contamination occurs. The salt may be picked up with a shovel and can be recycled if necessary.
7.6.4 Oil for Heat Transfer in Trough Applications
The current heat transfer fluid (Monsanto Therminol VP-1) is an aromatic hydrocarbon,
biphenyl-diphenyl oxide. The oil is classified as non-hazardous by U.S. standards but is a
hazardous material in California. When spills occur, contaminated soil is removed to an on-site
bio-remediation facility that uses indigenous bacteria in the soil to decompose the oil until HTF
concentrations have been reduced to acceptable levels. In addition to liquid spills, there is some
level of HTF vapor emissions from valve packing and pump seals during normal operation, but
such emissions are well within currently permissible levels.
7.6.5 Potential for Greenhouse Gas Reduction
Solar thermal power plants that are not hybridized with fossil fuel generate no direct emissions
of CO2, methane, or other greenhouse gases. Even when hybridized, the solar-generated portion
of the plant’s output is emissions-free. Consequently, all solar-thermal power plants provide
greenhouse gas emissions reductions. In addition, should a CO2 emissions-reduction mandate be
enacted in the future, solar thermal power could become an important component of a CO2
emissions-reduction strategy and could participate in CO2 emissions trading.
As discussed in Chapter 12, the CO2 emissions-reduction potential of a renewable energy power
plant is a function of the generation mix of the existing generation system, while the effective
CO2 emissions-reduction cost is a function of the CO2 emission rate and the average generation
costs of the base system and the renewable energy power plant.
7.7 Design, Deployment and O&M Issues
Issues associated with design, deployment, and O&M include the reliability and maintainability
of sun-tracking drive systems and balance-of-plant systems, operation under extreme weather
conditions (wind, precipitation, dust), and labor requirements. As noted in the “Cost and
7-41
10581090
Solar Thermal
Economic Issues” discussion, there are currently few solar thermal systems in commercial
operation yielding limited real-world O&M experience.
Through much of the 1990s, the DOE-Sandia-NREL solar thermal program worked closely with
the operators of the trough plants in southern California to reduce O&M costs for these systems.
That work focused primarily on the five 30-MWe plants—SEGS III through VII—located at
Kramer Junction, California. Through that effort, participants learned a great deal about
optimizing O&M procedures and about hardware improvements that would increase energy
production and reduce O&M costs. As a result, these costs have been substantially reduced, and
the O&M requirements for mature commercial plants based on similar hardware can be
estimated with improved confidence.
The solar collector assembly (SCA) (e.g., the mirrored trough unit) of a trough system rotates
around the horizontal north/south axis to track the sun through the sky during the day. The axis
of rotation is located at the collector center of mass to minimize the tracking power required, and
the drive system uses hydraulic rams to position the collectors. The SEGS plants were designed
for a nominal lifetime of 30 years, and the SCAs on those systems were designed for normal
operation in winds up to 25 mph (40 km/h) and with somewhat reduced accuracy in winds up to
35 mph (56 km/h). They are designed to withstand a maximum wind of 70 mph (113 km/h) in
their stowed position, aimed 30 degrees below the eastern horizon.
Actual experience in California has resulted in many lessons learned. During the period 1992
through 1998, the Sandia-industry O&M improvement program provided a 31% increase in
annual solar energy production from the 150-MWe solar park at Kramer Junction. In addition,
annual O&M costs per megawatt were reduced by 19%. Together, these advances resulted in
a reduction in O&M costs/kWh of 38%, from about 4¢/kWh in 1992 to about 2.5¢/kWh in 1998,
not including costs for natural gas fuel. The advances in collector design and receiver elements
described previously are also likely to increase annual energy performance through
improvements in wind damage resistance, integrity and protection of metal-to-glass seals,
stability of optical alignment of collector-reflectors and receiver elements, and increased
efficiency of receivers.
The following are some specific advances at Kramer Junction:

• Although the mirrors have proven to be among the trough systems’ most reliable
components, differential thermal expansion caused some separation of the mirrors from their
mounting pads. The problem was resolved through the use of new mounting materials and
adhesives. In general, mirrors can be maintained without degradation through simple periodic
washing. Front-surface mirrors have the potential to yield higher reflectivity, but must
demonstrate long-term performance, durability, and washability.
• Mirror breakage at the edges of the collector fields in high winds was a major problem that
often caused cascading breakage of encapsulated glass heat collection elements and other
mirrors. A means of strengthening the edge-mounted mirrors with a backing made of
fiberglass-resin successfully reduced breakage of these mirrors. Additional work developed
operating strategies for high wind conditions, with the aim of allowing plant operation in
7-42
10581090
Solar Thermal
high winds to continue and avoid loss of solar energy while also ensuring that vulnerable
collectors would be stowed in a safe position before wind damage could occur. Although
some valuable insights were developed, results of this work were inconclusive.
• By 1992, approximately 15% of the 50,000 heat collection elements (HCE) at the Kramer
Junction park were damaged, either through loss of vacuum in the space between the receiver
and the concentric glass tube or through loss of the glass tube entirely. Because loss of the
glass reduced performance to a much greater extent than loss of the vacuum, workers
developed a temporary repair procedure in which a new glass tube could be affixed to the
HCE. Then after several adjacent HCEs developed problems, all could be replaced at the
same time. In addition, strategies to protect the glass-to-metal seal have reduced the
incidence of vacuum loss in the HCEs. Finally, Solel and Schott have developed new HCE
designs that use an improved selective receiver-tube coating with reduced emissivity and
greatly increased resistance to oxidation and the associated degradation of optical and
thermal properties in the event of vacuum loss. Protection of the vacuum seal is also
enhanced in these new designs. Future HCE designs should use new tube materials to
minimize bowing problems, allow broken glass sleeves to be replaced in situ, and continue to
improve coating materials that efficiently transmit solar energy to the working fluid.
• The Luz LS-3 collector initially showed higher performance than earlier Luz collectors,
owing to higher concentration ratio and higher peak optical efficiency. Over time, however,
these units would lose their alignment, causing significant performance degradation.
Subsequent changes in the mechanical design of the LS-3 have corrected this problem, and
LS-3 collectors used in SEGS VIII and IX showed substantial improvement in alignment
maintainability. Operators also devised procedures for identifying collectors that were
developing alignment problems.
• Because the solar collector assemblies track the sun, piping connections to the receiver tubes
must tolerate motion. Originally, corrugated flex hoses served this purpose. Reliability of
these was marginal, and occasionally a catastrophic failure would occur, resulting in a fire
fed by the heat-collection oil. The corrugations also caused noticeable pressure losses in the
heat-collection loop. Many of these hoses have been replaced with a ball-joint arrangement
that has greatly improved the reliability of these connections. Ball joints are highly likely to
be used in future trough plants. Better seals for pumps in the thermal collection loop
represent another advance, which has greatly improved pump reliability.
Several key issues remain for trough systems, including the following:
• Despite substantial progress, O&M costs are still quite high. The trough community projects
that these costs can be reduced to 1¢/kWh or less, but this remains to be demonstrated.
• Although incremental advances have been made, substantial reductions in capital costs have
not been demonstrated.
• Increasing plant size, including thermal storage, and further increasing total collector area to
fill the thermal storage and increase plant capacity factor are likely to result in significant
cost reductions. Early experience with the molten-salt storage system of the Andasol-1 plant
appears very promising in this regard.
7-43
10581090
Solar Thermal
• Molten salt storage appears to have good prospects for use in trough systems, particularly
because of its success with the central receiver approach. However, the storage subsystem
contributes more cost to a trough system than to a central receiver system, both because
lower temperature differences are involved, which increases the salt inventory per unit of
energy storage, and because salt cannot, as yet, be used as the heat collection fluid.
Therefore, oil-to-salt heat exchangers are required that add a cost component beyond the salt
storage tanks.
• Alternative heat collection fluids have been proposed for trough systems, such as molten salt
and water-steam. These offer higher peak performance due to higher operating temperatures,
which could translate into lower costs. However, the technical difficulties of handling these
fluids without leaks or freezing throughout large collector fields in a wide range of
environmental conditions are significant. This indicates that credible cost projections for
near-term trough systems are limited to those based on close derivatives of the systems
deployed in southern California.
Other O&M issues include tracking maintenance via computerized maintenance management
software, performing periodic collector alignment, developing system design specifications and
operating procedures to allow quick start-up, and developing an effective O&M organization to
handle the special demands of operating a solar trough plant.
Prior to commercial operation, large solar systems in utility-size power plants need to pass a
performance acceptance test. NREL has undertaken the development of interim guidelines for
parabolic trough solar plants to recommend test procedures that can yield results of a high level
of accuracy consistent with good engineering knowledge and practice. As of September 2010,
those guidelines were in draft form and are not publicly available; however, the report, Utility-
scale Parabolic Trough Solar Systems—Performance Acceptance Test Guidelines [14], should
be issued by the end of 2010.
The Solar Two plant was constructed during the mid-1990s. Initial operation began in June 1996,
and the plant was operated in various experimental modes through April 1999. The original
intent was to complete system checkout during the first several months, then conduct a one-year
experimental evaluation program during which the characteristics of the plant would be fully
explored and documented without emphasis on power production. This was to be followed by a
two-year period of simulated commercial operation in which optimum power production was the
only goal.
In practice, initial operation uncovered serious problems with the system design. The trace
heating subsystem had to be completely reworked, and the steam generator developed an early
tube rupture. These repairs and others set back the schedule by about 18 months. Consequently,
the experimental evaluation and the power production phases were conducted concurrently over
about a 12-month period. During this period, the design-point (i.e., peak) performance of the
system and major components was well characterized. However, operators did not collect longer-
term operating performance information, so they could not determine actual annual system
performance. Some performance projections were possible, based on a combination of modeling
and actual data.
7-44
10581090
Solar Thermal
Peak component and system efficiencies for Solar One and Solar Two are shown in Table 7-8.
In addition to measured values for the two plants, this table shows goals for both Solar Two and
a commercial plant. The table shows that the net peak efficiency for Solar Two was projected to
be essentially the same as that actually observed for Solar One. For Solar Two, analysts factored
some acknowledged deficiencies in the heliostat field into the projected goal, but these were to
be offset by a higher receiver efficiency (which arises from a reduction in physical size, owing
to improved fluid heat transfer and, thus, reduced radiation losses). Parasitic losses would also
be higher for Solar Two, primarily because of the need for trace heating.
Table 7-7
Peak Central-Receiver System Efficiency Components
Solar One Solar Two Solar Two Commercial

Category
Actual (%) Goal (%) Actual (%) Plant Goal (%)
Mirror Reflectivity 90.3 90.7 90.7 94
Field Efficiency 76 72 66 71
Field Availability 99 98 94 99
Mirror Corrosion 100 97 97 100
Mirror Cleanliness 95 95 90 95
Receiver Efficiency 78 87 88 88
Storage Efficiency N/A 99 99 99
Steam/Turbine Generator 33 34 34 42
Gross Peak Efficiency 16.6 17.2 14.7 23.2
Parasitic Loss Factor 90 88 87 93
Net Peak Efficiency 15 15 13 22
Solar Two’s actual peak performance was somewhat less than the goal. This resulted primarily
because of deficiencies in the heliostat field, as discussed earlier. It is reasonable to assume that
those deficiencies would be eliminated in a new power plant. It is remarkable that the inclusion
of energy storage in the Solar Two plant produced an almost negligible impact on the overall
plant efficiencies.
Of greater significance are annual system efficiencies. While these were not measured for Solar
Two, an attempt to estimate them was made by the Sandia evaluation team. These estimates are
shown in Table 7-9, along with measured values for Solar One and goals for Solar Two and a
commercial plant. The table also shows values adjusted for the known deficiencies of Solar Two.
7-45
10581090
Solar Thermal
Table 7-8
Annual Average Central Receiver System Efficiency Components
Solar One Solar Two Solar Two Commercial

Category
Actual (%) Goal (%) Adjusted (%) Plant Goal (%)
Plant Availability 90 90 90 91
Overall Field Performance 58 50 51.5 56
Receiver Efficiency 65 76 76 79
Piping Efficiency 99 99 99 99
Storage Efficiency N/A 97 97 99
Steam/Turbine Generator 30 32 41 41
Gross Annual Efficiency 10.1 10.5 13.9 16.3
Parasitic Loss Factor 72 73 73 90

Net Annual Efficiency 7.3 7.7 10.1 14.7
The heliostat field represents the largest single capital investment in a central receiver power
plant. Relatively few heliostats have been manufactured to date, and their cost remains high.
In contrast to solar trough drive systems, which move on only one axis, heliostats must move
in two axes and reflect the sun’s rays onto the central tower with a high degree of accuracy.
Thus, related O&M is more complex and costly.
While it is a suitable working fluid in many respects, molten salt has proven problematic in
others. It can be troublesome to maintain in a molten state, requiring a fairly complex heat
trace system employing electric wires attached to the outer surface of pipes. Valves can also
be a problem in a molten salt system, requiring special packing and extended bonnets. Design
improvements and standardization in valve design would reduce risk and ultimately reduce
O&M costs. Based on O&M experience at Solar Two, subsequent central receiver projects
are likely to employ new materials or thermal management practices that ease some of the
problems observed.
Since Solar Two was decommissioned and disassembled, the most ambitious and noteworthy
central receiver projects are the currently operating 11-MW PS10 and 20-MW PS20 systems, as
well as the 17-MW Gemasolar project under construction and scheduled to begin operation in
April 2011. All are located in Spain. PS10 and PS20 use saturated steam as a heat transfer fluid,
while Gemasolar will use molten salt. Abengoa, which owns PS10 and PS20, has not released
detailed operational cost or performance information for those plants. More generally, Abengoa
has shared that day-to-day operation of the plants has gone as expected, and that overall their
performance has been satisfactory.
7-46
10581090
Solar Thermal
7.7.3 Dish/Engine
Like central receiver heliostats, dish/engine systems must track the sun in two axes and with
extreme accuracy. This introduces increased complexity and potential for O&M problems.
Dish/engine systems must be stowed when wind speeds are too high, typically about 35 mph
(16 m/s). If intended for remote or autonomous operation, a dish/engine system must therefore be
equipped with reliable sun and wind sensors to determine whether conditions warrant operation.
Dish/engine systems need a parabolic reflective surface to focus sunlight on a receiver; however,
that does not necessarily require a mirror. Although a reflective surface of silver deposited on
glass is the most durable option, low-cost and lightweight polymer films stretched over hoops
and drawn into a spherical shape with a vacuum have also been used successfully on a small
scale, with hope that their relatively lower cost will offset their greater fragility.
Other systems of concern include the receiver, the engine—most commonly a Stirling engine,
although Brayton cycle engines are also being investigated—engine cooling systems, and
the alternator. In particular, the Stirling engine must demonstrate acceptable efficiency and
reliability before commercial deployment can be seriously considered. In general, dish/engine
systems would benefit from the introduction of a commercial engine system produced in volume.
Detailed operating experience and O&M data on the 1.5-MW Maricopa plant, which began
operation in January 2010, are not yet available. The units are reportedly operating as expected.
The only significant demonstrations of CLFR technology to date have been the Liddell Solar
Thermal Station in Australia, the 5-MW Kimberlina project in California, and the Puerto Errado
1 project in Spain. All three entered service in 2008. Detailed operating experience and O&M
data have not yet been released for any of them. However, all reportedly operated as expected
and provided their owners with sufficiently encouraging information to develop more ambitious
projects.
The high-concentration solar thermal electric technologies discussed in this chapter will
prove most practical and economical in regions with abundant direct-normal sunlight. This
requirement somewhat constrains suitable locations for solar thermal in comparison to other
solar technologies such as flat-plate photovoltaics, which can make use of diffuse or indirect
sunlight. In addition to the systems successfully operating in southern California, Nevada,
Arizona, Spain, Germany, and Israel, other nations pursuing solar thermal power development
include Italy, Greece, Japan, Russia, France, Egypt, India, Morocco, Algeria, Brazil, Iran,
Australia and Mexico.
The 1991 bankruptcy of Luz International and its subsidiaries had a large impact on the viability
of solar thermal systems for over a decade and considerably affected the pool of expertise
available. However, the recent rebuilding of interest in central-station solar power plants has
7-47
10581090
Solar Thermal
begun attracting both investors and engineering companies. The following list of developers,
consultants, and manufacturers includes those currently engaged in developing and providing
solar thermal technologies. This list is provided for informational purposes only; no endorsement
is implied.
Abengoa Solar
Based in Spain, Abengoa is committed to developing PV, solar thermal and industrial heat
technologies for commercial, industrial and utility applications.
www.abengoa.com
Acciona Energia
C/Yanguas y Miranda, 1-5º, 31002-Pamplona, Spain. Acciona’s mission is to be leaders in
the management of infrastructures, services and renewable energy, actively contributing to social
well-being and sustainable development. Its annual revenue exceeds $5 billion and it operates on
five continents. Solargenix Energy, an Acciona subsidiary based in Sanford, North Carolina, is
developing a range of solar thermal systems using compound parabolic collectors for broad use
in buildings and electricity generation. Through Solargenix, Acciona also acquired intellectual
property from the former Duke Solar Energy.
www.acciona-energia.com
www.solargenix.com
Arizona Public Service Co.

P.O. Box 53999, Phoenix, Arizona 85072. Through its Solar Test and Research (STAR) Center,
APS is a utility leader in researching and implementing many types of solar energy systems.
www.aps.com
AREVA
Headquartered in Paris, France, AREVA acquired U.S.-based Ausra in February 2010 with the
stated intention of using Ausra’s experience in compact linear Fresnel reflector (CLFR) systems
to anchor an international solar energy business. The new “AREVA Solar” continues to market
CLFR as a solar steam generator for power and industrial applications.
www.areva.com/
Black & Veatch

With over 30 years experience with assessing renewable energy technologies, Black & Veatch
has participated in hundreds of projects throughout the world. They provide engineering,
consulting and construction services.
www.bv.com
BrightSource Energy, Inc.

1999 Harrison Street, Suite 500, Oakland, California 94612. BrightSource Energy, Inc. designs
and builds large scale solar plants that deliver solar energy in the form of steam and/or electricity
to industrial and utility customers worldwide.
www.brightsourceenergy.com
7-48
10581090
Solar Thermal
CIEMAT
(Research Centre for Energy, Environment and Technology), Carretera Senés s/n-04200
Tabernes (Almería), Spain. CIEMAT is a Research Public Institution attached to the Spanish
Ministry of Education and Science. The Solar Platform of Almería, jointly managed by CIEMAT
and DLR, is one of the most important European Centers on Solar Thermal Energy.
www.ciemat.es/portal.do
Cobra
Cardenal Marcelo Spinola, 10, 28016-Madrid, Spain. Cobra is an engineering, operations,
installation and maintenance provider that supports several industries, including energy projects.
The company has over sixty year of experience and employs over 22,000 professionals.
www.grupocobra.com
DLR (German Aerospace Center)

Collogne, Germany. This public non-government center focuses on aviation, space flight, and
energy technology in close cooperation with partners from academia and industry. An
experienced solar plant contractor and operating agent, DLR is involved in solar trough
feasibility studies in Spain and elsewhere.
www.dlr.de
eSolar Inc.
130 West Union Street, Pasadena, California 91103, is a developer of a novel, modular, 46-MW
system for building distributed central-receiver plants.
www.esolar.com
FlabegSolar
Also known as Flabeg Solar International, this company is a subsidiary of the Pilkington Group,
one of the world’s leading glass manufacturers. FlabegSolar manufactures a highly-specialized
glass called Optisol™ that is used for solar reflectors, flat-plate collectors, and PV applications.
Pilkington supplied the parabolic trough reflectors for SEGS I through IX and is active in solar
thermal feasibility studies and development activities.
www.flabeg.com
Flagsol
Flagsol GmbH is a German joint venture between Solar Millennium AG and Ferrostaal AG to
develop parabolic trough solar thermal plants 50 to 250 MW in size, including the Andasol
plants in Spain and the Kuraymat plant in Egypt.
www.flagsol-gmbh.com/flagsol/index.html
Fluor
6700 Las Colinas Blvd., Irving, Texas. Fluor is active in solar, wind, geothermal, and biomass
activities around the world. In the solar field, it develops both photovoltaic and solar thermal
projects. In August 2009, Fluor won a contract to design eSolar’s 46-MW central receiver plants,
and, in February 2010, signed an agreement to develop a 50-MW CSP plant in Spain.
7-49
10581090
Solar Thermal
Infinia
6811 West Okanogan Place, Kennewick, Washington 99336, is a developer of small-scale
(3-kW) dish-Stirling systems and has recently begun marketing them for utility-scale
applications.
www.infiniacorp.com
IST Abengoa Solar Inc. (Formerly, Industrial Solar Technology Corp.)

11500 West 13th Ave, Lakewood, Colorado 80215. IST designs, manufactures, installs and
operates large-scale parabolic trough collector systems for industrial and commercial water
heating, steam generation, and absorption cooling.
www.industrialsolartechnology.com
Nexant Corp.
44 Montgomery St., San Francisco, California, 94104. In addition to its California headquarters,
Nexant has offices in Colorado, London, and Bangkok. Nexant was a key partner in early solar
power tower projects, including Solar Two, and designed the molten-salt storage system for the
Andasol trough plants in Spain.
www.nexant.com
NextEra Energy Resources

P.O. Box 14000, Juno Beach, Florida 33408. Formerly FPL Energy and renamed in January
2009, NextEra Energy Resources owns the SEGS parabolic trough plants in California.
www.nexteraenergyresources.com
Paneltec Corp.
Lafayette, Colorado 80026. Paneltec is involved in the development and manufacture of mirror
facets using sandwich panel technology. The company custom-laminates honeycomb and foam
core panels for a variety of scientific and industrial applications.
www.panelteccorp.com
Parsons
100 West Walnut St., Pasadena, California 91124. Engineering and construction.
www.parsons.com
R.W. Beck
1125 17th St., Suite 1900, Denver, Colorado 80202.
www.rwbeck.com
Schott Solar
SCHOTT AG, Hattenbergstr. 10, 55122 Mainz, Germany.
www.schott.com/solar/
SkyFuel Inc.
10701 Montgomery Blvd. NE, Ste. A, Albuquerque, New Mexico 87111, is the developer of
parabolic trough systems using low-cost non-glass reflectors.
www.skyfuel.com
7-50
10581090
Solar Thermal
SENER
With offices in Spain, Portugal, Argentina, Mexico, the U.S., and Poland, is an engineering,
consultancy and systems integration company currently involved in several major solar thermal
plant projects in Spain and the U.S.
www.sener.es
Siemens
In October 2009, the large German electrical equipment manufacturer Siemens acquired Solel
Solar Systems, which owned most of the assets of Luz Industries and maintained a solar thermal
production line and test site. Siemens is very active in both solar thermal and photovoltaic
research, development and demonstration.
http://www.energy.siemens.com/hq/en/power-generation/renewables/solar-power/
Solar Millennium
Naegelsbachstr. 40, D-91052 Erlangen, Germany. Solar Millennium offers services for the
construction and operation of large-scale solar-thermal power plants. They specialize in
parabolic trough and solar chimney power plants. A U.S. subsidiary, Solar Millennium LLC, is
based in Oakland, California. In 2009, Solar Millennium entered into a joint venture with MAN
Ferrostaal named Solar Trust of America LLC, which has several projects lined up in the
United States, including power purchase agreements with Nevada Energy and Southern
California Edison.
www.solarmillennium.de
www.solartrustofamerica.com/
SolarReserve, LLC
2425 Olympic Blvd, Ste 6040W, Santa Monica, California 90404, has developed a reference
plant configuration for a concentrating solar power tower project using the proprietary molten
salt receiver, thermal storage and delivery system developed by Pratt & Whitney Rocketdyne,
Inc. and Hamilton Sundstrand Corporation, both of which are wholly owned subsidiaries of
United Technologies Corporation. SolarReserve is owned by USRG Power & Biofuels Fund II,
L.P. and Hamilton Sundstrand Corporation and was formed to develop utility-scale solar power
projects using this technology.
www.solar-reserve.com
Stirling Energy Systems (SES)

Biltmore Lakes Corporate Center, 2920 E. Camelback Road, Suite 150, Phoenix, Arizona 85016.
SES has used McDonnell Douglas intellectual property and partnered with national laboratories
to develop a dish/Stirling engine unit that it plans to use in 800 to 1750 MW of power plants for
power-purchase agreements announced in 2005. In May 2008, NTR plc of Dublin Ireland
acquired controlling interest in SES and founded Tessera Solar.
www.stirlingenergy.com
Tessera Solar
Tessera Solar North America, headquartered in Houston, Texas and Tessera Solar International,
based in London, United Kingdom, are exclusively responsible for global deployment of the
SunCatcher™ solar dish Stirling system manufactured by SES.
www.tesserasolar.com
7-51
10581090
Solar Thermal
WorleyParsons
With offices in 38 countries, WorleyParsons provides professional engineering, construction and
operation services to solar projects throughout the world. The company is technology-neutral,
with over 5000 MW of solar thermal under development.
www.worleyparsons.com
In addition to the vendors and organizations listed above and in the technology monitoring
guide (Table 7-2), the reader can visit the following Internet websites. Note that while these sites
contain information and contacts that may be of interest, EPRI cannot vouch for the accuracy or
quality of all of their content
U.S. DOE Energy Office of Efficiency and Renewable Energy

www.eere.energy.gov
U.S. DOE Concentrating Solar Power Program
The U.S. DOE administers the SunLab program through two of its national laboratories:
Sandia National Laboratory in New Mexico, and the National Renewable Energy Laboratory
in Colorado. SunLab operates the National Solar Thermal Test Facility.
http://www.sandia.gov/csp/csp_r_d_sandia.html
U.S. DOE/USA Solar Trough Initiative
Funded by the U.S. DOE and operated by SunLab, TroughNet is a comprehensive source for
government efforts in solar thermal technology, projects, research and development, and other
related resources.
www.nrel.gov/csp/troughnet/
U.S. National Renewable Energy Laboratory (NREL)
www.nrel.gov/csp/
American Solar Energy Society (ASES)

ASES is a national organization dedicated to advancing the use of solar energy for the benefit
of U.S. citizens and the global environment. The society sponsors conferences and publishes a
magazine and white papers on the subject. ASES is the U.S. section of the International Solar
Energy Society (ISES).
www.ases.org
www.ises.org
Solar Energy Industries Association (SEIA)

Established in 1974, SEIA is the national trade association of the U.S. solar energy industry. As
the voice of the industry, SEIA works with its 1000 member companies to make solar a
mainstream and significant energy source by expanding markets, removing market barriers,
strengthening the industry and educating the public on the benefits of solar energy.
www.seia.org
7-52
10581090
Solar Thermal
Solar Electric Power Association (SEPA)

Formerly the Utility PhotoVoltaic Group (UPVG), SEPA is a non-profit collaboration
comprising more than 700 utilities, energy service providers, and the PV industry working to
create and encourage commercial use of new solar electric power technology and business
models. SEPA also created and maintains the “Solar Data and Mapping Tool,” a web-based
utility that provides project data for solar power projects in the United States
(www.solarelectricpower.org/solar-tools/solar-data-and-mapping-tool.aspx).
www.solarelectricpower.org
SolarPACES
Solar Power and Chemical Energy Systems (SolarPACES) is an international cooperative
organization managed by the International Energy Agency. Member countries (currently 16)
develop solar thermal technologies by addressing technical problems associated with
commercialization and market development.
www.solarpaces.org
ESTELA
The European Solar Thermal Electricity Association (ESTELA) is an organization of developers,
manufacturers, utilities, engineering firms, and research institutions created to support the
emerging solar thermal industry in Europe.
www.estelasolar.eu
7.9 References
1. Renewable Energy Technology Characterizations. EPRI and U.S. Department of Energy:
2. A. Zavoico, W. Gould, B. Kelly, and I. Pastoril, “Solar Power Tower (SPT) Design
Innovations to Improve Reliability and Performance—Reducing Technical Risk and Cost,”
presented at the ASME 2001 International Solar Energy Conference, Washington, D.C.:
April 2001.
3. Kelly, B., Herrmann, Ulf, and Hale, Mary Jane, “Optimization Studies for Integrated Solar
Combined Cycle Systems,” presented at the ASME 2001 International Solar Energy
Conference, Washington, D.C.: April 2001.
4. Stone, K., Liden, B., Ellis, E., Sattar, T., and Mancini, T., “SES/Boeing Dish Stirling
System Operation,” presented at the ASME 2001 International Solar Energy Conference,
Washington, D.C.: April 2001.
5. Diver, R., Andraka, C., Rawlinson, K., Thomas, G., and Goldberg, V., “The Advanced
Dish Development System Project,” presented at the ASME 2001 International Solar Energy
Conference, Washington, D.C.: April 2001.
6. Mayette, J., Davenport, R., and Forristall, R., “The Salt River Project SunDish-Stirling
System,” presented at the ASME 2001 International Solar Energy Conference, Washington,
D.C.: April 2001.
7. Holl, R.J., Status of Solar-Thermal Electric Technology. EPRI, Palo Alto, CA: 1989.
GS-6573.
7-53
10581090
Solar Thermal
8. DeMeo, E., Solar-Thermal Electric Power: 2003 Status Update. EPRI, Palo Alto, CA:
March 2003. 1008463.
9. Price, H., Lupfert, E., Kearney, D., Zarza, E., Cohen, Gl, Gee, R., and Mahoney, R.,
“Advances in Parabolic Trough and Solar Power Technology,” Journal of Solar Energy
Engineering, Vol. 124: May 2002.
10. Stirling Engine Assessment. EPRI, Palo Alto, CA: October 2002. 1007317.
11. Leitner, A. “Fuel from the Sky: Solar Power’s Potential for Western Energy Supply,”
NREL/SR-550-32160. Golden, CO: National Renewable Energy Laboratory, July 2002.
12. New Mexico Central Station Solar Power: Final Summary Report. EPRI, Palo Alto, CA
and the Public Service Company of New Mexico, El Paso Electric, San Diego Gas &
Electric, Southern California Edison, Tri-State Generation & Transmission Association,
and Xcel Energy: 2008. 1016344.
13. “Case Study: Gemasolar Central Tower Plant,” SENER Engineering and Systems Inc.,
presentation at Concentrated Solar Power Summit conference, San Francisco, June 2010.
14. Utility-scale Parabolic Trough Solar Systems—Performance Acceptance Test Guidelines,
NREL: August 2010. Draft report.
15. Solar Thermal Electric Technology: 2009, EPRI, Palo Alto, CA: June 2010. 1021386.
16. Concentrating Solar Thermal Technology, EPRI, Palo Alto, CA: March 2009. 1018577.
17. Australian Sustainable Energy: Zero Carbon Australia Stationary Energy Plan, Energy
Research Institute, University of Melbourne: July 2010
18. Solar Energy Technologies Program Peer Review, DOE Energy Efficiency and Renewable
Energy (EERE), Washington D.C.: May 2010.
19. Solar Thermal Technology Status and Performance and Cost Estimates—2010, EPRI, Palo
Alto, CA: 2010. 1022504.
7-54
10581090
8
OCEAN TIDAL ENERGY
Installed Tidal • About 3 MW worldwide; 60 kW in the United States.

Capacity • Estimated annual incremental U.S. capacity additions:
(as of November
2010) 2011: 5 MW
2012: 35 MW
• Estimated cumulative capacity by 2025: 500 MW.
Tidal Energy • TISEC is an emerging technology. Marine Current Turbine (UK), Open
Conversion (TISEC) Hydro (Ireland), Verdant (USA), Ocean Renewable Power Corporation
Technology (USA), New Energy (Canada), and Clean Current (Canada) prototype
Readiness devices were tested in 2010.
(as of October 2010) • Marine Current Turbines deployed the world’s largest grid-connected pre-
commercial prototype, the 1.2-MW SeaGen in Strangford Narrows,
Northern Ireland in April 2008. The device first delivered power to the
local grid in December 2008, and has since delivered more than 2 GWh
of power to the grid, thereby achieving a 66% capacity factor.
• Verdant pulled its redesigned turbines out of the river at its Roosevelt
Island Tidal Energy (RITE) site in the East River, N.Y. They plan to
submit an application for a Pilot Project License to FERC, allowing
commercial operation. If granted, the license will allow Verdant Power to
build out the RITE Project to 30 turbines.
• Ocean Renewable Power Corp. received $10 million in funding from DOE
(to be matched by $11.1 million in non-federal funding) to build, install,
operate, and monitor a commercial-scale array of five grid-connected
devices in Cobscook Bay, Eastport, Maine.
• Nova Scotia Power deployed a 1-MW Open Hydro tidal power turbine in
the Bay of Fundy’s Minas Passage in November 2009. The turbine was
pulled from the water in May 2010, prior to its planned removal in the fall
of 2010, when it was discovered that two turbine blades had broken off.
• In August 2010, the European Marine Energy Centre (EMEC) added two
test berths and more than two miles of additional cabling needed for
device hookup, bringing the total number of test berths to seven.
• Atlantis Resources deployed its 1.2-MW AK-1000 tidal turbine at EMEC
in August 2010.
• Snohomish County Pubic Utility District (SnoPUD) received $10 million in
funding from DOE (to be matched by $10.1 million in non-federal funding)
to deploy, operate, monitor, and evaluate two 10-m Open Hydro turbines
at SnoPUD’s Admiralty Inlet Pilot Tidal Demo Project.
• In November 2010, Neptune Renewable Energy Ltd announced that it
successfully completed in-water tests of a full-scale prototype of its
Proteus NP1000 tidal stream power generator.
8-1
10581090
Ocean Tidal Energy
Economic Status • EPRI economic feasibility studies indicate TISEC technology is

competitive at a half dozen or so “very good” sites in the United States
(average annual power density greater than 2 kW/m2 and meeting other
required site attributes).
Environmental • Proper care in siting, installation, operation, and decommissioning may
Impact enable tidal energy technology to be one of the more environmentally
benign electricity generation technologies.
• Results from numerical modeling, laboratory flume studies, and field
monitoring conducted in 2009 and 2010 suggest that adverse interactions
between fish and individual turbines are unlikely to be significant.
• Pilot demonstration testing is needed to better understand the interactions
between the devices and their environment. Adaptive management will be
used to incorporate new information into decision-making processes that
will address project build out and cumulative effects.
Regulatory Status • The Federal Energy Regulatory Commission (FERC) has primary
jurisdiction for licensing tidal in-stream power plants under the Federal
Power Act (FPA). The time, cost, and complexity of the U.S. regulatory
process can be difficult for tidal project developers.
• There are currently 16 active, FERC-issued Preliminary Permits for tidal
projects as of October 3, 2010, seven of which were filed in 2010; there
are two Preliminary Permit applications pending as of October 4, 2010.
• A FERC decision is pending on an application for a 5-MW exemption from
licensing for the TideWorks Project, Sasanoa River, Maine.
Government • European governments (particularly in the UK, Ireland, Portugal and
Support of Tidal Denmark) as well as those in Japan, New Zealand, and Australia support
Energy Technology the development of TISEC technology and are now providing subsidies to
stimulate a commercial market.
• DOE initiated a Waterpower R&D Program in FY 2008 with a
Congressionally mandated $10 million, which was followed by another
Congressionally mandated $40 million for FY 2009 and $50 million for FY
2010. The majority of these funds have been allocated to marine and
hydrokinetic technologies.
Trends to Watch • SnoPUD filed a Draft License Application with FERC on December 28,
2009, and is progressing towards a pilot plant installation in 2011–2012,
with $10 million in funding from DOE and an additional $10.1 million in
non-federal funding.
• Verdant Power will apply to FERC for a commercial Pilot Project License
by the end of 2010.
• The Fundy Ocean Research Centre for Energy (FORCE) has received $20
million from the government of Canada, and has signed an $11 million
contract for the production and installation of four subsea cables in its
Minas Passage test site in 2011. The cables, with a combined capacity of
64 MW, will connect tidal devices to the Nova Scotia grid
• Additional megawatt-scale demonstration and early commercial projects
are expected to be deployed over the next decade in Canada, Europe,
Korea, Australia, New Zealand, and the United States.
• Major players in the wind and conventional hydropower industries are
acquiring significant stakes in hydrokinetic device development and
manufacturing companies. The technical and business experience they
bring to hydrokinetic generation may accelerate the development of the
tidal generation industry.
8-2
10581090
Ocean Tidal Energy
8.1 Introduction
The history of tidal power goes back to at least the Middle Ages. Primitive tidal mills were
operated in England in the 11th century. Numerous tidal mills were built in Western Europe,
Canada, and the United States during the 18th century. As a result of the industrial revolution
and subsequent widespread use of electric power, tidal mills became obsolete.
Actual construction of “first-generation” modern tidal power plants (dam or barrage systems) has
been limited to one large, 200-MW facility at LaRance, France; a small, 20-MW experimental
plant at Annapolis, Nova Scotia that is still operating today; and a number of small experimental
plants in Russia and China.
“Second generation” or in-stream tidal energy technology (i.e., using the kinetic energy from
unimpounded tidal currents) is a new technology which has heightened interest in the
commercial development of tidal energy conversion technology in Europe, Korea, Australia,
New Zealand, Canada, and the United States over the past five years. This chapter addresses
second-generation tidal in-stream energy conversion (TISEC).
The tides are the result of gravitational forces exerted by the Moon and the Sun on the oceans of
the Earth, with the Moon having the predominant influence. The changing relative position of
these bodies causes the surface of the oceans to rise and fall periodically, as illustrated in Figure
8-1. The gravitational forces of the Sun and the Moon and the centrifugal forces due to the
rotation about the Earth-Moon center-of-gravity create two “bulges” in the Earth’s oceans: one
closest to the Moon, and the other on the opposite side of the globe, as illustrated in Figure 8-2.
These “bulges” result in two tides per day, called semi-diurnal tides—the dominant tidal pattern
in most of the world’s oceans.
TISEC sites are typically located where there is a narrow channel or passage between two land
masses or an inlet into a bay, through which substantial volumes of tidal water must flow at high
speed. Tidal power sites should also have suitable seabed geology for proper anchoring of the
TISEC devices, and should be situated reasonably close to existing grid interconnection points.
Figure 8-1
Earth, Moon & Sun’s Influence on Tides
8-3
10581090
Ocean Tidal Energy
Figure 8-2
The Earth Tidal Bulge
An illustration of a typical open-rotor, horizontal-axis water turbine TISEC device is shown in

Figure 8-3. For such a water turbine, the hydrokinetic power per unit cross sectional area (in this
case, the area of the rotor blades) is equal to one half the density of water multiplied by the cube
of the water speed. This relationship is illustrated in Figure 8-4.
Figure 8-3
Typical Hydro Kinetic Water Turbine
(P/A)flow = 0.5 x density of water x V3
V
A
Figure 8-4
Tidal Power Density
8-4
10581090
Ocean Tidal Energy
8.2 U.S. Tidal In-Stream Energy Highlights: Mid-2009 to Late-2010

Interest in ocean tidal in-stream renewable energy continues to grow in the United States. This
section provides a brief account of notable federal and state developments as well as device
developer activities within the sector during the second half of 2009 through late 2010.
8.2.1 Federal and State Highlights
8.2.1.1 Federal-Level Activities
In 2010, the U.S. Congress appropriated $50 million to conduct research and development on
advanced water power energy generation technologies, including both marine and hydrokinetic
technologies (wave, tidal, ocean current, in-stream hydrokinetic and ocean thermal), as well as
conventional hydropower (any technology that uses a dam or diversionary structure). This marks
a significant increase in funding from the $10 million appropriated in 2008, which was the first
year wave and marine hydrokinetic power research was supported by DOE.
The DOE’s wave and marine hydrokinetic power research is focused on assessing the potential
recoverable energy from these resources in the United States and facilitating the development
and deployment of technologies to fully realize this potential. Wave and marine hydrokinetic
technologies represent an opportunity for the United States to engage directly in an emerging
area of science and discovery, while developing an entirely new suite of renewable energy
technologies available to reduce emissions and help states meet renewable energy portfolio
standard (RPS) targets.
The DOE’s priorities for tidal power include:

• Facilitating the deployment of prototypes, and collecting data on the energy conversion
performance and environmental impacts of the devices;
• Determining the available, extractable, and cost effective resources in the United States;
• Characterizing and comparing the wide variety of existing marine hydrokinetic technologies;
• Improving technology performance and reliability, and reducing technology development
costs; and
• Minimizing the cost, time, and negative impacts associated with siting projects.
In 2008, the majority of DOE funding for tidal power was awarded to specific technology and
project development efforts, selected through a competitive process. Those awarded efforts
included:
• The design, fabrication and testing of an improved turbine blade design structure for Verdant
Power, Inc.
• A program to conduct in-water testing and demonstration of tidal flow technology as a first
step toward deploying a commercial tidal power facility by the Snohomish Public Utility
District (SnoPUD), a municipal utility in Washington State.
8-5
10581090
Ocean Tidal Energy
• The selection and funding of two National Marine Renewable Energy Centers—one at the
University of Hawaii, and a second run jointly by Oregon State University and the University
of Washington. Further, the DOE funded a market acceleration program, consisting of
nationwide resource wave and tidal hydrokinetic resource assessments and a collaborative
project to address navigation and environmental issues as well as clarify the permitting
process.
In 2009, DOE support for tidal power included funding of:

• An EPRI project to conduct desktop modeling studies of strike probability and flume testing
of fish interactions with hydrokinetic turbines
• Various projects related to siting and evaluation of tidal energy projects.
8.2.1.2 State of Alaska Projects
EPRI estimates that Alaska possesses over 90% of the U.S. tidal hydrokinetic energy resource. A
key limitation to the use of these resources for electricity generation is that most of the resource
is found in remote areas and not adjacent to any electric transmission infrastructure necessary to
provide power export capabilities at significant scales.
ORPC Cook Inlet Project (P-12679). On April 1, 2010, Ocean Renewable Power Corp. (ORPC)
filed a Preliminary Permit application for the proposed Cook Inlet Tidal Energy Pilot Project,
with a 5-MW capacity.
Turnagain Arm Tidal Energy Project (P-13509). Turnagain Arm Tidal Energy Corporation was
issued a Preliminary Permit for its project of the same name on February 5, 2010. The proposed
project consists of 8-mile long and 7.5-mile long tidal fences containing 220 10-MW Davis
Turbines (a modification of the Darrieus, cross-flow turbine); total capacity is 2200 MW. The
project site is located in Cook Inlet on the Kenai Peninsula, Alaska.
Icy Passage Tidal Project (P-13605). Natural Currents Energy Services was issued a
Preliminary Permit for its Icy Passage Tidal Project on April 30, 2010. This 300-kW project
would be located in Icy Passage, near Gustavus, Alaska.
Gastineau Channel Tidal Project (P-13606). Natural Currents Energy Services was issued a
Preliminary Permit for its Gastineau Channel Tidal Project on April 30, 2010. This 400-kW
project would be located in Gastineau Channel, near Juneau, Alaska.
East Foreland Tidal Energy Project (P-13821). ORPC filed an application for a Preliminary
Permit with FERC on August 2, 2010. The 5-MW project would be located in Cook Inlet near
Nikiski, Alaska.
Killisnoo Tidal Energy Project (P-13823). Natural Currents Energy Services filed an application
for a Preliminary Permit with FERC on August 5, 2010. The 250-kW project would be located in
Kootznahoo Inlet, northeast of Killisnoo Island, near the City of Angoon, in southeastern Alaska.
8-6
10581090
Ocean Tidal Energy
8.2.1.3 State of Washington Projects
Admiralty Inlet Tidal Energy Project (P-12690). On March 2, 2010, the Public Utility District
No. 1 of Snohomish County, Washington (SnoPUD) filed an application with FERC to renew its
Preliminary Permit for its Admiralty Inlet Tidal Energy Project. FERC issued the renewed
Preliminary Permit on July 8, 2010. This project would be located in Admiralty Inlet in the
northwestern portion of Puget Sound, between the Olympic Peninsula and Whidbey Island,
Washington. The proposed project consists, in part, of two 10-m, 500-kW capacity turbines to be
manufactured by Open Hydro. In October 2008, the DOE provided a $1.2 million grant to the
utility to support the tidal energy effort at Admiralty Inlet, and in September 9, 2010, DOE
announced that it had awarded the project an additional grant of $10 million in funding, which
will be matched by $10.1 million in non-federal funding. SnoPUD submitted its Draft License
Application with FERC on December 28, 2009, and is working with resource agencies and other
stakeholders to prepare and submit its Final License Application and associated documents by
March 31, 2011. SnoPUD expects to begin deploying the turbines in late 2011.
Deception Pass Tidal Energy Project (P-12687). FERC issued a Preliminary Permit for
SnoPUD’s Deception Pass Tidal Energy Project on August 4, 2010. The proposed project would
consist, in part, of four 20-m, 1.6-MW horizontal-axis turbines and would be located in Puget
Sound within Deception Pass, between Whidby Island and Fidalgo Island. The Preliminary
Permit is a reissuance of a Preliminary Permit held by SnoPUD. Under the previous Preliminary
Permit, SnoPUD filed a pre-application document (PAD) for the project on January 31, 2008.
SnoPUD engaged in other activities, such as consultation with resource agencies and other
stakeholders, that were directed toward submission of a License Application. Under the reissued
Preliminary Permit, SnoPUD proposes to submit a Draft License Application for the Deception
Pass Tidal Energy Project by August 4, 2012.
U.S. Navy Demonstration Project in Puget Sound. Through a Congressionally directed and
funded project, the Navy will install turbines from Verdant Power for a demonstration
installation in Puget Sound. Unlike the Snohomish PUD project, however, the primary goal of
the Navy’s pilot project is aimed at furthering the state of tidal energy technology by gathering
operational and environmental assessment data. The Navy plans to temporarily place three tidal
turbines in the waters on the west side of Admiralty Inlet off Marrowstone Island. The turbines,
which are to be installed in early fall of 2011 or 2012, will power two buildings and a parking lot
at the Navy's ammunition depot on Indian Island, near Port Townsend.
8.2.1.4 State of California Projects
San Francisco Bay Tidal Energy Project II (P-12585). On September 30, 2008, Golden Gate
Energy (GGE) filed with FERC an application for a Pilot Project License, which FERC denied.
On October 1, 2008, GGE was reissued Preliminary Permit to evaluate the feasibility of the
project and prepare a license application. Once again, on February 4, 2010, GGE was reissued a
Preliminary Permit. On September 1, 2010, GGE was sent a notice of probable cancellation of
the Preliminary Permit, because GGE had not filed the mandated six-month progress report;
however, GGE filed that progress report on September 10, 2010. The future of this project is
unknown to EPRI.
8-7
10581090
Ocean Tidal Energy
8.2.1.5 State of New York Projects
Roosevelt Island Tidal Energy (RITE) Project (P-12611). Since 2002, Verdant Power has been
gathering operational and environmental data from turbines at its Roosevelt Island site in the
East River, N.Y. The current Preliminary Permit, issued February 17, 2009, encompasses a
proposed array of 30 turbines with an installed capacity of 1 MW in the East Channel, and a
proposed array of 100 turbines with an installed capacity of 4 MW in the West Channel. Verdant
plans to submit a Final License Application for a commercial Pilot Project License by the end of
2010. During 2010, Verdant re-evaluated the design of its 5-m turbine, and consulted with
participating agencies regarding monitoring of project impacts.
The RITE project has received key support from the New York State Energy Research and
Development Agency (NYSERDA).
Astoria Tidal Energy Project (P-13730). On May 12, 2010, New York Tidal Energy Company
filed an application with FERC for a Preliminary Permit for its Astoria Tidal Energy Project in
an area of the East River in New York extending from north of Roosevelt Island through an area
known as Hell Gate, to Stony Point. The permit application specifies 50 to 150 turbine units of
0.5 to 2 MW each, for a maximum installed capacity of 300 MW.
East River Tidal Energy Pilot Project (P-12665). New York Tidal Company filed an application
for a Preliminary Permit on May 12, 2010. The location of the proposed project is east of Wards
Island at Hell Gate in the East River, southwest of the Triborough Bridge, in New York City,
N.Y. The proposed project would have a capacity of 20 kW.
8.2.1.6 State of Massachusetts Projects
Massachusetts Ocean Management Plan. The Commonwealth of Massachusetts finalized its

Ocean Management Plan in December 2009. The plan has four goals:
1. Balance and protect the natural, social, cultural, historic, and economic interests of the
marine ecosystem through integrated management.
2. Recognize and protect biodiversity, ecosystem health, and the interdependence of
ecosystems.
3. Support wise use of marine resources, including renewable energy, sustainable uses, and
infrastructure.
4. Incorporate new knowledge as the basis for management that adapts over time to address
changing social, technological, and environmental conditions.
The plan states that goal number 3 above, “was achieved by identifying use areas and
promulgating enforceable management measures that: identify locations and performance
measures for allowable uses and infrastructure; require renewable energy development of
appropriate scale, minimize conflicts with/impact to existing uses and resources, develop
measures for reconciling use conflicts with fisheries, and streamline permitting.”
8-8
10581090
Ocean Tidal Energy
New England Marine Renewable Energy Center, Massachusetts Dartmouth Marine Energy
Research Center. On September 15, 2010, the New England Marine Renewable Energy Center
announced that it had received a federal grant of $750,000 for annual operations and $750,000
for study of advance techniques for assessing offshore wind and hydrokinetic renewable energy
resources. The researchers come from several local universities and institutions, including the
University of Massachusetts, MIT, the Woods Hole Oceanographic Institute, the University of
New Hampshire, and the University of Rhode Island.
Part of the center’s goal is also to support early-stage ventures to commercialize technology spun
out of university research. It opened a clean energy laboratory in July 2009 to measure and test
prototype marine generation technologies. The center provides incubation space to three marine
energy firms: wave power technology developer Resolute Marine Energy Inc., tidal power
developer Ocean Renewable Power Corp., and underwater unmanned vehicle developer Ocean
Server Technology Inc.
Cape Cod Tidal Energy Project (P-13828). Free Flow Power filed an application with FERC for
its Cape Cod Tidal Energy Project on August 9, 2010. The project area is the Cape Cod Canal,
near the Massachusetts towns of Sandwich, Sagamore, and Buzzards Bay. The project has a
proposed capacity of up to 20 MW, comprising up to 2,000 10-kW turbines, each 3 m in
diameter.
8.2.1.7 State of Maine Projects
Ocean Renewable Power Corp. received $10 million in funding from DOE (to be matched by
$11.1 million in non-federal funding) to build, install, operate, and monitor a commercial-scale
array of five grid-connected devices in Cobscook Bay, Eastport, Maine. An award of $1.2
million to ORPC from the Maine Technology Institute to support that work was announced
October 14, 2010.
In September, 2010, the U.S. Department of Commerce awarded $1.4 million to the City of
Eastport and Washington County to establish the Maine Marine Energy Center, an advanced
composite material manufacturing facility that can produce the complex components and sub-
assemblies required by the ocean energy industry.
On April 9, 2009, a $951,500 federal appropriation to the University of Maine was announced.
Researchers there will lead a collaborative effort to develop Maine's tidal power resource. The
funding comes from a Congressional initiative developed by the Maine delegation, and will be
used to assess current prototypes and turbine models that can be submerged in the ocean to
produce power using tidal currents. Although the University of Maine will lead the project,
Maine Maritime Academy and Portland-based ORPC are partners in the ongoing research.
Researchers will also move forward to evaluate the potential environmental impact of harnessing
tidal energy off the coast of Eastport in the Western Passage of Passamaquoddy Bay.
In late summer 2009, ORPC began extensive in-water testing of the commercial design of its
turbine generator unit.
8-9
10581090
Ocean Tidal Energy
ORPC Cobscook Bay Tidal Energy Project (P-12711) and Western Passage Tidal Energy
Project (P-12680). ORPC filed a Draft Hydrokinetic Pilot License Application with FERC on
July 24, 2009 for both projects. In preparation for writing this application, ORPC developed
environmental study plans for its Eastport sites and has been collecting data from the sites
regarding tidal current velocity, marine geophysical characteristics, benthic characteristics and
terrestrial site review. ORPC filed applications for reissuance of the Preliminary Permits for
these projects on July 1, 2010.
ORPC completed fabrication, assembly and shop testing of its 60-kW, beta pre-commercial tidal
generator unit (TGU) in early 2010 and the device has been deployed in Cobscook Bay since
March 2010.
Town of Wiscasset Tidal Resources Project (P-13329). FERC issued the Town of Wiscasset a
Preliminary Permit for the Wiscasset Tidal Resource Project on May 28, 2009. The Chewonki
Foundation submitted the second six-month progress report on behalf of the Wiscasset Tidal
Resources Project on May 27, 2010. Project activities included stakeholder outreach, discussion
with University of Maine about partnership for site characterization and initial site
characterization. The Preliminary Permit cites 4 to 40 OCGen hydrokinetic turbines with a total
installed capacity of 1 to 10 MW; however, the project proponents anticipate applying for a Pilot
Project License, which limits the project capacity to 5 MW.
Homeowner Tidal Power Electric Generation Project (P-13345). On July 1, 2009, FERC issued
a Preliminary Permit for the Homeowner Tidal Power Electric Generation Project to be located
on the Kennebec River in Sagadahoc County, near Phippsburg, Maine. The proposed project
would comprise six hydrokinetic turbines with a combined capacity of 60 kW, in addition to
transmission lines connected directly to individual homes and appurtenant facilities. As of the
end of 2009, the project applicant anticipated pursuing a license exemption.
Damariscotta River Hydrokinetic Tidal Energy Project (P-13646). FERC issued a Preliminary
Permit for the Damariscotta Tidal Project on May 19, 2010. The proposed project would be
located on the Damariscotta River in Lincoln County, Maine. The project would have a total
installed capacity of 250 kW provided by 10 to 20 EnCurrent hydrokinetic generator units.
Kendall Head Tidal Energy Project (P-13801). On September 24, 2010, FERC accepted
ORPC’s application for a Preliminary Permit for the Kendall Head Tidal Energy Project. The
project would consist of four OCGen hydrokinetic tidal devices, each consisting of two 150-kW
turbine generator units for a combined capacity of 1.2 MW; an anchoring support structure; a
mooring system; a 2,700-foot submersed cable connecting the turbine generating units to a shore
station; an 8,500-foot, 34.5-kV transmission line connecting the shore station to an existing
distribution line; and appurtenant facilities. The project would be located in the Western Passage,
Washington County, Maine.
8.2.1.8 State of New Hampshire Projects
General Sullivan and Little Bay Bridges Tidal Energy Project (P-13503). The University of
New Hampshire Center for Ocean Renewable Energy (UNH/CORE) filed a Preliminary Permit
application for the General Sullivan and Little Bay Bridges Tidal Energy Project to be located on
8-10
10581090
Ocean Tidal Energy
the Piscataqua River in Rickingham and Stafford Counties, New Hampshire. FERC issued the
Preliminary Permit on September 30, 2009. The proposed project comprises a single 10-foot-
wide by 35-foot-long test platform suspending a variety of hydrokinetic devices, including a
Gorlov Helical turbine generation unit into the river, plus appurtenant facilities. Because
electricity generated by the system would power the test platform, the project would require no
transmission line. UNH/CORE intends to develop the site as a nationally recognized testing and
evaluation facility for marine hydrokinetic turbines, with no intention to connect with the electric
grid. Thus, UNH/CORE expects to seek an exemption from a FERC license.
During 2010, UNH/CORE used vessels of opportunity to acquire current velocity measurements
with an Acoustic Doppler Current Profiler (ADCP). Test platform design and fabrication activity
accelerated as funding became available. The new platform will be 64 feet long and 34 feet wide,
and will enable deployment of devices up to 4 m in diameter. The platform is expected to be
ready for deployment by late spring 2011, and initial testing of a turbine may occur in the
spring/summer of 2011.
8.2.1.9 State of New Jersey Projects
Hoffmans Marina Tidal Energy Project (P-13682). Natural Currents Energy Services filed an
application for a FERC-issued Preliminary Permit on May 5, 2010, for the Hoffmans Marina
Tidal Energy Project, which would be located on Manasquan River, in Monmouth County, New
Jersey. The proposed project would consist of 10 20-kW turbines with a combined capacity of
200 kW and would be operated to power the marina and associated office building on site.
Highlands New Jersey Tidal Energy Project (P-13725). Natural Current Energy Services filed
an application for a Preliminary Permit on May 6, 2010, for the Highlands New Jersey Tidal
Energy Project, which would be located on the Shrewsbury River, in Monmouth County, New
Jersey. The project would utilize 20 150-kW Natural Currents Red Hawk Tidal In-Stream
Energy Conversion (TISEC) modules with a total combined capacity of 3 MW.
8.2.2 U.S. Developer Highlights
8.2.2.1 Verdant Power
Verdant completed Phase 2 of its Roosevelt Island Tidal Energy (RITE) project in late 2008, for
which it deployed its Generation 4 turbine. Through the use of 24 hydroacoustic transducers and
a DIDSON high definition sonar, evidence was gathered that showed minimal environmental
impacts. Separately, Verdant experienced and solved failures of rotors blades and hubs.
Verdant has applied to FERC to install and operate 30 turbines (each turbine rated at about 34
kW for a total of roughly 1 MW). A full expansion of the project would use about 300 turbines
and has the potential to generate 10 MW.
During the period December 2009 to April 2010, Verdant Power completed efforts to re-evaluate
various elements of the Generation 5 design of its kinetic hydropower system (KHPS). Verdant
plans to submit a Final License Application by December 15, 2010.
8-11
10581090
Ocean Tidal Energy
In May, 2010, Verdant Power signed a memorandum of understanding with the China Energy
Conservation Environment Protection Group to develop tidal energy projects in China.
8.2.2.2 Vortex Hydro Energy
Michigan-based Vortex Hydro Energy LLC, founded in 2004, manufactures a technology

nicknamed VIVACE that is based on the extensively studied phenomenon of vortex induced
vibration (VIV) to extract useful energy from ocean, river, tidal and other water currents. VIV
results from vortices forming and shedding on the downstream side of a bluff body in a current.
Vortex shedding alternates from one side to the other, thereby creating a vibration or oscillation
as illustrated in Figure 8-5.
Figure 8-5
Vortex Induced Vibrations Oscillates Objects in Fluid Currents.
On July 30, 2010, Vortex Hydro Energy announced that it will deploy a prototype device in the
St. Clair River, Michigan starting in spring 2011.
8.2.2.3 Ocean Renewable Power Corporation
ORPC fabricated its 60-kW beta turbine generating unit in 2009 to early 2010 and deployed it in
March 2010 on a floating barge in Cobscook Bay, near Eastport, Maine. Operation is expected to
continue through the winter months of 2010–2011.
8.3 Worldwide Tidal In-Stream Energy Highlights: Mid-2009 to Late-2010
This section provides a brief summary of notable activities within the sector during the second
half of 2009 and through October 2010.
8.3.1 Canada
Fundy Ocean Research Centre for Energy Developments. The Fundy Ocean Research Centre
for Energy (FORCE) has received $20 million from the government of Canada, and signed an
$11 million contract for the production and installation of four subsea cables at its Minas Passage
test site in 2011. The cables, with a combined capacity of 64 MW, will connect tidal devices to
the Nova Scotia grid.
8-12
10581090
Ocean Tidal Energy
This facility will allow for the testing of in-stream turbine devices in tidal waters that are known
to possess the most demanding conditions in the world. Throughout the course of the test period,
the devices will be monitored for wear, environmental impacts and performance that will lead to
further technological developments.
Three tidal energy developers have been selected to install commercial prototype turbines at this
test facility: Clean Current of British Columbia, Open Hydro of Ireland, and Marine Current
Turbines of the UK. The first device to be installed was a 1-MW Open Hydro turbine, which was
deployed in the fall of 2009. The device was removed from the water after it was discovered that
two of its blades had broken off.
The project sites were selected via a one-year, $1-million study. They lie on sediment-free
bedrock and have a linear current flow at speeds of up to 10 m/s during ebb and flood tide, more
than enough to turn the turbines. The estimated cost to deploy each project is $15 to $18 million
and total project cost is estimated to be $70 million.
The Canoe Pass Tidal Energy Company (CPTEC) Award from the Innovative Clean Energy
Fund. CPTEC will apply the $2 million award toward the installation of two 250-kW EnCurrent
turbines, developed by New Energy Corp of Calgary Alberta, at Canoe Pass adjacent to the foot
of Seymour Narrows north of the Campbell River. As of August 2010, CPTEC had completed
the environmental assessment for its project and expected to begin construction in January 2011
with completion expected in spring 2011.
8.3.2 United Kingdom (UK)
The UK is maintaining its stature as the global leader in tidal energy technology development. In
an effort to solidify its leadership position in TISEC development, the UK has established an
installed marine energy capacity goal of 2 GW by 2020 and adopted an energy policy designed
to attract and support TISEC developers and equipment testing.
On March 16, 2010, the Scottish government and crown estate announced a £4 billion project to
build 10 tidal and wave power sites with a combined capacity of 1.2 GW in the Orkney Island
and Pentland Firth. The projects are expected to require an additional investment of £1 billion to
construct new grid connections, ports, and other support facilities. These projects were described
as the world’s first commercial wave and tidal power schemes, with capacity evenly divided
between wave- and tide-based generation. The tidal projects are:
• SSE Renewables Development Ltd, 200 MW, at Cost Head
• SSE Renewables Holding Ltd and OpenHydro Site Development Ltd, 200 MW, at Cantick
Head
• Marine Current Turbines Ltd, 100 MW, at Brough Ness
• Scottish Power Renewables UK Ltd, 100 MW at Ness of Duncansby
European Marine Energy Center Developments. The UK has a prototype test facility at the
European Marine Energy Center (EMEC) in the Orkney Islands north of Scotland that is home to
many tidal energy developers. Marine currents at the test site reach almost 4 m/sec (7.8 knots) at
8-13
10581090
Ocean Tidal Energy
spring tides. The facility offers five test berths at depths ranging from 25 m to 50 m in an area 2-
km across and approximately 4-km in length. From each developer berth, the subsea cables
follow the seabed and then pass under the beach and into an external housing next to the
substation. An adjacent building on site holds a Scottish and Southern Energy transformer where
11 kV is transformed to 33 kV. The Orkney Island of Eday, Westray, and Sandy are linked by
subsea cables that form a ring through the Northern Isles and feed into the grid.
The first developer to use the site was OpenHydro, whose open center turbine was installed in
late 2006 (see Figure 8-6). Testing took place throughout 2007 and grid connection occurred in
late 2007. In November 2007, OpenHydro successfully changed out its turbines at the Fall of
Warness. In May 2008, the company successfully supplied electricity to the National Grid during
its testing procedure, marking the first time electricity had been produced and delivered to the
National Grid by a tidal energy device in the UK. Recent developer-specific activities are
described in the following subsections; however, an important upgrade to the tidal and wave
energy infrastructure at EMEC occurred with the delivery of 11 km of 20 kV subsea cable which
allowed EMEC to expand their test sites to include three additional grid-connected test berths –
two at the tidal test site and another at the wave test site.
Figure 8-6
Open Hydro Device at the EMEC Fall of Warness Tidal Test Site
TISEC developers in the UK with mid-2009 to late-2010 updates to report are:

• Marine Current Turbines
• Pulse Tidal
• Swanturbines
• Tidal Energy Ltd
8-14
10581090
Ocean Tidal Energy
• Atlantis Resources
• Hammerfest Strøm.
8.3.2.1 Marine Current Turbines (MCT)
MCT successfully completed the deployment of its 1.2-MW SeaGen Tidal System. The MCT
prototype was deployed in April 2008 in Strangford Narrows (UK) and was the world’s first
demonstrated TISEC unit with a power rating exceeding 1 MW. The SeaGen system (Figure 8-7)
delivered electricity into the UK grid for the first time in early July 2008, and has since delivered
more than 2 GWh of power to the grid, thereby achieving a 66% capacity factor. The power
generated by SeaGen is being purchased by the Irish energy company ESB Independent for its
customers in Northern Ireland and the Republic of Ireland. SeaGen has the capacity to generate
power to meet the average electricity needs of around 1,000 homes. Siemens acquired a stake in
Marine Current Turbines in early 2010.
Figure 8-7
Marine Current Turbines SeaGenTM
8.3.2.2 Pulse Tidal
The Pulse Tidal technology is based on twin hydrofoils positioned across the tidal flow. Moving
water pushes the foils either up or down according to the angle of the foil in the water. The
vertical forces act in the same way that air moving over a wing provides lift. A conventional
generator above the water surface is driven by the foils moving below the surface.
Pulse Tidal received funding from the UK Government’s Technology Program to research and
develop a 100-kW Pulse generator in the Humber estuary close to Immingham (Figure 8-8). The
generator fed power ashore directly to supply a large chemical works.
8-15
10581090
Ocean Tidal Energy
Figure 8-8
Pulse Tidal Generator
The mean water level the Pulse Stream 100 operates in at Humber is only 9 m; with 4 m of tidal
range either side of that, the Pulse Tidal concept proved for the first time the potential for tidal
stream energy from shallow waters. Pulse Tidal has begun environmental studies at the planned
site of the world’s first commercial shallow water tidal energy site at Kyle Rhea, the strait
between the Isle of Skye and the Scottish mainland, and plans to start generating there in 2012.
8.3.2.3 Swanturbines
A 300-kW demonstration device is being assembled to be installed at the European Marine

Energy Centre (EMEC) in Orkney. The “Cygnet” device is illustrated in Figure 8-9. This £7
million project will demonstrate the robust design and installation philosophy of the
Swanturbines technology. In early April 2010, trials confirmed the feasibility of deploying
Swanturbines’ Cygnet device at EMEC.
8-16
10581090
Ocean Tidal Energy
Figure 8-9
Swanturbines Cygnet
8.3.2.4 Tidal Energy Ltd
Tidal Energy Ltd is developing the DeltaStream technology illustrated in Figure 8-10. Located in
Cardiff, Wales, the company is funded by Eco2 and Carbon Connection. DeltaStream uses the
same concept as a wind turbine together with ship propeller technology. The DeltaStream is a
nominal 1.2-MW unit which sits on the seabed without the need for a positive anchoring system,
generating electricity from three separate horizontal axis turbines mounted on a common frame.
The use of three turbines on a single triangular frame produces a low center of gravity enabling
the device to satisfy structural stability requirements, including the avoidance of overturning and
sliding. Tidal Generation Ltd deployed and connected its 500-kW tidal turbine at EMEC in 2010.
Figure 8-10
Tidal Energy Ltd DeltaStream
8-17
10581090
Ocean Tidal Energy
8.3.2.5 Atlantis Resources
Atlantis Resources unveiled its AK1000 tidal turbine on August 12, 2010, and deployed it 12
days later at EMEC in 35-m water. The twin-rotors of the device are 18 m in diameter and will
produce 1 MW at a water velocity of 2.65 m/s. The turbine will be deployed at EMEC and tested
for three years. Power generated by the device will be dispatched to the local grid.
8.3.2.6 Hammerfest Strøm
On August 17, 2010, Hammerfest Strøm, a Norwegian company, announced contracts totaling
£4 million to construct the first of 10 of its HS1000 tidal turbines at EMEC in 2011. They
deployed an Acoustic Doppler Current Profiler (ACDP) and turbulence meter at the site in
preparation for deployment of their HS1000 device in 2011.
8.3.3 Ireland
8.3.3.1 OpenHydro
The development of the company’s Open-Centre technology began in the United States during
the early 1990s, and OpenHydro has over a decade of experience in developing and testing the
Open-Centre Turbine in marine conditions. OpenHydro was formed in 2005 after negotiating
world rights to the Open-Centre technology in late 2004. The Open-Centre Turbine is designed
to be deployed directly on the seabed. It can be located at depth and presents no navigational
hazard.
OpenHydro’s installation in September 2008 (Figures 8-11 and 8-12) is the world’s first
successful deployment of a tidal turbine mounted directly on a gravity base on the seabed. The
installation took place at EMEC in Scotland.
Figure 8-11
Open Hydro Gravity Base Installation Illustration
8-18
10581090
Ocean Tidal Energy
Figure 8-12
Open Hydro Gravity Base Photo
To solve the problem of vessel availability and enable launches of the tidal power devices,
OpenHydro commissioned the world’s first specialist barge to install seabed-mounted tidal
turbines (Figure 8-13).
Figure 8-13
OpenHydro Deployment Vessel
OpenHydro unveiled its next generation 10-m diameter, 1-MW Open Centre turbine in
September 2009 and deployed that turbine in the Bay of Fundy’s Minas Passage in November
2009 in partnership with Nova Scotia Power and the Fundy Ocean Research Centre for Energy
(FORCE). The device was fabricated and shipped to the Bay of Fundy from OpenHydro’s
Technical Center at Carlingford Lough, Ireland. The turbine was pulled from the water in May
2010, prior to its planned removal in the fall of 2010, when it was discovered that two of the
turbine blades had broken off.
8-19
10581090
Ocean Tidal Energy
On October 5, 2009, Open Hydro received notice of a €2 million grant from Sustainable Energy
Ireland’s (SEI) Ocean Energy Prototype Research and Development Programme. The grant will
be used to design and develop OpenHydro’s next generation 16-m Open-Centre turbine, subsea
base, and installation barge.
On March 16, 2010, OpenHydro and its partner, SSE Renewables, were awarded exclusive rights
to develop a 200-MW tidal energy farm at the Cantick Head site in the Pentland Firth off the
north coast of Scotland. On July 6, 2010, OpenHydro was awarded a £1.85 million grant from
Scotland’s Wave and Tidal Energy: Research, Development and Demonstration Support
(WATERS) fund. The funds will be used to design, develop, manufacture, and test a power
conversion and control system for connecting arrays of tidal turbines to the grid.
8.3.4 Continental Europe
Electricite de France (EDF) Group has selected Irish company OpenHydro Group to build the
first underwater tidal turbines at Paimpol-Brehat in Brittany, France. The effort is part of its pilot
project to build a tidal turbine farm to generate electricity from tidal energy.
Announced in October 2009, the OpenHydro partnership involves installing four to 10 tidal
turbines with a total rated capacity of between 2 MW and 4 MW. These will be connected to the
electricity grid beginning in 2011. The test unit at Paimpol-Brehat will reportedly enable the
technology to be tested in real-world conditions and permit assessment of impacts on the marine
environment. In the future, other technologies tested by EDF could be deployed to take
advantage of the Paimpol-Brehat region’s powerful tidal currents.
8.3.5 Australia
Clarence Strait Tidal Energy Project. In early September, 2010, Australia’s Northern Territory
granted a site license to Tenax Energy for a proposed 200-MW project in Clarence Strait
between the northern coast of Australia and the Tiwi Islands, near the city of Darwin. The license
allows Tenax to undertake environmental studies, which are needed to obtain other regulatory
approvals. Tenax is considering OpenHydro turbines for this project.
8.4 Tidal Power and Energy Resources
EPRI has researched available data sources and developed a methodology for estimating in-
stream tidal current resource and performance [1], available at
http://oceanenergy.epri.com/streamenergy.html.
8.4.1 Data Sources
The National Ocean Service (NOS) Center for Operational Oceanographic Products and Services
(CO-OPS) collects and distributes observations and predictions of water levels and currents to
ensure safe, efficient, and environmentally sound maritime commerce. The center manages the
8-20
10581090
Ocean Tidal Energy
National Water Level Observation Network (NWLON) and a national network of Physical
Oceanographic Real-Time Systems (PORTS) in major U.S. harbors.
The CO-OPS NOS predictions page (http://co-ops.nos.noaa.gov/tide_pred.html) is separated

into two sections. The first, Tidal Predictions, provides tidal predictions at a few thousand
stations around the United States, Caribbean, and Pacific Islands. The second, Tidal Current
Predictions, provides tidal current prediction for a few hundred U.S. and Canadian tidal current
reference stations, as well as time and speed adjustment to allow the calculations to predict tidal
currents at a few thousand secondary stations. Both tide and tidal current predictions are
provided for the present calendar year. The predictions are updated in late October or early
November to include the following calendar year.
Current velocity tables may be found at http://co-ops.nos.noaa.gov/curents04/currpred.html.

Both tide and tidal current predictions are based upon analyses of observations at reference
stations. Unlike tide stations, which are normally located along the shoreline, most tidal current
stations are located offshore in channels, rivers, and bays. Tidal current stations are often named
for the channel, river, or bay in which they are located or for a nearby navigational reference
point.
Figure 8-14 shows the locations of the CO-OPS reference stations for tidal current prediction in
the U.S.
Figure 8-14
United States Tidal Current Reference Stations
8-21
10581090
Ocean Tidal Energy
8.4.2 Tidal Power Flux and Average Annual Energy
The power density of ocean tides is expressed in kilowatts per square meter (kW/m2) of area
perpendicular to the direction of flow. Annual average power density ranges up to about
2
4 kW/m for a very fast-moving tidal stream.
To estimate mean annual power density, it is first necessary to know the speed of the surface
currents as a function of time. The power density calculation must account for change in current
speed with depth, and horizontal variability across the width of the channel. This calculation
yields an estimate for the mean annual, depth-averaged, width-averaged tidal stream power
density. Figure 8-15 shows a tidal current frequency distribution for a single transect in Maine’s
Western Passage.
In 2005, EPRI performed a site survey and assessment study for two U.S. states (Maine [2]
and Massachusetts [3]), two Canadian provinces (Nova Scotia [4] and New Brunswick), and
three other U.S. sites: Knik Arm, Alaska [5]; Tacoma Narrows, Washington [6]; and Golden
Gate, California [7]. In 2006, EPRI performed a tidal energy resource assessment for southeast
Alaska [8].
Frequency (hours per
1000
800
600
year)
400
200
0
0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 3.3 3.5 3.7 3.9
Depth-Averaged Velocity
(mid-point of speed bin, m/sec)
Figure 8-15
Annual Average Tidal Current Speed Probability Distributions for Dog Island Transect,
Western Passage, Maine
Figure 8-16 shows the available tidal energy resource for the five U.S. feasibility study sites
and one additional site, North Inian Pass in Southeast Alaska. North Inian Pass may be the
largest tidal energy site in North America.
8-22
10581090
Ocean Tidal Energy
Knik Arm
0.9 TWh/yr
Admiralty
No Inian Inlet
Pass 14 1.7 TWh/yr
TWh/yr
Tacoma
0.9 TWh/yr
Western
Passage
0.9 TWh/yr
Golden
Gate SF
2.1 TWh/yr
Figure 8-16
Annual Tidal Energy Resource for Six U.S. Tidal Current Sites
Following the site surveys, EPRI performed design, performance, cost, and economic feasibility
studies for five U.S. sites and two Canadian sites. The results of this work are documented in
reports available from EPRI’s Ocean Energy website (http://oceanenergy.epri.com) [5-7, 9-12].
8.4.3 Extractable Tidal Power and Energy
Estimating the portion of tidal energy at a given site that can be extracted without causing
unacceptable ecological impact is far more complex than calculating the available tidal energy at
a given site. Since tidal energy technology is modular, energy extraction is scalable and can be
limited to avoid significant ecological impact. Ian Bryden [13] suggest that 10% of the available
energy can be extracted without undue modification of the flow characteristics, while Black &
Veatch [14] suggested a limit of 20%.
Recent research by Dr. Ian Bryden [13, 15], now at the University of Edinburgh, Scotland,
indicates that the extractable resource may be significantly greater than earlier reported,
depending on the characteristics of the site. Extraction of kinetic power from tidal streams alters
the tidal regime in an estuary by reducing flow volumes, constricting the tidal range, and altering
the timing of tidal events. However, the magnitude of these impacts depends strongly on the
level of power extraction, estuary geometry, the nonlinear dynamics of in-stream turbines, and
the natural tidal regime. An understanding of the various fluidic effects of large-scale kinetic
power extraction is an essential first step in a more detailed investigation of ecosystem impacts.
The modular nature of tidal energy technology also has an effect on natural processes. It would
be counterproductive for the user to over-develop any particular tidal current resource because,
at a certain point, adding more turbines ceases to generate additional kinetic energy. The tidal
stream is in effect a self-regulating energy supply and any significant reduction in water transfer
reduces the energy capture.
8-23
10581090
Ocean Tidal Energy
EPRI sponsored a Ph.D. dissertation at the University of Washington on the Effect of Kinetic
Power Extraction [16] that resulted in publication of a peer-reviewed paper [17]. The
dissertation, published in early 2009, noted:
“The hydrodynamic effects of extracting kinetic power from tidal streams present unique
challenges to the development of in-stream tidal power. In-stream tidal turbines
superficially resemble wind turbines and extract kinetic power from the ebb and flood of
strong tidal currents. Extraction increases the resistance to flow, leading to changes in
tidal range, transport, mixing, and the kinetic resource itself. These far-field changes have
environmental, social, and economic implications that must be understood to develop the
in-stream resource.
“This dissertation describes the development of a one-dimensional numerical channel

model and its application to the study of these effects. The model is applied to determine
the roles played by site geometry, network topology, tidal regime, and device dynamics.
A comparison is also made between theoretical and modeled predictions for the
maximum amount of power which could be extracted from a tidal energy site. The model
is extended to a simulation of kinetic power extraction from Puget Sound, Washington.
“In general, extracting tidal energy will have a number of far-field effects in proportion to
the level of power extraction. At the theoretical limit, these effects can be very significant
(e.g., 50% reduction in transport), but are predicted to be immeasurably small for pilot-
scale projects. Depending on the specifics of the site, far-field effects may either augment
or reduce the existing tidal regime. Changes to the tide, in particular, have significant
spatial variability. Since tidal streams are generally subcritical, effects are felt throughout
the estuary, not just at the site of extraction.”
8.4.4 Tidal Power Forecasting
Since tides can be computed from the relative positions of the Sun, Earth, and Moon, the
available tidal power as a function of time can be computed years in advance. Weather has only
an insignificant effect on tidal currents. The timing of the peak tidal power advances by 50
minutes each day.
8.5 Tidal In-Stream Energy Conversion (TISEC) Development
Tidal power research programs in industry, government, and at universities in the UK, Ireland,
France, Italy, Korea, Canada, Australia, New Zealand, and the United States over the last six
years have established an important foundation for the emerging tidal power industry.
8-24
10581090
Ocean Tidal Energy
8.5.1 Harnessing Tidal Energy
Tidal energy extraction is complex and many device designs have been proposed. It is helpful
to introduce these designs in terms of their physical arrangements and energy conversion
mechanisms. A hydrokinetic energy conversion device can be classified as one of three types:
• Axial flow: the axis of rotation is parallel to the direction of water flow.
• Cross flow: the axis of rotation is perpendicular to the water stream and may be oriented at
any angle, from horizontal to vertical, with respect to the water surface.
• Non-turbine: oscillatory hydrofoil, vortex induced motion, or hydro Venturi device.
Figure 8-17 illustrates axial and cross flow-types of turbines.
Axial and Cross Flow Turbines Marine Current Turbines Axial Turbine
(Courtesy Marine Current Turbines)
Lucid Energy Cross Flow Vertical Turbine ORPC Cross Flow Horizontal Turbine
(Courtesy Lucid Energy) (Courtesy ORPC)
Figure 8-17
Tidal Energy Conversion System Configurations
8-25
10581090
Ocean Tidal Energy
The subsystems for a hydrokinetic turbine typically include a blade or rotor, which converts the
energy in the water to rotational shaft energy; a drive train, which usually comprises a gearbox
and a generator; a support structure for the rotor and drive train; and other equipment, including
controls, cables, and interconnection equipment. Additional ways of classifying these devices
include:
• Support structures: These may be either gravity foundation based, attached to a monopile
foundation, or anchored or moored and allowed to “fly” in the stream using buoyancy and/or
dynamic pressure forces. TISEC devices may also be hung from a floating platform that is
anchored and moored to maintain its location.
• Open vs. Ducted: Turbine rotors may be either open, much like wind turbines, or enclosed in
a duct or shroud. Since the energy is a function of the cross-sectional area of the flow
directed through the swept area of the turbine, using a duct of a given cross sectional area is
the equivalent of using an open rotor with the same cross-sectional area. The wind energy
industry has found that adding length to a rotor blade is more economical than adding a duct
to increase the cross-sectional area and power of a wind machine. However, the economics
may differ for wind turbines.
• Fixed vs. variable pitch blades: Pitch control is used to limit power, maximize the efficiency
of the turbine in variable flows, and enable bi-directional operation. There are other ways to
accomplish these three functions with fixed-blade turbines and many different design
concepts for implementing them.
• Closed-center vs. open-center hubs: Instead of a fixed hub and rotating blades, a design
variation uses an outer fixed rim and an inner rotating bladed disc. The potential benefit of an
open-center design is eliminating the need for a gearbox by encapsulating the stator of a
generator on the rim of the machine.
• Savonius vs. Darrieus vertical-axis turbines: First invented in Finland, the Savonius turbine is
S-shaped when viewed from above. This drag-type vertical-axis turbine turns relatively
slowly, but yields a high torque. The Darrieus turbine was invented in France in the 1920s.
Often described as looking like an eggbeater, this type of turbine has vertical blades that
rotate into and out of the flow. Using hydrodynamic lift, these turbines can capture more
energy than drag devices.
• Helical vs. cycloidal: Aerodynamic lift-type vertical axis turbine blade configuration. A
helical blade traces a three-dimensional curve that lies on a cylinder or cone, such that its
angle to a plane perpendicular to the axis is constant. A cycloidal blade has the shape of a
curve traced by a point on the circumference of a circle that rolls on a straight line.
There are other types being investigated, such as hydro Venturi and hydrofoil or vortex induced
oscillation. However, they are not of sufficient practical importance at this time to be described
in this chapter. Information on these devices is contained in an EPRI report [18] on TISEC
devices available at EPRI’s Ocean Energy website (http://oceanenergy.epri.com).
8-26
10581090
Ocean Tidal Energy
8.5.2 TISEC System Developers
Today, a number of companies are leading the commercialization of technologies to generate

electricity from tidal currents. In 2005, EPRI requested information from all known TISEC
device developers [18]. Table 8-1 lists known developers as of November, 2010; it excludes
those with only concept designs and patents.
Table 8-1
Offshore In-Stream Tidal Energy Conversion Device Developers
Device Developer(1) Development

Device Name Type
Website Status(2)
AeroHydro Research and technology Unknown Oscillatory Experimental
www.arhta.com Wings
AquaMarine Power Neptune Dual-rotor Technology
www.aquamarinepower.com Horizontal Axis Demonstration
Atlantis Resources Nereus and Solon Ducted Technology
www.atlantisresourcescorporation.com Horizontal Axis Demonstration
BioPower BioStream Oscillatory Experimental
www.biopowersystems.com Biomemetric
Blue Energy Canada Inc Ocean Turbine Vertical Axis Experimental
www.bluenergy.com
Clean Current Tidal Turbine Ducted Early Commercial
www.cleancurrent.com Generator Horizontal Axis
Engineering Business Stingray Oscillatory Technology
www.engb.com Wing Demonstration
Free Flow Power Unknown Ducted Early Commercial
www.freefllowpower.com Horizontal Axis
Free Flow 69 Osprey Vertical Axis Experimental
Hammerfest STROM AS Unknown Horizontal Axis Technology
www.e-tidevannsenergi.com Demonstration
Hydro Green Energy Krouse Turbine Horizontal Axis Experimental
www.hgenergy.com
Lucid Energy Gorlov Helical Vertical Axis Technology
www.lucidenergy.com Turbine (GHT) Demonstration
Lunar Energy Ltd Rotech Tidal; Ducted Horiz. Technology
www.lunarenergy.co.uk Turbine Axis Demonstration
Marine Current Turbines Ltd, Seaflow and Dual-rotor Commercial
www.marineturbines.com SeaGen Horizontal Axis Demonstration
Neptune Renewables Proteus Cross Flow Technology
http://www.neptunerenewableenergy. Demonstrations
com/tidal_technology.php
New Energy Corporation EnCurrent Turbine Vertical Axis Technology
www.newenergycorp.ca Demonstration
Ocean Flow Energy Evapod Horizontal Axis Technology
http://www.oceanflowenergy.com/ Demonstration
8-27
10581090
Ocean Tidal Energy

Offshore In-Stream Tidal Energy Conversion Device Developers
Device Developer(1) Development

Device Name Type
Website Status(2)
Oceana Energy TIDES Unknown Experimental
www.oceanaenergy.com/
Open Hydro Open Centre Ducted Horiz. Commercial
www.openhydro.com Turbine Axis Demonstration
Ocean Renewable Power Corp OCGen Cross-Flow Technology
www.oceanrenewablepower.com Axis Demonstration
Ponte de Archimede Kobold Vertical Axis Commercial
www.pontediarchimede.it Demonstration
Pulse Stream Pulse Stream 100 Horizontal Axis Technology
Demonstration
Seapower Int’l AB EXIM Vertical Axis Technology
www.seapower.se Demonstration
SMD Hydrovision TidEl Horizontal Axis Technology
www.smdhydrovision.com Demonstration
Swan Turbines Swan Turbine Horizontal Axis Technology
www.swanturbines.co.uk Demonstration
Tidal Energy Ltd DeltaStream Open Rotor Unknown
Tidal Generation Unknown Horizontal Axis Technology
www.tidalgeneration.co.uk Demonstration
Tidal Hydraulics Tidal Hydraulic Horizontal Axis Technology
www.dev.onlinemarketinguk.net?THG Generator Demonstration
?index.html
Tidal Sails Tidal Sail Cross-flow Experimental
www.tidalsails.com Axis Sail
Tidal Stream Tidal Stream Horizontal Axis Technology
www.tidalstream.co.uk device Demonstration
UEK Underwater Ducted Horiz. Commercial
www.uekus.com Electric Kite Axis Demonstration
Verdant Power Free Flow Turbine Horizontal Axis Commercial
www.verdantpower.com Demonstration
Voith Siemans Unknown Horizontal Axis Technology
www.voith.com/press/546530.htm Demonstration
Vortex Hydro Energy VIVACE Point Absorber Laboratory
www.vortexhydroenergy.com
(1) This list excludes individual inventors with conceptual level technology
(2) The following definition of development status was used
• Laboratory testing stage
• Experimental – Subscale at sea testing
• Technology Demonstration – Large size engineering prototype at sea testing whose purpose is to test for function
and performance
• Commercial Demonstration – Large size manufacturing prototype at sea testing whose purpose is to test for
commercial viability
• Early Commercial – Offering many units of large size for purposes of generating and selling the electricity produced
8-28
10581090
Ocean Tidal Energy
8.5.3 Survival in Storms and Hostile Marine Environments
Tidal energy conversion devices can be deployed entirely below the water’s surface and thereby
avoid exposure to the full brunt of atmospheric storms. Other types of devices have
superstructures that pierce the surface; while such structures are exposed to atmospheric storms,
servicing of above-water equipment may be facilitated. Relative to long-term survival in the
environment, oil and gas platforms using anti-corrosion and biofouling technology are surviving
50 years.
8.5.4 Effect of Tidal Power Plants on the Environment
Given proper care in siting and scaling of projects, tidal in-stream power can be one of the more
environmentally benign electrical generation technologies. Early demonstration and commercial
tidal in-stream power plants should include rigorous monitoring to record both environmental
impacts as well as natural variation at nearby undeveloped sites.
To date, a single in situ monitoring study has been conducted by the Verdant Power RITE
project. Anecdotal information from numerous temporary testing activities in the United States,
Canada, UK, and other countries has not observed any harm to aquatic life. TISEC device blades
rotate very slowly (around 10 rpm for an 18-meter diameter rotor) and the speed at the tip of the
blade is about 10-12 m/s (22-27 mph). The devices are self limiting in that any faster speeds
result in cavitation.
Two TISEC test programs with thorough environmental assessments of sea life and
environmental effects were initiated in 2008: the Verdant RITE project in New York’s East
River and the MCT SeaGen project in Strangford, UK. Results from these projects have not yet
been published. More recently, environmental studies have been undertaken to assess fish
interactions with Ocean Renewable Power Corporation’s beta TideGen turbine in Cobscook Bay.
These results, also, have yet to be published.
EPRI is currently testing interaction between hydrokinetic turbines and fish in experimental
flumes at Alden Research Laboratories in Holden, Massachusetts, and at the USGS Conte
Anadromous Fish Research Laboratory in Turners Falls, Massachusetts. The results of these
studies, sponsored by industry and the DOE, will be available in early- to mid-2011.
8.5.5 Permits for Tidal Power Plants
The novelty of TISEC technology has to date triggered conservative evaluations and an
extensive approval process at the federal, state, and local levels. As of October 2010, there were
no FERC licenses issued for operation of tidal energy projects. Table 8-2 lists the 16 active,
issued Preliminary Permits and the 13 pending Preliminary Permits for tidal hydrokinetic
projects. Many of these Preliminary Permits are have been reissued either because the prior
permit expired or to accommodate substantive changes in the project description. Preliminary
Permits are valid for 36 months from the month of issuance. Additional information can be
obtained at the FERC website: www.ferc.gov/industries/hydropower/indus-
act/hydrokinetics/permits.asp).
8-29
10581090
Ocean Tidal Energy
Table 8-2
Current Issued and Pending FERC Tidal In-Stream Preliminary Permits (as of October 4, 2010)
Capacity Issued/
Docket No Project/Permit Name (MW) State Filing Date Issue Date Pending
P-12585 San Francisco Bay Tidal Energy Project 10 CA . 02/04/2010 Issued
P-12611 Roosevelt Island Tidal Energy 5 NY . 02/17/2009 Issued
P-12679 Cook Inlet Tidal Energy 5 AK 04/01/2010 . Pending
P-12680 Cobscook Bay OCGen Power Project 70 ME 07/01/2010 . Pending
P-12687 Deception Pass Tidal Energy 6.4 WA . 08/04/2010 Issued
P-12690 Admiralty Inlet Tidal Energy 1 WA . 07/08/2010 Issued
P-12704 Half Moon Cove 9 ME 04/12/2010 . Pending
P-12711 Western Passage OCGen Power Project 70 ME 07/01/2010 . Pending
P-12718 Wards Island Tidal Power 0.096 NY . 04/17/2009 Issued
P-13015 Edgartown-Nantucket Tidal Energy 10 MA . 03/31/2008 Issued
P-13245 Indian River Tidal Energy 10 DE . 01/07/2009 Issued
P-13247 Kingsbridge Marina Tidal Energy 0.04 NJ . 12/12/2008 Issued
P-13298 Port Clarence 0.3 AK . 05/08/2009 Issued
P-13329 Town of Wiscasset Tidal Resources 10 ME . 05/28/2009 Issued
P-13345 Homeowner Tidal Power Elec Gen 0.06 ME . 07/01/2009 Issued
P-13503 General Sullivan and Little Bay Bridge 0 NH . 09/30/2009 Issued
P-13509 Turnagain Arm Tidal 2200 AK . 02/05/2010 Issued
P-13605 Icy Passage Tidal 0.3 AK . 04/30/2010 Issued
P-13606 Gastineau Channel Tidal 0.4 AK . 04/30/2010 Issued
P-13646 Damariscotta Tidal 0.25 ME . 05/19/2010 Issued
P-13682 Hoffmans Marina NJ Tidal Energy Project 0.2 NJ 03/10/2010 . Pending
P-13725 Highlands NJ Tidal Energy Project 3 NJ 05/06/2010 . Pending
P-13730 Astoria Tidal Energy Project 300 NY 05/12/2010 . Pending
P-13731 East River Tidal Energy Pilot Project 0.02 NY 05/12/2010 . Pending
P-13801 Kendall Head Tidal Energy Project 50 ME 06/29/2010 . Pending
P-13821 East Foreland Tidal Energy 5 AK 08/02/2010 . Pending
P-13823 Killisnoo Tidal Energy 0.25 AK 08/05/2010 . Pending
P-13828 Cape Cod Tidal Energy 20 MA 08/09/2010 . Pending
P-13830 Half Moon Cove Tidal 20 ME 08/09/2010 . Pending
8.5.6 Overview of Regulatory Status for Tidal Power Plants
The regulatory permitting and licensing processes associated with U.S. ocean wave and tidal
projects can be quite complex, lengthy, and costly. Indeed, regulatory issues may be the primary
barrier to ocean energy development in the United States.
Since 1920, construction and operation of a non-federal hydroelectric project in the United States
has required a license issued by FERC in accordance with the Federal Power Act. In 2004, as a
result of legal interpretation from FERC, unconventional tidal and wave energy hydroelectric
projects were placed under FERC jurisdiction. An EPRI report [19], available at EPRI’s Ocean
Energy website (http://oceanenergy.epri.com/streamenergy.html) provides a detailed description
of the U.S. regulatory environment.
8.5.6.1 FERC Pilot Project License

In 2008, FERC rolled out a licensing process for hydrokinetic pilot projects tailored to meet the
needs of entities interested in testing new technology, including connection with the interstate
grid, while minimizing the risk of adverse environmental impacts. The goal of the Pilot Project
licensing process is to allow developers to test new hydrokinetic technologies, determine
appropriate siting of these technologies, and confirm their environmental effects, while
maintaining FERC oversight and agency input and reducing the licensing requirements
compared to what they would be for a Commercial License of 30-50 years duration. Pilot Project
8-30
10581090
Ocean Tidal Energy
Licenses are restricted to projects of 5 MW or less. Licensed Pilot Projects must be removed at
the end of the license term (nominally 5 years) unless a Commercial License is obtained. Pilot
Projects must be thoroughly monitored and may be terminated prior to expiration of the license
term if unacceptable effects of the project are observed. A Pilot Project License is not an option
for proposed projects located at environmentally sensitive sites.
8.6 Design, Performance, Cost, and Economic Feasibility
8.6.1 TISEC Sites
EPRI has investigated the attributes required for a good tidal energy site. Reports that document
the EPRI site assessments for Maine [2], Massachusetts [3], Nova Scotia [4], and New
Brunswick [20] are available at EPRI’s Ocean Energy website
(http://oceanenergy.epri.com/streamenergy.html).
There are many factors to consider when evaluating potential sites for a tidal energy project.
Primary factors include:
• High annual current flow measured by both flow volume and velocity;
• Proximity to a harbor/marina with sufficient depth, size, and infrastructure to support
assembly and deployment of the plant as well as maintenance operations;
• Proximity to a transmission and distribution system that can transfer power from the tidal
plant to a substation interconnection point;
• Water depth that allows the device to be mounted on the seabed with sufficient overhead
clearance for navigation, if required;
• Minimal potential for conflicting use of sea space;
• Receptivity of the local community;
• Availability of a local labor force to be trained for employment in this new industry.
8.6.2 TISEC Devices Studied
EPRI performed a point design for both a single-unit demonstration and a commercial wave
power plant for sites in five U.S. states (Alaska [5], Washington [6], California [7],
Massachusetts [9], and Maine [10]) and two Canadian Provinces (New Brunswick [11] and Nova
Scotia [12]). These designs were used to estimate cost and performance. Performance estimates
were developed using tidal current data predictions obtained from NOAA. Cost estimates were
developed by creating a detailed breakdown of the various cost centers and outlines of
installation and operation procedures, and by cross checking them with a variety of sources,
including local operators, the design team, local manufacturers, and similar offshore projects in
the oil and gas and offshore wind industries.
EPRI independently estimated plant system costs based on the Marine Current Turbine SeaGen
dual 18-m diameter rotor device design for the five tidal sites in the United States and the two
tidal sites in Canada. Table 8-3 describes the cost estimates for two of the seven sites, Tacoma
Narrows in Washington State and the Western Passage between Maine and New Brunswick.
8-31
10581090
Ocean Tidal Energy
Table 8-3
Cost and Performance Estimates for Selected Tidal Power Plants
0.83 MW 10 MW 0.72 MW 45.8 MW

Rated Capacity Western Western Tacoma Tacoma
Passage ME Passage ME Narrows WA Narrows WA
Plant Capacity (number of units x size, MW) 1 x 0.83 12 x 0.83 1 x 0.716 64 x 0.716
Annual Energy (MWe/year) 3,480 41,732 1,875 120,000
Plant Nameplate Rating (MW) 0.83 10 0.75 45.8
Technology Description – Axial Open Rotor SeaGen SeaGen) SeaGen SeaGen
MCT
Physical Plant
2
Seabed, m 500 125,000 500 750,000
Scheduling
1
Development time, Months Note 1 Note 1 Note 1 Note 1
Construction time, Months 12 24 12 24
Capital Cost ($/kW)
2
Month/Year Dollars Dec 2009 (2) Dec 2006 Dec 2006 Dec 2006
On shore transmission and grid I/C 266 592 524 84
Subsea cables 144 23 28 20
Power Conversion Modules 1,576 987 1,576 729
Structural sections 571 558 1,018 932
Subsea cable installation 1,806 345 1,042 229
Turbine installation 1,921 655 2,223 496
3
Construction Mgmt and Commissioning Note 3 Note 3 Note 3 Note 3
3
Contingencies Note 3 Note 3 Note 3 Note 3
Total Plant Cost (TPC) 6,284 2,679 6,411 2,493
AFUDC (interest during construction) 535 195 544 186
Total Plant Investment 6,819 2,874 6,955 2,679
4 4 4 4 4
Due Diligence, Eng & Permitting, Etc. 341 144 348 134
Total Capital Requirement 7,160 3,018 7,303 2,813
5 5
Yearly O&M Costs (% of TPC per yr) N/A 4.2 N/A 3.7
Unit Availability (%) 95 98 95 98
Technology Development Rating Pre Com’l Pre Com’l Pre Com’l Pre Com’l
1. Development time for permitting is an unknown at this early point with emerging ocean energy technology.
2. The costs are in December 2005 dollars in References [10 through 16] and were adjusted by 2.5%% inflation to Dec 2009.
3. Construction management, commissioning, and contingency costs are built into each of the subsystems.
4. Assumed to be 5% of TPI; Cost of permitting is an unknown at this early point with emerging ocean energy technology.
5. O&M costs for a pilot plant cannot be estimated.
8-32
10581090
Ocean Tidal Energy
These cost estimates are in December 2006 constant dollars, are based on an assumed set of
financial incentives, and do not include permitting and engineering costs. In order to avoid
significant environmental impacts, the energy extraction was conservatively limited to 15% of
the available energy resource.
8.6.3 TISEC Economic Feasibility
Table 8-4 describes the economic results for the five U.S. feasibility evaluation sites. These
cost and cost-of-electricity (COE) estimates were escalated from 2005 to December 2009,
assuming an escalation rate of 2.5%/year. The levelized cost of electricity is in 2009 constant
dollars and is based on an assumed set of financial incentives. The capital cost estimate includes
a 5% allowance for due diligence and permitting costs and excludes engineering costs. In two
cases, California and Maine, the 15% extraction limit could not be reached due to the relatively
small high-current area limiting the number of existing technology turbines which could be
deployed.
Table 8-4
Cost Estimates for Selected U.S. Feasibility Evaluation Sites (Dec-2009 $)
WA MA ME
AK Knik CA Golden
Tacoma Muskeget Western
Arm Gate
Narrows Channel Passage
Rated Power (MW) 50 46 44 4 10
Annual Average Power 14.6 13 16.5 1.6 4.6

(MW)
Number of Turbines 66 64 40 9 12
Total Plant Cost ($M) 122 114 100 19 27
Yearly Level O&M

Costs ($M/yr) 4.5 4.2 4.0 0.6 1.1
Annual Energy (GWh) 128 120 129 1.5 41.7
Utility Gen COE

(cents/kWh) 10.2 10.0 7.3 9.5 6.2
Muni Gen COE
(cents/kWh) 7.8 7.9 5.4 6.6 4.6
TISEC technology has similarities with wind technology and has benefited from learning
experiences and cost reductions that have occurred in that industry. Additional reductions from
the TISEC cost shown in the table above will be realized through value engineering and
economies of scale.
8-33
10581090
Ocean Tidal Energy
8.7 Environmental Impact Issues
Given proper care in scale, siting, deployment and operation, EPRI expects that tidal in-stream
power will be one of the most environmentally benign electricity generation technologies. Early
demonstration and commercial offshore tidal power plant projects should include rigorous
monitoring of the environmental effects of the plant and similar rigorous monitoring of a nearby
undeveloped site in its natural state (so that natural variation can be distinguished from TISEC-
induced effects).
Between mid-2009 and October 2010, two major reports on the environmental effects of tidal in-
stream energy were published [21, 22]. One of the reports, “Potential Environmental Effects of
Marine and Hydrokinetic Energy Technologies,” was prepared in response to the Energy
Independence and Security Act (EISA) of 2007. Section 633(b) of the EISA called for a report to
be provided to Congress that addresses (1) the potential environmental impacts of marine and
hydrokinetic energy technologies; (2) options to prevent adverse environmental impacts; (3) the
role of monitoring and adaptive management; and (4) the necessary components of an adaptive
management program.
The EISA-mandated Report to Congress was prepared based on a review of peer-reviewed

literature, project documents, and U.S. and international environmental assessments of new
technologies. The information was supplemented by contacts with technology developers,
experts in state resource and regulatory agencies and non-governmental organizations, and input
and reviews by federal agencies (NOAA Fisheries, Minerals Management Service, U.S. Fish and
Wildlife Service, National Park Service, Bureau of Indian Affairs, Federal Energy Regulatory
Commission). Excerpted findings from the report follow.
"There are numerous conceptual designs for converting the energy of waves, river and
tidal currents, and ocean temperature differences into electricity. Most of these
technologies remain at the conceptual stage—they have not yet been tested in the field or
as prototype, full-scale devices. Consequently, there have been few studies of their
environmental effects. Most considerations of the environmental impacts have been in the
form of predictive studies and environmental assessments that have not yet been verified.
"The assessments have identified common elements among these technologies that may
pose a risk of adverse environmental effects. These potential impacts include the
alteration of currents and waves; alteration of substrates and sediment transport and
deposition; alteration of habitats for benthic organisms; noise during construction and
operation; emission of electromagnetic fields; toxicity of paints, lubricants, and
antifouling coatings; and interference with animal movements and migrations. Project
installation and operation will change the physical environment. Effects on biological
resources could include alteration of the behavior of animals, damage and mortality to
individual plants and animals, and potentially larger, longer-term changes to plant and
animal populations and communities. Some effects are expected to be minor, but the
potential significance of many of the environmental issues cannot yet be determined
owing to a lack of experience with operating projects.
"Although there have been few environmental studies of these new concepts, a
preliminary indication of the importance of each of these issues can be gained from
published literature related to other technologies, e.g., noises generated by similar marine
8-34
10581090
Ocean Tidal Energy
construction activities, EMF emissions from existing submarine cables, and

environmental monitoring of active offshore wind farms. Experience with other, similar
activities in freshwater and marine systems will also provide clues to effective impact
minimization and mitigation measures that can be applied to these new renewable energy
technologies. However, some aspects of the environmental impacts are unique to the
technologies, and will require operational monitoring to determine the seriousness of the
effects. This is particularly true for the cumulative effects of large numbers of ocean
energy or hydrokinetic devices that will comprise fully built-out projects. Impacts to
bottom habitats, hydrographic conditions, or animal movements that are inconsequential
for a few units may become serious if large, multi-unit projects exploit large areas in a
river, estuary, or nearshore ocean. For some environmental issues it will be difficult to
extrapolate predicted effects from small to large numbers of units because of
complicated, non-linear interactions between the placement of the machines and the
distribution and movements of aquatic organisms. Assessment of these cumulative effects
will require careful environmental monitoring as the projects are deployed.
"Evaluation of monitoring results might be usefully conducted in an adaptive

management framework. There are numerous state and federal agencies and
environmental laws and regulations that will influence the development of marine and
hydrokinetic technologies. Federal licensing of these renewable energy projects is the
responsibility of the Federal Energy Regulatory Commission and the Minerals
Management Service. Their licensing decisions will include input from other federal and
state agencies, tribes, environmental groups, and other stakeholders. After a licensing
decision has been made and operation of the energy project has begun, the identification
(and correction) of environmental impacts will depend on appropriate monitoring.
"The ability to modify the project in order to mitigate unacceptable environmental

impacts identified by operational monitoring might be based on application of adaptive
management principles reflected in the project license conditions. In the context of
marine and hydrokinetic energy technologies, adaptive management is a systematic
process by which the potential environmental impacts of installation and operation could
be evaluated against quantified environmental performance goals during project
monitoring. Early information about undesirable outcomes could lead to the
implementation of additional minimization or mitigation actions which are subsequently
re-evaluated. An adaptive management process is particularly valuable in the early stages
of technology development, when many of the potential environmental effects are
unknown for individual units, let alone the eventual build out of large numbers of units.
Basing the environmental monitoring programs on adaptive management principles, as
advocated by many resource and regulatory agencies, will take advantage of ongoing
research and monitoring to help refine technology designs and to improve environmental
acceptability of future installations."
The second major report [21], sponsored by DOE, sought to develop a framework for identifying
key environmental concerns associated with marine renewable energy projects, including TISEC
projects. Both of these reports constitute scoping documents that identify the range of issues
potentially faced by the industry as a whole and by developers of individual projects. Much work
remains to be done to narrow the scope of the data-gathering requirements placed upon project
developers [23]
8-35
10581090
Ocean Tidal Energy
8.8 Installed Capacity and Estimated Growth
To date, worldwide installed TISEC capacity is about 3 MW. The devices are prototypes and no
commercial tidal power plants have yet been constructed; however, several commercial projects
are in the planning and permitting stage. As of November, 2010, the installed TISEC capacity in
the U.S. was 60 kW (the testing of two 34-kW Verdant turbines in the East River in New York is
complete and the turbines have been removed). Table 8-5 presents EPRI's estimate of tidal in-
stream capacity (in MW) that will come on line (i.e., be commissioned and selling electricity into
the grid to end users) in the United States from 2011 through 2015.
Table 8-5
Installed and Planned U.S. Tidal Energy Capacity
Developer Project Name-Site 2011 2012 2013 2014 2015

(Proposed Capacity, MW)
Verdant1 East River, NY 20 40

ORPC Western Passage ME 5 20 40
Oceana2 Golden Gate, CA 5 20
SNOPUD (25) Admiralty Strait, WA 5 20
Chevron (80) Cook Inlet, AK 5 20

Massachusetts Tidal3 (300) Vineyard Sound MA
UEK3 (40) Piscatagua River ME
Edgartown3 (10) Nantucket Sound, MA
Maine Tidal Energy3 (100) Kennebec River, ME
Astoria Tidal Energy3 (300) East River, NY

TOTAL NEW YEARLY CAPACITY 5 35 20 100 40
CUMULATIVE 6 41 61 161 201

1. Verdant has removed the experimental turbines and filed a Pilot License application with FERC for 5 MW.
Assumes extractable rated capacity of the East River is greater than 40 MW; however, EPRI has not studied
this site.
2. FERC declined Oceana’s application for a pilot license. Nevertheless, we are making an optimistic
assumption that someone will develop the Golden Gate site since it may be one of the few really
economically attractive sites in the contiguous United States and since PG&E and the City of San Francisco
had previously joined with Oceana (the preliminary permit holder) to develop this site.
3. EPRI is not confident that these project developers have the skills required to develop these project and has
therefore not included them in our capacity estimate.
Those projections assume that FERC and other regulatory agencies change their practices to
enable tidal in-stream plants to be permitted at about the same schedule and cost as on-land wind
plants, and that Congress provides the same incentives to tidal in-stream energy as for wind
energy.
8-36
10581090
Ocean Tidal Energy
In general, EPRI expects that tidal energy will experience a growth rate that is limited by
the resource availability in the lower 48 states, although any predictions depend largely on
overcoming regulatory barriers and securing government support and incentives for the emerging
tidal energy industry. EPRI estimates that there could be 500 MW of tidal energy plant capacity by
2025. If Alaska is able to harness its tidal energy resource in this time period, this estimate would
be much larger (Alaska is believed to have more than 90% of the total U.S. tidal energy resource).
If small (tens of kilowatts) hydrokinetic turbines rated at lower speeds (less than 2 m/s) become
economical, the number and distribution of viable sites could increase.
8.9 R&D Needs
With support from the DOE Office of Energy Efficiency and Renewable Energy’s Wind and
Hydropower Technologies Program, EPRI sponsored a workshop for the water power industry in
October 2008 [24]. The purpose of the workshop was to identify and prioritize research,
development, deployment, and demonstration (RDD&D) needs which will further the
deployment of conventional hydro/pumped storage and emerging wave and marine hydrokinetic
(MHK) technologies to increase domestic, low-carbon energy production. The priorities
identified are intended to shape both EPRI’s and the DOE’s research agendas, and will support
R&D initiatives throughout both the public and private sectors.
The US RDD&D Needs Workshop used the 12 topics from the U.K. Marine Energy technology
Roadmap [25] as the starting point for developing the U.S. technology needs:
• Resource Modeling
• Device Modeling
• Experimental Testing
• Moorings and Sea Bed Attachments
• Electrical Infrastructure
• Power Take Off and Control
• Engineering Design
• Lifecycle and Manufacturing
• Installation, O&M
• Environmental
• Standards
• System Simulation
Four other topics identified by the Steering Committee prior to the workshop and two other
topics identified by the participants of the workshop were:
• Materials—low cost, corrosion and biofouling
• Storage
8-37
10581090
Ocean Tidal Energy
• System Configuration Evaluations

• Vision, Goals, Objectives and Roadmap
• Master Generation and Transmission Plan
• Education
The three highest prioritized topical areas were:

• Testing (development including experimental through pilot demonstration)
• Environmental (which will require device testing and deployed projects)
• Standards.
8.10 Conclusion
Considerable potential exists for generating electrical power from tidal energy in the United
States and many other places in the world. However, most of the U.S. tidal energy potential
exists in the state of Alaska, where the demand is low and where there is no transmission
infrastructure to bring the energy into the lower 48 states. There are only a few very good tidal
sites in the lower 48 states. Those known to EPRI are:
• Golden Gate, California: Excellent tidal resource (both volume and power density), adjacent
to a large load center with grid infrastructure, a nearby port, and high electricity prices.
• Puget Sound, Washington: Excellent tidal energy resource (both volume and power density)
at a few sites, adjacent to a large load center with grid infrastructure, nearby port, but low
electricity prices.
• Western Passage, Maine: Excellent tidal energy resource (both volume and power density),
adjacent to transmission infrastructure for delivering the power to large load centers, nearby
port, and high electricity prices.
Verdant Power believes that the East River in New York is a good tidal site, as do Oceana and
Natural Currents, who also hold preliminary permits for sites on the East River. It remains to be
seen how FERC will treat multiple licenses on a single river, given that the projects may have
impacts on each other.
8.11 Internet Resources
EPRI tidal power reports are available at: www.epri.com/oceanenergy.
Department of Trade and Industry (UK): www.dti.gov.uk/renewable.
8-38
10581090
Ocean Tidal Energy
8.12 References
1. Methodology for Estimating Tidal Current Energy Resources and Power Production by Tidal
In-Stream Energy Conversion (TISEC) Devices, EPRI, Palo Alto, CA: September 2006.
EPRI-TP-001-NA (Revision 3).
2. Maine Tidal In-Stream Energy Conversion (TISEC): Survey and Characterization of
Potential Project Sites, EPRI, Palo Alto, CA: October 2006. EPRI-TP-003 ME Rev 1.
3. Massachusetts Tidal In-Stream Energy Conversion (TISEC): Survey and Characterization of
Potential Project Sites, EPRI, Palo Alto, CA: October 2006. EPRI-TP-003-MA (Revision 1).
4. Nova Scotia Tidal In-Stream Energy Conversion (TISEC): Survey and Characterization of
Potential Project Sites, EPRI, Palo Alto, CA: October 2010. EPRI-TP-003-NS (Revision 2).
5. System Level Design, Performance, Cost and Economic Assessment—Knik Arm Alaska Tidal
In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-006-AK.
6. System Level Design, Performance, Cost and Economic Assessment -- Tacoma Narrows
Washington Tidal In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-006-
WA.
7. System Level Design, Performance, Cost and Economic Assessment—San Francisco Tidal
In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-006-SF-CA.
8. Tidal In-Stream Energy Resource Assessment for Southeast Alaska, EPRI, Palo Alto, CA:
2006.
9. System Level Design, Performance, Cost and Economic Assessment -- Massachusetts
Juskeget Channel Tidal In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-
006-MA.
10. System Level Design, Performance, Cost and Economic Assessment—Maine Western
Passage Tidal In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-006-ME.
11. System Level Design, Performance, Cost and Economic Assessment—New Brunswick Head
Harbour Passage Tidal In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-
006-NB.
12. System Level Design, Performance, Cost and Economic Assessment—Minas Passage Nova
Scotia Tidal In-Stream Power Plant, EPRI, Palo Alto, CA: June 2006. EPRI-TP-006-NS.
13. Bryden, I. and S.J. Couch. 2006. ME1—marine energy extraction: tidal resource analysis.
Renewable Energy 31(2): 133-139.
14. Black & Veatch. 2004. UK, Europe, and Global Tidal Energy Resource Assessment. The
Carbon Trust, London. Marine Energy Challenge Report No. 107799/D/2100/05/1.
15. Bryden, I.G. and S.J. Couch. 2007. "How much energy can be extracted from moving water
with a free surface: a question of importance in the field of tidal current energy?" Journal of
Renewable Energy 32(11): 1961-1966.
16. Polagye, B. 2009. Hydrodynamic Effects of Kinetic Power Extraction by In-Stream Tidal
Turbines. Doctoral dissertation. University of Washington, Seattle, Washington.
8-39
10581090
Ocean Tidal Energy
17. Polagye, B., P.C. Malte, M. Kawase, and D. Durran. 2008. "Effect of large-scale kinetic
power extraction on time-dependent estuaries." Proceedings of the Institution of Mechanical
Engineers, Part A: Journal of Power and Energy 222(5): 471-484.
18. Survey and Characterization. Tidal In Stream Energy Conversion (TISEC) Devices, EPRI,
Palo Alto, CA: November 2005. EPRI-TP-004-NA.
19. Instream Tidal Power in North America: Environmental and Permitting Issues. Prepared by
Devine Tarbell & Associated, Inc. EPRI, Palo Alto, CA: June 2006. EPRI-TP-007-NA.
20. New Brunswick Tidal In-Stream Energy Conversion (TISEC): Survey and Characterization
of Potential Project Sites, EPRI, Palo Alto, CA: October 2006. EPRI-TP-003-NB (Revision
1).
21. Kramer, S., M. Previsic, P. Nelson, and S. Woo. 2010. Deployment Effects of Marine
Renewable Energy Technologies: Framework for Identifying Key Environmental Concerns in
Marine Renewable Energy Projects. Prepared by RE Vision Consulting, LLC for DOE. RE
Vision DE-003.
22. Report to Congress on the Potential Environmental Effects of Marine and Hydrokinetic
Energy Technologies. U.S. Department of Energy, Energy Efficiency and Renewable
Energy, Wind and Hydropower Technologies Program, Washington, D.C.: December 2009.
23. Jacobson, P.T. in press. Challenges and Opportunities in Tidal and Wave Power. In: K.R.
Rao (ed.). Energy and Power Generation Handbook. American Society of Mechanical
Engineers.
24. Prioritized Research, Development, Deployment and Demonstration (RDD&D) Needs:
Marine and Other Hydrokinetic Renewable Energy, EPRI, Palo Alto, CA: December 2008.
25. UKERC (UK Energy Research Centre). 2008. UKERC Marine (Wave and Tidal Current)
Renewable Energy Technology Roadmap. Summary Report.
8-40
10581090
9
OCEAN WAVE ENERGY
Installed Offshore • Less than 1 MW worldwide, and less than 0.1 MW in the United States.
Wave Capacity
• Estimated annual incremental U.S. capacity additions:
(as of October, 2010)
2011: 1.50 MW
2012: 13.30 MW
• EPRI estimates a U.S. cumulative capacity of approximately 200 MW by
2015 and 10,000 MW by 2025
Wave Energy • WEC is an emerging technology. About a half dozen full-scale prototype
Conversion (WEC) WEC devices have been demonstrated at sea over the past five years, as
Technology well as another dozen sub-scale prototypes.
Readiness
• The first phase (three Pelamis units at 0.75 MW each) of the world’s first
commercial wave plant was deployed in Portugal; it first transmitted
electricity to the grid in mid-2008. The three machines are now dockside
due to financial difficulties of the project majority owner.
Economic Status • The first U.S. commercial plant project in Reedsport, Oregon, was made
possible through public support, private investment, and state incentives.
• The first U.S. wave power purchase agreement was rejected by the
California Public Utility Commission in late 2008.
• The first U.S. wave power plant license issued by FERC for a 1-MW
project at Makah Bay, Washington was surrendered by Finavera.
Environmental Impact • Proper care in siting, installation, operation, and decommissioning may
enable ocean wave energy technology to be one of the more
environmentally benign electricity generation technologies.
• Pilot demonstration testing is needed to understand the interactions
used to acquire and incorporate new information into decision-making
processes that will address project build out and cumulative effects.
Regulatory Status • Wave power plant projects are being permitted in Europe (Pelamis plant
in Portugal, UK Wave Hub, Santano in Spain, etc.) and Australia.
• The time, cost and complexity of the U.S. regulatory process can be
difficult for wave power project developers.
• The Federal Energy Regulatory Commission (FERC) has primary
jurisdiction for licensing ocean wave energy under the Federal Power Act
(FPA), both in state waters and on the outer continental shelf.
• The FERC has developed a six-month license application process for pilot
demonstration plants.
9-1
10581090
Ocean Wave Energy
• The 2005 Energy Bill gives the Mineral Management Service (MMS) the
jurisdiction to lease lands on the outer continental shelf (OCS)—that is, 3
to 200 nautical miles offshore (except for the Gulf of Mexico where it is 12
to 200 nautical miles).
• FERC will no longer issue preliminary permits for wave plants on the OCS
but will continue to do so for wave plants in state waters.
• MMS lease rules for alternative energy on the OCS were issued in April
2009. There currently are no leases for wave plants on the OCS nor are
there any applications for leases for wave plants on the OCS.
Government Support • European governments (particularly in the UK, Ireland, Portugal and
of Wave Energy Denmark) as well as those in Japan, New Zealand, and Australia, support
Technology the development and deployment of WEC technology and are now
providing subsidies to stimulate a commercial market.
• DOE initiated a Waterpower R&D Program in FY 2008 funded with a
Congressionally-mandated $10 million, which was followed by an
additional $40 million for FY 2009.
• The Murkowski/Inslee Marine Renewable Energy Promotion Act of 2009
would authorize as much as $250 million a year (up from the current
authorization limit of $50 million per year) to expand federal research of
marine energy; take over the cost verification of new wave, current, tidal
and thermal ocean energy devices; create an adaptive management fund
to help pay for the demonstration and deployment of such electric
projects; and provide key additional tax incentives. This bill was approved
by the Senate Energy Committee in June 2009.
• It appears that the stimulus monies distributed to the DOE for renewable
energy will not include an allocation for marine energy.
Trends to Watch • Getting economical power from ocean waves will be difficult and require
the very best engineering skills. In addition, communication outreach and
negotiation skills for resolving conflict of sea space issues will be
required.
• More demonstration projects and early commercialization projects,
including multi-megawatt “wave farms,” are expected to be deployed over
the next decade in Europe, South America, and Australia (and the United
States if regulatory obstacles are overcome).
9.1 Introduction
Ocean waves are generated by the influence of wind on the ocean surface (Figure 9-1). Ripples
on the surface create a steep slope against which the wind can push and cause waves to grow. As
the wind continues to blow, the ripples become chop, fully developed seas and, finally, swells. In
deep water, the energy in waves can travel for thousands of miles until their energy is dissipated
on distant shores.
Individual waves represent an integration of all winds encountered as they travel over the ocean
surface. Sea states (comprising wave height, period, phase and direction) can be accurately
predicted more than 48 hours in advance. This predictability, along with the relatively small
short-term variability of the available resource, will allow time for grid operators to dispatch
9-2
10581090
Ocean Wave Energy
other generation resources to balance demand with supply. This is a major advantage over wind
and solar generation. Other characteristics of wave energy that make it especially attractive for
electricity generation are its high power density (kW/meter of wave crest length) and the
potential for being relatively environmentally benign if properly sited, sized, deployed, operated
and maintained.
Ocean waves are composed of orbiting particles of water, as illustrated in Figure 9-2. At the sea
surface, the diameter of water particle orbits is equal to wave height. Orbital motion decays
exponentially with depth, and its amplitude is only 4% of its surface value at half a wavelength
below the surface. The wave orbital motions are not significantly affected by the bottom in water
deeper than half a wavelength. The vector field of particle motion is illustrated in Figure 9-3.
Figure 9-1
Wind Blowing over Fetch of Water Produces Waves
Figure 9-2
Particle Motion in Different Water Depths
9-3
10581090
Ocean Wave Energy
Figure 9-3
Vector Field for Particle Motion in Waves [1].
In deep water, wavelength is directly proportional to wave period squared. Therefore, a 10-
second wave is four times longer than a 5-second wave, and it will begin to feel the bottom in
water that is four times as deep. Since the rate at which a wave travels (its phase velocity) is
equal to wavelength divided by period, a 10-second period wave travels twice as fast as a 5-
second wave. The combined potential and kinetic energy of the waves travels at the velocity of
the wave group, which in deep water is equal to half the phase velocity.
Wave-power linear density (kilowatts per meter (kW/m) of wave crest width) is defined as the
flux of energy across a vertical plane intersecting the sea surface and extending to the depth of
no sub-surface orbital motion (which is half the wavelength of the longest harmonic component).
For a 16-second wave, this depth is 200 m, which is the approximate depth of the continental
shelf edge.
The power of ocean waves is expressed in kilowatts per meter of wave crest front. Figure 9-4
depicts the power flux as the energy that crosses through a vertical plane 1 meter in width.
Annual averages range from 10 kW/m to 100 kW/m depending on site location.
Figure 9-4
Wave Power Flux
9-4
10581090
Ocean Wave Energy
9.2 U.S. Wave Energy Highlights: Mid-2009 to Late-2010
Interest in ocean wave renewable energy continues to grow in the United States. This section
presents a brief accounting of notable federal and state developments as well as device
manufacturer activities within the sector from the second half of 2009 through October 2010.
9.2.1 Federal- and State-Level Highlights
9.2.1.1 Federal-Level Developments
In 2010, the U.S. Congress authorized $50 million to conduct research and development on
technologies (wave, tidal, ocean current, in-stream hydrokinetic, and ocean thermal), as well as
conventional hydropower (any technology that uses a dam or diversionary structure). The FY09
appropriation was $40 million, which marked a significant increase in funding from the $10
million appropriated in 2008, which was the first year wave and marine hydrokinetic power
research was supported by DOE.
DOE’s wave and tidal power research is focused on assessing the potential recoverable energy
from these resources in the United States and facilitating the development and deployment of
technologies to fully realize this potential. DOE’s priorities for wave and hydrokinetic power
include:
• Facilitating the deployment of prototypes and collecting data on the energy conversion
performance and environmental impacts of the devices;
• Determining the available, extractable and cost effective resources in the United States;
• Characterizing and comparing the wide variety of existing wave and hydrokinetic
technologies;
costs; and
• Minimizing the cost, time and negative impacts associated with siting projects.
In 2008, the majority of DOE funding for wave and tidal power was awarded to specific
technology and project development efforts, selected through a competitive process. These
awarded efforts included:
• Preparation of detailed design, manufacturing and installation drawings of a bi-directional air
turbine for application in a floating oscillating water column.
• Engineering design, baseline environmental studies, and license construction and operation
applications to help Pacific Gas and Electric (PG&E), the largest investor-owned utility in
California, develop a hub to deploy wave energy converters and connect them to the grid.
9-5
10581090
Ocean Wave Energy
In addition, DOE selected and funded two National Marine Renewable Energy Centers: one at
the University of Hawaii, and a second run jointly by Oregon State University and the University
of Washington. Further, the DOE funded a market acceleration program, consisting of
nationwide wave and tidal hydrokinetic resource assessments, and a collaborative project to
address navigation and environmental issues as well as clarify the permitting process. EPRI was
selected by the DOE to conduct the national wave energy resource assessment. EPRI expects to
complete the national wave energy resource assessment in early 2011.
Two FY09 solicitations were issued by the Wind and Hydropower Technologies Program of the
DOE in April 2009: 1) a Funding Opportunity Announcement (FOA) directed at industry
partners and industry-led teams; and 2) a Program Announcement (PA) directed at DOE national
laboratories to address technical challenges in water power development as well as market
acceptance barriers. The industry FOA consisted of six parts:
1. Marine and Hydrokinetic Energy Conversion Device or Component Design and
Development
2. Marine and Hydrokinetic Site-specific Environmental Studies/Information
3. Advanced Water Power Market Acceleration Projects/Analysis and Assessments
4. Hydropower Grid Services
5. Environmental Mitigation Effectiveness
6. University Hydropower Research Program.
The Topic 3 Market Acceleration Projects/Analysis and Assessments consisted of six subparts:
• An assessment of off-shore ocean current energy resources along the U.S. coastline,
excluding tidal currents, to determine maximum practicably extractable energy.
• An assessment of in-stream hydrokinetic energy resources—defined as energy that can be
extracted from free-flowing water in rivers, lakes, streams or man-made channels without the
use of a dam or diversionary structure—in the United States to determine maximum
practicably extractable energy.
• An assessment of projected life-cycle costs for ocean thermal energy conversion in the
United States over time.
• An assessment of global and domestic U.S. ocean thermal energy resources to determine
maximum practicably extractable energy.
• An assessment of projected life-cycle costs for wave, tidal, ocean current, and in-stream
hydrokinetic power in the United States over time.
• An assessment of the energy resources available from installing power stations on non-
powered dams and in constructed waterways and the construction of new pumped storage
facilities in the United States to determine maximum practicably extractable energy.
9-6
10581090
Ocean Wave Energy
On September 9, 2010, DOE announced the results of its FY2010 solicitation from marine and
hydrokinetic energy technology development projects. Substantial match with non-federal
dollars is required of grantees under the DOE funding opportunity. Wave energy proposals
selected for funding included:
• $2.4 million grant to Ocean Power Technologies (OPT) to deploy a full-scale 150-kW
PowerBuoy system at its Reedsport Oregon site
• $1.8 million grant to Northwest Energy Innovations to verify ocean wavelength performance
for a new point absorber wave energy device.
• $2.4 million grant to Wavebob, LLC, to develop and test an advance power take-off device
for their point absorber wave energy device.
• Several smaller grants for projects related to wave energy device and device component
development and monitoring.
On August 3, 2009, Minerals Management Service (now the Bureau of Ocean Energy
Management, Regulation, and Enforcement; BOEMRE) and FERC issued guidance on the
regulation of hydrokinetic projects on the Outer Continental Shelf (OCS).
On October 26, 2010, DOE, BOEMRE, and the National Oceanic and Atmospheric
Administration (NOAA) announced awards of nearly $5 million for environmental research
projects to advance ocean renewable energy. Funded projects included:
• Baysian Integration for Marine Spatial Planning and Renewable Energy Siting (Parametrix,
$499,000)
• Characterization & Potential Impacts of Noise Producing Construction & Operation
Activities on the OCS (Cornell Lab of Ornithology, $499,000)
• Developing Environmental Protocols and Modeling Tools to Support Ocean Renewable
Energy and Stewardship (University of Rhode Island, $745,000)
• Evaluating Acoustic Technologies to Monitor Aquatic Organisms at Renewable Sites
(University of Washington, $746,000)
• Protocols for Baseline Studies and Monitoring for Ocean Renewable Energy (Pacific Energy
Ventures, $499,000)
• Renewable Energy Visual Evaluations (University of Arkansas, $497,000)
• Sub-Seabed Geologic Carbon Dioxide Sequestration Best Management Practices (University
of Texas-Austin, $497,000)
• Technology Roadmap for Cost Effective, Spatial Resource Assessments for Offshore
Renewable Energy (University of Massachusetts – Marine Renewable Energy Center,
$748,000)
9-7
10581090
Ocean Wave Energy
9.2.1.2 State of Hawaii Projects
In October 2008, the University of Hawaii (UH) was selected as one of two national sites for
development of a Marine Renewable Energy Center (MREC). The Hawaii MREC is managed by
the Hawaii Natural Energy Institute (HNEI) at UH and funded by DOE at approximately $1
million per year with equivalent cost share from UH and industrial partners. The center will be
an international partnership between academia, industry, local and federal government agencies,
and NGOs. The objectives of the Hawaii project are to facilitate the development and
implementation of commercial wave energy systems with one or more systems deployed and
supplying power to the local grid at greater than 50% availability within five years, and to assist
the private sector to move ocean thermal energy conversion (OTEC) systems beyond proof-of-
concept to pre-commercialization.
The MREC will work closely with energy developers to conduct supporting research on system
performance and survivability, grid integration and environmental impacts, including completing
necessary environmental studies and assisting industrial partners to acquire required permits.
Partners include local engineering firms familiar with the permitting process.
MREC proposes to build upon current and proposed marine energy projects in Hawaii to
accelerate establishment of up to three field test facilities for hydrokinetic systems and one for
OTEC component testing. Proposed wave energy test sites include Pauwela Point on the
northeast coast of Maui, in cooperation with Maui Electric Company and Oceanlinx; the
Kaneohe Marine Corps Base on Oahu, where Ocean Power Technologies maintains an ongoing
program; and off the Makai Research Pier located west of Makapuu Point on the eastern tip of
Oahu. The latter site is proposed for obtaining long-term data series on wave energy resources,
research on corrosion and innovative materials, and an easily accessible site for deploying and
testing small wave energy conversion devices and components. The pier is already permitted for
a range of marine research activities.
U.S. Navy–OPT Project at Kaneohe Bay. Ocean Power Technologies continues testing of its
PowerBuoy a mile offshore at the Kaneohe Bay Marine Corps Base on Oahu. This project is
exempt from FERC permitting and licensing requirements. An Environmental Assessment
conducted to comply with the National Environmental Policy Act (NEPA) resulted in a Finding
of No Significant Impact (FONSI).
Oceanlinx Maui Wave Energy Project (P-13521). FERC issued a Preliminary Permit to
OceanLinx Hawaii, LLC, on November 25, 2009, for it wave energy project off the coast of
Maui, about 0.6 miles north of Pauwela Point. The proposed project would deploy oscillating
water column wave energy conversion technology with a combined nameplate capacity of 2.7
MW. During the first year of the project, Oceanlinx worked with Maui Electric Company
(MECO) to refine the design and routing of the undersea cable to connect the Project to MECO’s
grid. MECO reported a preliminary cost estimate of $11–$15 million. This estimate significantly
exceeds the initial estimates of $5 million; consequently, the project partners are exploring cost-
saving alternatives, including deploying a smaller device (up to 1.0 MW) in shallower water
closer to shore. Commencement of environmental studies awaits selection of a lower-cost
deployment site closer to shore.
9-8
10581090
Ocean Wave Energy
9.2.1.3 State of Oregon Projects
The Oregon Innovation Council granted $4.2 million to create the non-profit Oregon Wave
Energy Trust (OWET) in 2007. It is OWET’s mission to establish the state as the preeminent
developer of wave energy in the United States, with the goal of producing 500 MW of clean
power from its ocean—about 3%-5% of the state’s energy—by 2025. To achieve this mission,
OWET’s strategy is to maintain “technology neutrality” and focus its resources on reducing the
barriers hindering the emerging wave energy industry’s movement forward toward commercial
development. OWET’s activities are grouped within four major program areas:
• Stakeholder Education and Engagement: Specific activities include: a) Coastal Community
Open Houses, b) creating a statewide network of coastal economic and community
development advisors, c) facilitating development of organized fishermen groups to
participate in wave energy planning, d) showcasing at consumer events, e) producing a wave
energy conference, and more.
• Regulatory and Policy: Specific activities include development of Regulatory Roadmaps to
help wave energy developers navigate the complex network of state and federal permit and
license requirements. OWET actively monitors state legislative activities and provides wave
energy industry information to legislative representatives and committees.
• Market Development: OWET recently completed a Utility Market Initiative. This extensive
project sought to produce an effective market strategy to integrate wave energy projects into
the electric utility system and establish technical requirements to connect into the grid [2].
OWET hopes to create a utility “pull” and wave energy “push” to help meet the target
production goals. In addition to the Utility Market Initiative, an economic assessment of the
wave energy industry in Oregon—which will include economic data for the fishing and
crabbing industries—is underway. On March 25, 2010, OWET released its Wave Energy
Infrastructure Assessment in Oregon report [3].
• Research: OWET directs and funds environmental and applied research projects to answer
key questions about wave energy development. It has completed baseline assessment studies
on seabirds and whale migration at the proposed wave energy project sites. During 2009 and
2010, OWET supported additional research on seabirds, crab distribution, EMF, sediment
transport, and create a planning tool to model cumulative effects of wave energy projects.
Coos Bay OPT Wave Park (P-12749). On August 10, 2010, FERC issued a Preliminary Permit
for the Coos Bay OPT Wave Park Project. The project would be located in the Pacific Ocean
between 2.5 and 3.0 miles off the coast near Coos Bay, Oregon. The project would consist of 200
PowerBuoys having a combined, installed capacity of 100 MW, as well as an approximately 3.4-
mile subsea transmission cable and other facilities.
Reedsport OPT Wave Park (P-12713). On February 1, 2010, OPT submitted a License
Application (LA) for the 1.5-MW Reedsport OPT Wave Park project. The company plans to
deploy a single PowerBuoy by the end of 2010 and begin deployment of the additional nine
Power Buoys in summer 2011. FERC accepted the LA on June 1, 2010, deeming the Application
ready for environmental analysis. On July 28, 2010, OPT and several intervenors to the FERC
licensing process entered into a Settlement Agreement resolving issues associated with issuance
of an original license for the project, and requesting a license term of 35 years.
9-9
10581090
Ocean Wave Energy
Reedsport OPT Wave Park, Phase III (P-13666). Ocean Power Technologies (OPT) filed a
Preliminary Permit application on February 2, 2010, for Phase III of its Reedsport commercial
wave energy park. Phase III is intended to expand the project from the initial 1.5-MW project to
50 MW, comprising 100 500-kW PowerBuoys. The Phase III project is intended to expand the
size of the project site, potentially utilizing the transmission cable and corridor associated with
the 1.5-MW project. OPT also intends to use the results of studies conducted at the 1.5-MW
project to inform the evaluation of the License Application for the 50-MW project.
Northwest National Marine Renewable Energy Center. Oregon State University (OSU) and the
University of Washington (UW) were awarded $6.25 million by DOE to develop the Northwest
National Marine Renewable Energy Center (NNMREC). OSU plans to deploy various devices
and subsequent environmental measurement devices at a location near Newport to study new
technologies and evaluate potential environmental effects.
9.2.1.4 State of California Projects
Pacific Gas and Electric Humboldt WaveConnect Project (P-12779). On October 28, 2010,
PG&E announced that it was suspending the Humboldt WaveConnect project. The
announcement cited permitting challenges and costs as reasons for suspending the project.
PG&E’s service territory borders 960 km of Pacific coastline, with wave power densities of 20–
40kW/m, making wave power a renewable energy resource of strategic value to the utility. The
WaveConnect project’s goal was to assess wave energy’s potential and examine the regulatory
and environmental issues associated with such a facility’s development. WaveConnect was
intended to provide the infrastructure to test small arrays of commercial wave energy conversion
devices in California, and therefore allow the emerging wave power industry and PG&E to gain
an understanding of the full life cycle of deployed technologies (Figure 9-5).
Figure 9-5
Schematic of PG&E WaveConnect System
9-10
10581090
Ocean Wave Energy
In late 2008, DOE selected PG&E to receive a cost-sharing grant of $1.2 million. In early 2009,
PG&E received a decision of approval from the California Public Utility Commission to spend
$4.8 million of ratepayer-based funds toward the estimated design and licensing costs for the
pilot WaveConnect project.
PG&E received preliminary permits from FERC for the Humboldt site and another site in
Mendocino County in early 2008. In early May 2009, PG&E announced that it was dropping the
Mendocino County study as Port Noyo Harbor in Fort Bragg, California, was found to be
unsuitable to support a pilot wave energy project.
Central Coast WaveConnect (P-13641). FERC issued PG&E a Preliminary Permit for the
Central Coast WaveConnect Project on May 14, 2010. This project would be located off the
Santa Barbara County coast and is similar in concept to the Humboldt WaveConnect project.
Potential advantages of the Central Coast site include potential for large-scale build-out, longer-
term operation, and availability of baseline environmental data.
Green Wave Mendocino Project (P-13053) and San Luis Obispo Project (P-13052). FERC
took final action on October 26, 2010 canceling Green Wave Energy Solutions’ two projects off
the cities of Mendocino, and Morro Bay, California. These Preliminary Permits were cancelled
for failure to file a notice of intent (NOI) to file a license application and a pre-application
document (PAD) for each project.
Sonoma Coast Hydrokinetic Energy Project Del Mar Landing (P-13376), Sonoma Coast
Hydrokinetic Energy Project Fort Ross (South) (P-13377), and Sonoma Coast Hydrokinetic
Energy Project Fort Ross (North) (P-13378). FERC issued Preliminary Permits to the Sonoma
County Water Agency for these three projects on July 9, 2009. Each of these projects would
deploy either oscillating water column devices and/or buoy-type wave energy conversion devices
with a combined capacity of 2 to 5 MW. The Preliminary Permit applications noted possible
future expansion of each project up to 40 to 200 MW.
SWAVE Catalina Green Wave Energy Project (P-13498). FERC issued a Preliminary Permit
for the SWAVE project to Scientific Applications & Research Associates, Inc. (SARA) on
September 15, 2009, for an array of 10 to 40 SWAVE buoys approximately 0.75 mile off the
west coast of Santa Catalina Island. SARA voluntarily surrendered the Preliminary Permit for the
project on October 13, 2010, after determining that the project is not feasible at this time.
San Onofre OWEG Electricity Farm (P-13679). FERC issued JD Products, LLC, a Preliminary
Permit on October 29, 2010, for the proposed San Onofre OWEG Electricity Farm Project. The
project would consist of up to 11,443 Ocean Wave Electricity Generation (OWEG) units, an
experimental technology, with an estimated total installed capacity of 3,186 MW. The land-
based portions of the project would be located on a site adjacent to the San Onofre Nuclear
Generating Station (SONGS). The application notes that SONGS Station 1 is inoperative, and
Stations 2 and 3 will be decommissioned by 2022. The application proposes to use the
distribution infrastructure currently associated with SONGS. The initial plan is to install only 50
of the experimental OWEG units for an estimated capacity of 8.9 MW.
9-11
10581090
Ocean Wave Energy
9.2.1.5 State of Washington Projects
Grays Harbor Ocean Energy Project (P-13058). FERC cancelled the Preliminary Permit for this
project on September 21, 2010. Grays Harbor Ocean Energy Company was issued the
Preliminary Permit on July 31, 2008, but failed to submit the semi-yearly progress reports,
Notice of Intent, and Draft Application as required by the Permit.
9.2.1.6 State of Maine Projects
Castine Harbor and Bagaduce Narrows (P-12777). Maine Maritime Academy filed a
declaration of intention and petition for relief from the requirement to obtain a license from
FERC to deploy and test hydrokinetic devices at two coastal sites (Bagaduce Narrows and
Castine Harbor) in Hancock County, Maine. FERC declared on March 23, 2010, that the
contemplated activities do not require licensing under the Federal Power Act.
9.2.2 U.S. Developer and Project Deployment Highlights
Ocean Power Technologies
Ocean Power Technologies (OPT) is pacing domestic wave energy development. In addition to
OPT, the United States has roughly a half dozen developers that are at an early stage of
technology development. Recent highlights from OPT projects include:
• Reedsport, Oregon: On July 28, 2010, OPT signed a Settlement Agreement with 11 federal
and state agencies and three non-governmental stakeholders for its 10-buoy, 1.5-MW
commercial power project off the coast at Reedsport, Oregon. This agreement followed
submission of the Final License Application for the project on February 1, 2010. OPT has
initiated the permitting and licensing process to expand this site to 50 MW capacity.
• Kaneohe Bay, Hawaii: On September 27, 2010, OPT announced that it had completed the
first-ever grid connection of a wave energy device in the United States at the Marine Corps
Base Hawaii (MCBH), Kaneohe Bay, Oahu, Hawaii. The device was deployed on December
14, 2009 about three quarters of a mile offshore in approximately 100 feet of water.
• Cornwall, UK: OPT is poised to be the first customer at the South West of England Regional
Development Agency’s (SWRDA) "Wave Hub" project. The Wave Hub system underwent
its first full test in early November 2010.
9.3 Worldwide Wave Energy Highlights: Mid-2009 to Late-2010

Interest in ocean wave renewable energy is strong worldwide. This section presents a brief
accounting of notable activities within the sector outside of the United States during the second
half of 2009 and through October 2010.
9-12
10581090
Ocean Wave Energy
9.3.1 United Kingdom
The UK is maintaining its stature as the global leader in wave energy technology development.
In an effort to solidify its leadership position in WEC development, the UK has established an
installed marine energy capacity goal of 2 GW by 2020 and adopted an energy policy designed
to attract and support WEC developers and equipment testing. The country contains three wave
testing facilities and is home to many wave energy developers.
9.3.1.1 UK Wave Energy Developers
WEC developers in the UK with updates to report include AquaMarine Power, Checkmate Sea
Energy, Orecon, and Pelamis WavePower.
9.3.1.1.1 Aquamarine Power
Aquamarine Power's Oyster system (Figure 9-6) was officially launched at the European Marine
Energy Centre (EMEC) in Orkney on November 20, 2009. The Oyster is a hydroelectric wave
energy converter that consists of an oscillator fitted with pistons and fixed to the near-shore sea
bed. Each passing wave activates the Oscillator, pumping high pressure water through a sub-sea
pipeline to the shore. Onshore, conventional hydroelectric generators convert this high-pressure
water into electrical power. The Oyster has been under development by Aquamarine Power since
2005, in partnership with the marine energy research group at Queens University, Belfast.
9.3.1.1.2 Checkmate Sea Energy
Checkmate is aiming for commercial production of its Anaconda wave energy conversion
technology by 2014. Full-scale devices are expected to be up to 200 meters in length and capable
of generating 1 MW each. In the first half of 2010, Checkmate conducted a series of scale-
model, wave-tank trials at Strathclyde University, Glasgow, Scotland. These tests were designed
to assess fatigue life and power output. The Anaconda is made from fabric and natural rubber,
and uses the incoming waves to drive a turbine in its tail as illustrated in Figure 9-7.
9-13
10581090
Ocean Wave Energy
Figure 9-6
Aquamarine Power Oyster
Figure 9-7
Illustration of Checkmate Anaconda
9.3.1.1.3 Orecon
Orecon Limited’s Multi Resonant Chamber (MRC) employs the oscillating water column
principle. Orecon was forced to close in early 2010 after investors withdrew funding and UK
government support was directed elsewhere. Less than a year earlier, Orecon became the latest
wave energy development partner at Wave Hub, taking the place of Australian company
Oceanlinx, which employs the same oscillating water column principle and had decided to
conduct its next deployment in Australian waters.
9-14
10581090
Ocean Wave Energy
9.3.1.1.4 Pelamis WavePower
In July 2010, E.ON deployed a second-generation Pelamis WavePower device at EMEC.

Following four days of successful testing, the device was removed for inspection and prepared
for redeployment.
9.3.1.2 UK Wave Energy Test Centers
The UK's wave testing facilities include the New and Renewable Energy Center (NaREC) with
sub-scale prototype testing in a wave tank; the European Marine Energy Center (EMEC) with
single full-scale prototype testing in natural waters; and Wave Hub, for arrays of full-scale
prototypes tested in natural waters. Each facility is described below.
9.3.1.2.1 NaREC
NaREC is a leading research and development platform for new, sustainable, and renewable
energy technologies located in Blythe, England. Its range of development, testing and
consultancy services work to support the evolving energy industry and transform innovative new
technologies into commercial successes. NaREC provides the emerging marine renewables
industry the support it needs to transform winning concepts into commercial successes. NaREC
services include:
• Complete in-house prototype development facilities for wave technology, including a wave
tank
• Mechanical and electrical design engineering and procurement
• Electrical engineering consultancy and support for power conversion and drive train
development
• Complete system testing from marine environment to grid connection.
• Resource and feasibility assessment and consultancy
• Market analysis and research
• Project management, funding and investment coordination.
In July 2009, the Department for Energy and Climate Change awarded £10 million to NaREC for
Project Nautilus, a rotary test rig that will facilitate development of marine renewable energy
drive trains. The facility will allow developers to test devices ranging from medium-scale
prototypes to full-scale commercial devices.
9.3.1.2.2 The European Marine Energy Centre (EMEC)
In operation since 2003, EMEC has to date tested the Pelamis Wave Energy Converter prototype
at its facility. EMEC envisions adding three new devices—Ocean Power Technologies' 150-kW
PowerBuoy, Oyster, and a second-generation P2 Pelamis WavePower. Figure 9-8 illustrates the
three WEC systems.
9-15
10581090
Ocean Wave Energy
The EMEC test site at Billia Coo in mainland Orkney provides the world’s only multi-berth,
open-sea test facility for wave energy converters. The EMEC offices and data facilities are in
Stromness. Orkney was chosen because of its natural and man-made resources. The wave test
facility site receives uninterrupted Atlantic waves of up to 15 meters. Orkney is also the most
northerly community connected to the UK national grid, has excellent harbor facilities, and a
significant professional community experienced in working with renewable energy.
Aquamarine
Oyster
OPT
150 kW
Pelamis P2
OPT
PowerBuoy 150
Figure 9-8
Expected EMEC Testing Configuration in 2011
9.3.1.2.3 The UK Wave Hub
Wave Hub is a £42 million test facility for wave energy technology. In early September 2010,
Wave Hub deployed a 12-ton, grid-connected socket on the ocean floor 10 miles offshore of
Cornwall, South West England in 55 meters of water. Four 300-meter “tails” extending from the
hub will each constitute a test berth for deploying a wave energy conversion device. The project
funders were the South West Regional Development Agency (£12.5 million), the European
Regional Development Convergence Fund Programme (£20 million), and the UK government
(£9.5 million).
9-16
10581090
Ocean Wave Energy
9.3.2 Continental Europe
9.3.2.1 WaveRoller
The WaveRoller prototype of a bottom-mounted flat-plate oscillating device developed by the

Finnish company AW-Energy was deployed in April 2007 at Peniche, 100 km north of Lisbon.
In 2008, AW-Energy announced plans to construct a 1-MW plant. The company now has all the
necessary permits to install the project and connect it to the grid.
9.3.2.2 Ongoing Testing and Development in Denmark, Norway and Sweden
A grid-connected wave energy test site at Nissum Bredning in the northwestern corner of
Denmark was built to enable various technology developers the ability to test and demonstrate
their technologies at different scales. A 24-m long, 1:10 scale model of one such WEC device,
the Wave Star, was tested at Nissum Bredning until August 2008. Figure 9-9 shows the Wave
Star at Nissum Bredning with the buoys raised in the survival position. On September 18, 2009,
Wave Star deployed its next generation device at Hanstolm, Denmark. The 40-m long device is a
test section of their commercial 500-kW machine (Figure 9-10).
Figure 9-9
Wave Star 1:10 Machine with Buoys Raised
Figure 9-10
Wave Star 1:2 Machine with Buoys Raised
9-17
10581090
Ocean Wave Energy
Floating Power Plant A/S has constructed a 37-m model for a full off-shore test at Vindeby
offshore wind turbine park, located off the coast of Lolland in Denmark (Figure 9-11). The test
system, named Poseidon 37, is 37 meters wide, 25 meters long, 6 meters high (to deck), and
weighs approximately 300 tons. The test plant was launched in Nakskov Harbour in May 2008
and was towed to the test site and installed in August 2008. Poseidon is based on the principle of
oscillating water columns. On June 14, 2010, Poseidon was redeployed at a test site in Onsevig,
Denmark, to begin Test Phase 2. The platform has demonstrated its stability even when the wind
turbines are operating at peak efficiency.
In Sweden, the Seabased AB system is being tested. It uses a three-phase, permanent-magnet,

linear generator and is especially developed to be used in ocean bed arrays and directly driven by
point absorbers (buoys) on the surface. Figure 9-12 shows the preparation of the generator for
launch off the coast of Lysekil, Sweden. The WEC unit consists of a buoy coupled directly to the
rotor of a linear generator by a rope. The tension of the rope is maintained with a spring pulling
the rotor downwards. The rotor moves up and down at approximately the same speed as the
wave. The linear generator has a uniquely low pole height and generates electricity at low wave
amplitudes and slow wave speeds. Directly driven linear wave energy converters are deployed
and coupled in arrays at intervals of 25 to 50 meters. The Swedish Energy Authority awarded
Seabased AB 139 million Swedish Kroner on February 11, 2010, toward the 250 million
Swedish Kroner needed to build a 400- to 500-unit, 10-MW project off the coast of Smögen,
Sotenäs, Sweden.
Figure 9-11
Floating Power Plant AS Poseidon
9-18
10581090
Ocean Wave Energy
Figure 9-12
Seabased AB Linear Generator
9.3.3 Australia, New Zealand and Tasmania
Australia, New Zealand and Tasmania are also pursuing wave energy projects. Three known
companies involved in developing wave energy in Oceana include BioPower Systems Pty,
Oceanlinx, and a partnership between Industrial Research Limited, the National Institute of
Water and Atmospheric Research (NIWA), and energy industry consultants Power Projects
Limited.
BioPower Systems Pty Ltd, a Sydney-based company, announced on March 1, 2010, that it had
secured land access and onshore development rights and project intellectual property for a
commercial-scale project at Port Fairy, Australia. The project plan calls for initial deployment of
a 250-kW bioWAVE system, to be followed by a commercial array of 1-MW bioWAVE units.
Preliminary assessments indicate the site could support up to 100 MW of installed capacity.
Figure 9-13 shows a subscale model of the bioWAVE.
Oceanlinx Limited, which manufactures an oscillating water column device (Figure 9-14), has
made steady advances toward a commercial device:
• Port Kembla (New South Wales, Australia): Oceanlinx deployed its third-generation, 1:3
scale model, pre-commercial device, known as the Mk3PC, on February 26, 2010, and
connected it to the grid on March 19. On May 14, 2010, the device broke free of its mooring.
Otherwise, the Mk3PC performed as designed, thereby validating the design for Oceanlinx’s
full-scale 2.5-MW commercial device.
• Hawaii: A Memorandum of Understanding was signed with Maui Electric Co. (MECO) in
Hawaii for up to 2.7 MW. MECO will contribute the transmission cables, both submerged
and land-based, and the project is scheduled to be deployed around the 2012 time frame.
9-19
10581090
Ocean Wave Energy
Figure 9-13
BioPower BioWave
Figure 9-14
Oceanlinx Oscillating Water Column
9.4. Wave Power and Energy Resources
A number of sources provide wave power and energy resource data for assessing potential sites.
These sources include in situ measurements, satellite measurements, and wind-wave models.
In 2003, EPRI developed a methodology for estimating wave energy resource and predicting
wave energy conversion performance [4]. The EPRI assessments are for offshore wave energy
conversion only and were based on the decades of statistical wave parameter data archived on
the NOAA NDBC website. (www.ndbc.noaa.gov).
EPRI’s preliminary estimate of the available U.S. offshore wave energy resource is 2,100
TWh/yr (exclusive of the Bering Sea north of the Aleutian Islands and exclusive of any resource
less than 10 kW/m). The estimate is broken down regionally in Figure 9-15.
9-20
10581090
Ocean Wave Energy
Total Energy = 2,100 TWh/yr (excluding the

Bering Sea) for sites with >10 kW/m
New England
Southern AK and Mid-Atlantic
1,250 TWh/yr 110 TWh/yr
WA, OR, CA
440 TWh/yr
Northern HI
300 TWh/yr
Extracting 15% and converting to
electricity at 80% yields 255 TWh/yr
Figure 9-15
U.S. Wave Energy Resources
In terms of extractable wave energy resource for our preliminary assessment, the EPRI team
assumed that 15% of the available resource could be extracted based on societal constraints of
30% coverage of the coastline with a 50% efficient wave energy absorbing device. Assuming
typical power train efficiencies of 90%, and a plant availability of 90%, the electricity produced
is about 260 TWh/yr, which is equal to an average power of 30,000 MW or a rated capacity of
about 90,000 MW. This is approximately equal to the total 2004 energy generation from
conventional hydro power, which is about 6.5% of total 2004 U.S. electricity supply.
Under DOE sponsorship EPRI, with Virginia Tech and NREL, began a new assessment of the
national U.S. offshore available and practically recoverable wave energy resource in September
2008. The final product is expected to be completed by March 2011. It will include a geospatial
database, verified and validated by a third party, that displays wave power densities for specific
geographic information system (GIS) coordinates along with user-selectable annual and monthly
statistical products. The database includes the probability distributions of sea state parameters,
which developers need to predict the annual and monthly energy yield of their devices and
projects.
The wave power density, significant wave height (Hmo), energy wave period (Te), the directions
of the primary wind-wave and primary and secondary swell waves, and bathymetry (iso-baths)
will be displayed on a GIS map from a depth of 50 meters out to either 200 meters depth or 50
nautical miles from shore (whichever is closer to the shoreline). The GIS map will include the
entire coastline of the United States, including Alaska, Hawaii, and Puerto Rico. Figure 9-16
shows the NWW3 grid resolution for the offshore United States.
9-21
10581090
Ocean Wave Energy
Figure 9-16
NOAA Wave Watch III Grid Resolution
The influence of the ocean floor reduces wave power levels in shallow waters (<50 m).
Submerged features such as canyons can also focus energy, leading to hot spots in close
proximity to shore. A number of shallow-water wave transformation models take into account
the bathymetry to calculate near-shore wave data. High-resolution bathymetry is required. The
input boundary condition for these shallow water models is the output from the NOAA
WAVEWATCH III model at the edge of the Outer Continental Shelf. Experience with shore-
based devices shows a need for extensive modeling in such locations.
9.4.1 Measurement Data Sources
The two largest inventories of long-term measured wave data in the United States are maintained
by the National Data Buoy Center (NDBC) of NOAA (www.ndbc.noaa.gov), and by the Coastal
Data Information Program (CDIP) of Scripps Institution of Oceanography (http://cdip.ucsd.edu/).
NDBC data buoys are equipped with strapped-down accelerometers for measuring wave
conditions derived from buoy heave response. Wave spectra are computed from 20-minute time-
series measurements of sea surface elevation changes, and these records are archived at one-hour
intervals. West Coast and Hawaii reference stations are shown in Figures 9-17 and 9-18,
respectively.
9-22
10581090
Ocean Wave Energy
Figure 9-17
West Coast Reference Stations
Figure 9-18
Hawaii Reference Stations (Point Makapuu is CDIP 0098)
9.4.2 Wind-Wave Model Data Sources
The operational ocean wave predictions of NOAA/NWS/NCEP are performed using the wind-
wave model NOAA WAVEWATCH III using operational products of NCEP as input. The wind-
wave model is implemented on global and regional scales, as shown in Figure 9-19.
9-23
10581090
Ocean Wave Energy
Figure 9-19
NOAA Wave Watch III Global Coverage
NOAA WAVEWATCH III [5, 6] is a third-generation wave model developed at NOAA/NCEP.

It is a further development of the model WAVEWATCH I, developed at Delft University of
Technology, and WAVEWATCH II, developed at NASA, Goddard Space Flight Center. NOAA
WAVEWATCH III differs from its predecessors in many important ways such as its governing
equations, model structure, numerical methods, and physical parameterizations.
NOAA WAVEWATCH III solves the spectral action density balance equation for wave number-
direction spectra. The implicit assumption of this equation is that water depth and current, as
well as the wave field itself, vary on time and space scales that are much larger than that of a
single wave. A further constraint is that the parameterizations of physical processes included in
the model do not address conditions where the waves are strongly depth-limited. These two basic
characteristics imply that the model can generally be applied on spatial scales (grid increments)
larger than 1 to 10 km and outside the surf zone.
All regional models obtain hourly boundary data from the global model. All models are run on
the 0000 GMT, 0600 GMT, 1200 GMT and 1800 GMT model cycles, and start with a 6-hour
hindcast to assure continuity of swell. The models provide 126-hour forecasts, with the exception
of the North Atlantic Hurricane model, which provides a 72-hour forecast.
Graphical products are maps and spectra; binary and text products are gridded binary (GRIB)
files, spectral data, and spectral bulletins. Detailed parameter definitions can be found in the
NOAAWAVEWATCH III Users’ Manual [6].
9-24
10581090
Ocean Wave Energy
9.4.3 Available Offshore Wave Resource Calculation Methodology
EPRI, Virginia Tech and NREL have jointly developed and are using a 10-step methodology to
calculate offshore wave energy resource estimates for the 2011 DOE Wave Energy Resource
Assessment Project. The methodology is presented below. It incorporates archived
WAVEWATCH III hindcast, fully-partitioned, spectral wave parameters at over 100,000 coastal
grid points off the U.S. coastline.
Step 1: Select deep-water "characterization stations" at depths greater than 200 m, where NDBC
buoys are co-located with WAVEWATCH III spectral output locations, and for each station
develop a joint probability distribution (JPD) table of overall significant wave height (Hm0) and
wave energy period (Te).
• Calculate Hm0 and Te using spectral moments (m0 and m-1) of the non-directional wind-wave
variance density spectrum (hereinafter referred to simply as the spectrum) from those times
when the measurement archive and hindcast archive overlap.
• Develop a measured JPD table and a hindcast JPD table for each calendar month, and
characterize sea state probability as the fraction of time per month that each Hm0 and Te
combination was measured or hindcast.
• Compile annual JPD tables by averaging all months for each sea state bin.
• For each measured spectrum and hindcast spectrum in the time series used to populate the
JPD tables, calculate wave power linear density (P) using the spectral formula for P.
• Plot hindcast P (y-axis) against measured P (x-axis) for all spectra in the time series at a
given characterization station and calculate the correlation coefficient between hindcast and
measured wave power density.
• Develop similar scatter plots and correlation coefficients for hindcast vs. measured
significant wave height and hindcast vs. measured wave energy period.
Step 2: For each annual JPD within each region, select a subpopulation of sea state bins,
including all sea states that have a >0.2% probability of occurrence (17.5 hr/yr).
Step 3: For each selected sea state bin, develop an average measured spectrum and an average
hindcast spectrum, calculate the wave power density associated with each of these average
spectra, and compare the results with measured and hindcast wave power densities determined in
Step 1 to check on the spectral averaging results.
Step 4: For each selected sea state bin, test a few different theoretical spectral formulas (e.g.
Bretschneider, JONSWAP, etc.) based on Hm0 and the peak wave period (Tp) of the partitioned
wind wave and multiple swell components, which are also produced by WAVEWATCH III at
each hindcast time step. Apply the theoretical spectral formula to the partitioned hindcast time
series to reconstruct the overall spectrum.
Step 5: For each "built-up" hindcast spectrum, calculate the wave power density and average all
of the wave power densities to produce a mean wave power density for that sea state bin.
9-25
10581090
Ocean Wave Energy
Step 6: Determine which theoretical spectral formula for the wave train partitions provides the
best agreement with the average wave power density from measured spectra (as determined in
Step 1) for a given region. It is anticipated that the theoretical formula providing the best fit will
differ among regions.
Step 7: Apply the “best fit” theoretical spectral formula to the wave train partitions at all
WAVEWATCH III grid points by region. The sea state parameters for all partitioned wave trains
(not just wind sea and two swell components) have been made available by NOAA for the period
February 1, 2005, through January 31, 2009, specifically for this study.
Step 8: Map the resulting annual and monthly values of Hm0, Te, and P for all WAVEWATCH III
coastal grid points located between the 50 m and 200 m depth contours, and extend mapping to
50 nautical miles offshore wherever the 200 m depth contour occurs within that distance.
Step 9: Keep track of the peak wave direction for each component wave train, and prepare wave
power linear density directional distribution roses for the local wind sea, the primary swell, and
the secondary swell at selected grid points
Step 10: Using the NREL database of MMS (now BOEMRE) lateral administrative boundaries,
calculate the total annual wave energy flux (terawatt-hours per year) offshore for each coastal
U.S. state and Puerto Rico, based on the following wave crossings:
• Across the mapping limit of 50 nautical miles offshore.
• Across the 200-m depth contour. Where this contour "wraps back on itself," the farther
offshore contour will be used, which indicates where waves first "feel the bottom."
• Across the 50-m depth contour. Where this contour “wraps back on itself,” the farther near-
shore contour will be used, indicating where waves finally leave the mapped zone, heading
towards shore, and where finer-resolution bathymetry and wave propagation models must be
used, using NOAA’s fully partitioned WAVEWATCH III data for input.
9.4.4 Practically Recoverable Wave Resource Calculation Methodology
The methodology to estimate the practically recoverable U.S. wave energy converted to
electrical energy will be developed in late-2010 to early-2011 by the EPRI-Virginia Tech-NREL
Project Team and its NOAA, military, university and device/project developer expert advisors.
9.4.5 Wave Power Forecasting
Reliable electric power system operation requires precise balancing of supply and demand. Grid
operators manage supply-demand balance on a minute-to-minute basis considering load forecasts
and using current resources, rules, and procedures. Accurate wave power forecasts will help
system operators meet the challenge of integrating wave-generated power with the electric grid.
In 2007, EPRI investigated the accuracy of forecasting wave power as a function of forecasting
time horizon [7]. In the original Bonneville Power Administration (BPA) co-funded wave energy
study, WAVEWATCH III forecasts were only available at NDBC buoy locations in very deep
9-26
10581090
Ocean Wave Energy
water 100 nautical miles or more from the coastline. The EPRI Project Team accomplished the
forecast accuracy study in two steps:
1. Virginia Tech compared the WAVEWATCH III forecast accuracy with far offshore buoys.
2. SAIC correlated time-lagged data from the far offshore buoys to a few buoys at about 50 m
depth, the depth currently favored for offshore wave power plants.
NOAA WAVEWATCH III now forecasts wave sea states shoreward to the 50-m depth contour.
EPRI conducted a study for Pacific Energy Ventures and Oregon Wave Energy Trust in the late
summer of 2009 making a one-to-one comparison of WAVEWATCH III forecasts with co-
located measurements at a NDBC buoy on the 50-m depth contour [8]. Time-lagged correlations
between far offshore and 50-m depth buoys are no longer required.
9.5 Wave Energy Conversion (WEC) Technology Description
Wave power research programs in industry, government, and at universities have established an
important foundation for the emerging wave power industry over the past decade. In the late
1970s and early 1980s, the UK regarded wave power as an alternative to nuclear generation and
had the most aggressive R&D program in the world. Although the program contributed to
important basic research on optimal control and tuning of wave power conversion devices, it
ultimately stalled as oil prices dropped and government funding ceased. In the past decade,
wave-powered generation has advanced, and in the last five years a half dozen full-scale
prototypes have been tested in natural waters.
9.5.1 Harnessing Wave Energy

Wave energy extraction is complex and many device designs have been proposed. Four of
the best known device concepts and their principle of operation are:
• Point Absorber: A bottom-mounted or floating structure that absorbs energy in all
directions. The power take-off system may take a number of forms, depending on the
configuration of displacers/reactors. The illustration shows a floating buoy; however, a point
absorber could be a bottom-standing device with an upper floater.
• Oscillating Water Column (OWC): At the shoreline, this could be a cave with a
blow-hole and an air turbine/generator in the blow hole. Near shore or offshore, an OWC
device is a partially submerged chamber with air trapped above a column of water. As waves
enter and exit the chamber, the water column moves up and down and acts like a piston. A
column of air, contained above the water level, is compressed and decompressed by this
motion to generate an alternating stream of high-velocity air in an exit blowhole. The air is
channeled through an air turbine/generator to produce electricity.
• Overtopping Terminator: A floating reservoir structure with reflecting arms and a ramp.
Arriving waves overtop the ramp and are contained in the reservoir. The collected water
turns turbines connected to generators as it flows out of the device.
• Attenuator or Linear Absorber: An example of the attenuator principle is a long floating
structure that is orientated perpendicular to the traveling wave front. The structure is
composed of multiple sections that move relative to each other, and this motion activates
hydraulic power converters that drive electrical generators.
9-27
10581090
Ocean Wave Energy
Figure 9-20 illustrates the fundamental principles described, while Figure 9-21 shows examples
of the four machine types summarized above. There are many other design concepts that are
related to these machine types.
Overtopping
Waves
overtopping
the ramp
Reservoir
Figure 9-20
Wave Energy Device Principles
OPT Oceanlinx
Power
Buoy
Pelamis
Wave Dragon
Figure 9-21
Wave Energy Device Concept
9-28
10581090
Ocean Wave Energy
9.5.2 WEC System Developers
Today, a number of small companies are leading the commercialization of technologies to

generate electricity from ocean waves. In 2004, EPRI requested information from all known
WEC device developers [9] and updated the survey in 2006 [10]. EPRI again updated the list of
known developers who have built and tested prototypes as of November 2008 (Table 9-1),
excluding those with only concepts or patents. Survey details are now maintained by the DOE,
which expects to release updated survey results in late 2010.
Table 9-1
Wave Energy Conversion Device Developers as of November 2008
(1) (2) (3)
Device Developer Website Device Name Type Development Status
Able Technologies Wave Pipe Point Absorber Laboratory Proof of Concept
abletechnologiesLLC. com
Artificial Muscles Unknown Point Absorber Experimental
www.artificialmuscles.com
Aquamarine Power Oyster Technology Demonstration
www.aquamarinepower.com
AW Energy WaveRoller Oscillatory Technology Demonstration
www.aw-energy.com
AWS Energy Archimedes Point Absorber Commercial
www.waveswing.com Wave Swing Demonstration
BioPower bioWave Oscillatory Technology Demonstration
www.biopowersystems.com
C-Wave Limited C-Wave Attenuator Experimental
www.cwavepower.com
Checkmate Sea Energy Anaconda Articulating Laboratory Proof of Concept
www.checkmateuk.com linear absorber
College of the North Atlantic Wave Pump Point Absorber Experimental
Ecofys Waverotor Hydrodynamic Experimental

www.ecofys.co.uk Lift
Energiesysteme GmbH ECOWAS III Unknown Experimental
http://members.aol.com/mamoenerg
y/
Finavera (formerly AquaEnergy) AquaBuoY Point Absorber Early Commercial
www.finavera.com
Fred Olsen Ltd Buldra Point Absorber Technology
Demonstration
Hidroflot s.L. Ocean Converter Multiple Pt. Laboratory
www.hidroflot.com Absorbers
Independent Natural Resources SEADOG – Point Absorber Early Commercial
www.inri.us water pump
Kinetic Wave Power Unknown Patent Laboratory Proof of Concept
www.kineticwavepower.com Pending
9-29
10581090
Ocean Wave Energy

(1) (2) (3)
Manchester Univ of Bobber Point Absorber Laboratory Proof of Concept
www.manchesterbobber.com
Motor Wave Motor Wave Point Absorber Experimental
www.motorwavegroup.com
Ocean Energy Ltd Ocean Energy Oscillating Technology Demonstration
www.oceanenergy.ie Buoy (OEBuoy) Water Column
Oceanlinx (formerly Energetech) Uiscebeatha Oscillating Commercial
www.oceanlinx.com Water Column Demonstration
Ocean Power Technologies PowerBuoy Point Absorber Early Commercial
www.oceanpowertechnologies.com
Ocean Wave Energy Company OWEC Point Absorber Laboratory Proof of Concept
www.owec.com
Ocenergy Wave Pump Point Absorber Laboratory Proof of Concept
www.ocenergy.com
OreCON Ltd MRC1000 Floating Pt Abs Technology Demonstration
www.orecon.com & OWC
Oregon State Univ Various direct Point Absorber Technology Demonstration
www.eecs.orst.edu/msrf drive buoys
Pelamis Wave Power Pelamis Attenuator Early Commercial
www.pelamiswavepower.com
Renewable Energy Holdings CETO Point Absorber Technology Demonstration
www.reh-plc.com
Renewable Energy Wave Pump REWP Point Absorber Experimental
www.renewableenergypump.com
Seabased AB Direct Driven Point Absorber Laboratory Proof of Concept
www.seabased.com Linear Gen
Seapower Floating Wave Point Absorber Experimental
www.seapower.com Pres. Vessel
SyncWave Energy SyncWave Point Absorber Technology Demonstration
www.syncwaveenergy.com
Surf Buoy SurfBuoy Point Absorber Experimental
www.cosmotheist.com/Surfbuoy.htm
Trident Direct Energy Point Absorber Laboratory Proof of Concept
www.tridentenergy.co.uk/index.php Conversion
Versabuoy Int’l VersaBuoy Point Absorber Experimental
www.vbuoy.com
Waveberg Water Pump Point Absorber Experimental
Wavebob Ltd Wavebob Point Absorber Technology

www.wavebob.com Demonstration
9-30
10581090
Ocean Wave Energy

(1) (2) (3)
Wave Dragon ApS Wave Dragon Overtopping Commercial
www.wavedragon.net Demonstration
Wave Energy AS Seawave Slot Overtopping Technology Demonstration
www.waveenergy.no Cone Generator
Wave Gen Offshore OWC Oscillating Commercial demo
www.wavegen.co.uk Water Column w/breakwater system;
floating system status
unknown
Wave Power Plant Sea Gate-1 Point Absorber Experimental
www.wavepowerplant.com
Wave Star Energy Wave Star Point Absorber Technology
www.wavestarenergy.com Demonstration
1. This list excludes individual inventors with conceptual-level-only technology
2. The principle of operation: Point Absorber, Attenuator, Overtopping or Oscillating Water Column
3. The following definition of development status was used
• Experimental – Subscale at sea testing
• Technology Demonstration – Large size engineering prototype at sea testing whose purpose is to test for
function and performance
• Commercial Demonstration – Large size manufacturing prototype at sea testing whose purpose is to test
for commercial viability
• Early Commercial – Offering many units of large size for purposes of generating and selling the electricity
produced
9.5.3 Survival in Storms and Hostile Marine Environments
Today’s wave energy conversion technologies are designed to survive a 100-year wave and are
the result of years of testing, modeling, and development by many developer organizations. Full-
scale prototypes have been deployed, although not continuously, in natural waters since 2004.
There are justified concerns over the survivability of devices. With relatively little real-world
operational experience, developers must prove that their device’s survivability design is
adequate. There will be device failures, but this is to be expected in the prototype stages. The
marine environment is extremely challenging and for devices to operate successfully in it will
require significant investment in ocean engineering. It should be noted, however, that oil and gas
platforms are surviving 50 years or more in similarly hostile marine environments.
9-31
10581090
Ocean Wave Energy
9.5.4 Effect of Wave Power Plants on the Environment
Given proper care in siting, deployment and operations, offshore wave power can be one of the
more environmentally benign electricity generation technologies. Cumulative effects can be
controlled by appropriately limiting deployment at project and regional scales. Early
demonstration and commercial offshore wave power plant projects should include rigorous
monitoring of the environmental effects of the plant and similarly rigorous monitoring of a
nearby undeveloped site in its natural state so that natural variation can be distinguished from
long-term effects of wave energy projects.
9.5.4.1. Report to Congress: Potential Environmental Effects of Marine and Hydrokinetic

Energy Technologies [11]
Section 633(b) of the Energy Independence and Security Act of 2007 (EISA) called for a report
to Congress that addresses the potential environmental impacts of marine and hydrokinetic
energy technologies, options to prevent adverse environmental impacts, the role of monitoring
and adaptive management, and the necessary components of an adaptive management program.
The EISA Report to Congress was prepared based on a review of peer-reviewed literature,
project documents, and U.S. and international environmental assessments of these new
technologies. The information was supplemented by contacts with technology developers;
experts in state resource, regulatory agencies, and non-governmental organizations; and input
and reviews by Federal agencies (NOAA Fisheries, Minerals Management Service (now the
Bureau of Ocean Energy Management, Regulation and Enforcement), U.S. Fish and Wildlife
Service, National Park Service, Bureau of Indian Affairs, FERC).
There are numerous conceptual designs for converting the energy of waves, river and tidal
currents, and ocean temperature differences into electricity. Most of these technologies remain at
the conceptual stage. They have not yet been tested in the field or as prototype, full-scale
devices. Consequently, there have been few studies of their environmental impacts. Most
considerations of the environmental impacts have been in the form of predictive studies and
speculative environmental assessments that have not yet been verified.
The assessments have identified common elements among these technologies that may pose a
risk of adverse environmental effects. Potential impacts include the alteration of currents and
waves; alteration of substrates and sediment transport and deposition; alteration of habitats for
benthic organisms; noise during construction and operation; emission of electromagnetic fields;
toxicity of paints, lubricants, and antifouling coatings; and interference with animal movements
and migrations. Project installation and operation will change the physical environment. Possible
effects on biological resources could include alteration of the behavior of animals, damage and
mortality to individual plants and animals, and potentially larger and longer-term changes to
plant and animal populations and communities. Some effects are expected to be minor, but the
potential significance of many of the environmental issues cannot yet be determined owing to a
lack of experience with operating projects.
9-32
10581090
Ocean Wave Energy
Although there have been few environmental studies of these new concepts, a preliminary
indication of the importance of each of these issues can be gained from published literature
related to other technologies (e.g., noises generated by similar marine construction activities,
EMF emissions from existing submarine cables, and environmental monitoring of active offshore
wind farms). Experience with other, similar activities in freshwater and marine systems can also
suggest effective impact minimization and mitigation measures that can be applied to these new
renewable energy technologies. However, some aspects of the environmental impacts are unique
to the technologies, and will require operational monitoring to determine the magnitude and
significance of the effects. This is particularly true for the cumulative effects of large numbers of
ocean energy devices that commercial-scale projects will comprise.
Impacts to bottom habitats, hydrographic conditions, or animal movements that are

inconsequential given a few units may be significant if large, multi-unit projects exploit large
areas in a near-shore ocean area. For some environmental effects it will be difficult to extrapolate
the effects from small to large numbers of units because of complicated, non-linear interactions
between the placement of the machines and the distribution and movements of aquatic
organisms. Assessment of these cumulative effects will require careful environmental monitoring
as the projects are deployed.
Evaluation of monitoring results might be usefully conducted in an adaptive management

framework. There are numerous state and federal agencies and environmental laws and
regulations that will influence the development of marine and hydrokinetic technologies. Federal
licensing of these renewable energy projects is the responsibility of FERC and BOEMRE. Their
licensing decisions will include input from other federal and state agencies, tribes, environmental
groups, and other stakeholders. After a licensing decision has been made and operation of the
energy project has begun, the identification (and correction) of environmental impacts will
depend on appropriate monitoring.
The ability to modify a project to mitigate unacceptable environmental impacts identified by

operational monitoring might be based on applying adaptive management principles reflected in
the project license conditions. In the context of marine and hydrokinetic energy technologies,
adaptive management is a systematic process by which the potential environmental impacts of
installation and operation could be evaluated against quantified environmental performance goals
during project monitoring. Early information about undesirable outcomes could lead to the
implementation of additional minimization or mitigation actions which are subsequently re-
evaluated. An adaptive management process is particularly valuable in the early stages of
technology development, when many of the potential environmental effects are unknown for
individual units, let alone the eventual build out of large numbers of units. Basing the
environmental monitoring programs on adaptive management principles, as advocated by many
resource and regulatory agencies, will take advantage of ongoing research and monitoring to help
refine technology designs and to improve environmental acceptability of future installations.
9-33
10581090
Ocean Wave Energy
9.5.5 Permits for Offshore Wave Power Plants
As of October 29, 2010, there were eight active FERC-issued Preliminary Permits, and one
Preliminary Permit was pending (Table 9-2). A Preliminary Permit gives the permit holder the
first right of refusal to a site for a three-year period to study the site and file a construction
License Application. Further information can be obtained from the FERC website:
http://www.ferc.gov/industries/hydropower/indus-act/hydrokinetics/permits.asp.
Table 9-2
FERC Active and Pending Preliminary Permits
Docket Capacity Filing Issued/

Project/Permit Name State Issue Date
No. (MW) Date Pending
P-12749 Coos Bay OPT Wave Park 100 OR 8/10/2010 Issued
P-13376 Del Mar Landing 5 CA 7/9/2009 Issued
P-13377 Fort Ross (South) 5 CA 7/9/2009 Issued
P-13378 Fort Ross (North) 5 CA 7/9/2009 Issued
P13498 SWAVE Catalina Green Wave 6 CA 9/15/2009 Issued
P-13521 Oceanlinx Maui 2.7 HI 11/25/2009 Issued
P-13641 Central Coast WaveConnect 100 CA 4/28/2010 Issued
Reedsport OPT Wave Park, 2/1/201
P-13666 48.5 OR Pending
Phase III 0
San Onofre OWEC Electricity 3/2/201
P-13679 8.9 CA 10/29/2010 Issued
Farm 0
9.5.6 Overview of Regulatory Status for Offshore Wave Power Plants
Agreements between FERC and the Mineral Management Service (now BOEMRE) in early
2009 have produced the permitting, licensing, and leasing framework depicted in Table 9-3.
Table 9-3
FERC, MMS, and State Lands Permitting, Licensing and Leasing Framework
State Seabed Lands (1) Federal Seabed Lands (2)
Preliminary Permits FERC None
Pilot Licenses FERC FERC
Construction and FERC FERC

Operation Licenses
Leases State Department BOEMRE

(1) To 3 nautical miles except in Texas and Gulf of Mexico states where state lands extend to 12 nautical miles.
(2) On the Outer Continental Shelf (OCS)
9-34
10581090
Ocean Wave Energy
The regulatory permitting, licensing, and leasing processes associated with U.S. ocean wave
projects can be quite involved, complex, lengthy, and costly. Indeed, regulatory issues represent
a significant barrier to ocean wave energy development in the United States.
Since 1920, construction and operation of a non-federal hydroelectric project in the United States
result of legal interpretation from FERC, wave energy hydroelectric projects were placed under
FERC licensing jurisdiction.
Meanwhile, the 2005 Energy Policy Act (EPACT05) gave jurisdiction for leasing on the Outer
Continental Shelf to the Department of Interior’s Mineral Management Service (now BOEMRE).
In addition to FERC and BOEMRE, approvals to install and operate a pilot or commercial
project are still required from many other federal, state, and local regulatory agencies (upwards
of 20 different agencies).
9.5.7 WEC Power Plant Footprints
Use of sea space by wave energy power plants is of critical concern to many environmental
advocacy and ocean area user groups. The oceans are held in trust by the federal and state
governments for the good of society as a whole, and these areas are currently used for multiple
purposes (i.e., commercial fishing, recreation, commercial shipping, dump sites, military
training, etc.).
The footprints of some of the existing technologies for a 10-MW and a 100-MW wave power
plant are shown in Table 9-4. The Orecon machine, the only tension-moored device in the table,
will have the smallest footprint. Slackly moored devices such as the Pelamis and the PowerBuoy
will require larger footprints. The Oyster is a near-shore device that operates only in the surge
zone.
9-35
10581090
Ocean Wave Energy
Table 9-4
WEC Device Areal Footprints
Absorber Dimensions Footprint Single Unit
Length (m) Width (m) Length (m) Width (m)
Pelamis P1 (1) 123 4.6 300 150
OPT 500 PowerBuoy (2) 18 18 100 100
AquaMarine Oyster (3) 12 18 30 30
Orecon MRC (4) 30 45 245 130

Farm Arrangement Footprint 10 MW
# Devices # Rows Length (km) Width (km)
Pelamis P1 (1) 19 2 1.0 0.90
OPT 500 PowerBuoy (2) 20 3 0.70 0.30
AquaMarine Oyster (3) 33 1 1.0 0.03
Orecon MRC (4) 6 1 1.47 0.13

Farm Arrangement Footprint 100 MW
# Devices # Rows Length (km) Width (km)
Pelamis P1 (1) 210 4 10 1.80
OPT PowerBuoy 500 (2) 200 3 8 0.30
AquaMarine Oyster (3) 333 1 10 0.03
Orecon MRC (4) 68 4 4.16 0.52

1. Based on preliminary design performed by EPRI for Oregon Pelamis Wave Power Plant [12].
2. Based on March 2008 Coos Bay Preliminary Application Document filed with FERC. Project size estimates
based on 100-m lateral spacing and 100 m between rows for PB500 PowerBuoys. Overall 100-MW project size
includes three transit lanes. Each lane is 400 m in length.
3. Based on Aquamarine Power website, 300-kW machine located at 15-m depth and 12 X 18-m size (EPRI
assumed 12-m lateral spacing)
4. Based on input provided by Orecon MRC rated at 1.5 MW.
9-36
10581090
Ocean Wave Energy
9.6 Design, Performance, Cost, and Economic Feasibility Issues
9.6.1 WEC Sites
EPRI has investigated the attributes required for a good wave energy site. EPRI site assessments
are documented for Hawaii [13], Washington [14], Oregon [15], and Maine [16]. These and
other reports are available from the EPRI’s Ocean Energy website at
http://oceanenergy.epri.com/waveenergy.html#reports.
There are many factors to consider when evaluating potential sites for a wave energy plant. The
primary factors are:
• Favorable wave energy climate
• A nearby harbor with sufficient depth, size, and port infrastructure to support the assembly
and deployment of the plant and maintenance operations
• A transmission and distribution system that can flow the power from the wave plant into the
grid and a substation interconnection point close to shore
• An existing easement for the submerged cable from the wave power plant to shore and
possibly under the beach to the substation, similar to the easement required for an outflow
pipe
• Bathymetry that provides a depth of about 50 m within two to three miles of shore;
• A sandy seabed (for anchor placement) and a sandy route to shore for trenching the
submerged cable
• Minimum conflicts of sea space use (e.g., fishing, crabbing, whale migration, etc.)
• Local labor to be trained for employment in this new industry.
9.6.2 WEC Design, Performance and Cost
In 2004, EPRI performed an Offshore Wave Power Feasibility Definition Study examining five
locations and two WEC technologies. Offshore Wave Power Plant Feasibility Reports have been
published for sites in Hawaii [17], Oregon [12], San Francisco [18, 19], Massachusetts [20], and
Maine [21]. Performance estimates were developed using local wave data obtained from
measurement buoys and performance data supplied by the manufacturer. Two manufacturers
provided sufficient information such that EPRI could perform a design and performance
analysis: Pelamis WavePower and Oceanlinx. Cost estimates were developed by creating a
detailed breakdown of the various cost centers and outlines of installation and operation
procedures, and by cross-checking them with a variety of sources, including local operators, the
design team, local manufacturers, and similar offshore projects in the oil and gas and offshore
wind industries. Table 9-5 shows the performance and cost estimates for a single-unit Pelamis
pilot plant and a 300,000 MWh/yr commercial-scale plant, one located on the East Coast at
Wellfleet, Massachusetts, and the other on the West Coast at Reedsport, Oregon.
9-37
10581090
Ocean Wave Energy
Table 9-5
Cost and Performance Estimates for Linear Absorber (Pelamis) Wave Power Plants
(December 2009 dollars2).
0.75 MW 103 MW 0.75 MW 90 MW
Rated Capacity and Site
Wellfleet MA Wellfleet MA Reedsport OR Reedsport OR
Plant Size (number of units x unit size, MW) 1 X 0.75 206 X 0.5 1 X 0.75 180 x 0.5
Annual Electrical Energy at Busbar (MWeh) 964 300,000 1,000 300,000
Plant Nameplate Rating (MW) 0.75 103 0.75 90
Physical Plant
Seabed, km^2 <0.5 18.5 <0.5 16.2
Scheduling
Development time, Months Note 1 Note 1 Note 1 Note 1
Construction time, Months 12 24 12 24
Capital Cost ($/kW)
On shore transmission and grid
interconnection $1,050 $66 $874 $32
Subsea cables $1,528 $53 $452 $23
Mooring $367 $264 $367 $264
Power Conversion Modules $2,316 $1,412 $2,316 $1,412
Structural sections $1,282 $555 $1,282 $555
Facilities $0 $132 $0 $150
Installation $954 $133 $10,653 $143
Construction Mgmt and Commissioning $755 $125 $633 $123
Contingencies Note 3 Note 3 Note 3 Note 3
Less State Renewable Inv Tax Credit (4 & 5) ($91) ($13) ($1,663) ($126)
Total Plant Cost (TPC) $8,159 $2,728 $5,318 $2,577
AFUDC (interest during construction) $693 $330 $452 $254
Total Plant Investment $8,852 $3,061 $5,770 $2,830
Owner Costs
Due Diligence, Engineering, Permitting,
Legal, Financial fees, etc $408 $136 $266 $129
Total Capital Requirements $9,260 $3,197 $6,035 $2,959
Yearly O&M Costs (% of TPC per year) N/A7 $0.53 N/A7 $0.48
10 Year One Time Retrofit Costs (% of TPC) N/A7 $1.07 N/A7 $1.03
Unit Availability (%) 85 95 85 95
Technology Development Rating Pre- Pre- Pre- Pre-
Commercial Commercial Commercial Commercial
1. Development time for permitting is an unknown at this early point with emerging ocean energy technology
2. The costs are in November 2004 dollars in References 15 through 19 and were adjusted with a 2.5 % inflation rate to Dec 2009
dollars
3. Contingency costs are built into each of the subsystems
4. The Oregon credit is 25% of the project cost up to a maximum of $10 million
5. The Massachusetts credit is 9.5% of the installation cost
6. Permitting cost is assumed to be 5% of Total Plant Cost; Permitting cost is an unknown at this early point with emerging ocean
energy technology
7. O&M costs for a pilot plant cannot be estimated
9-38
10581090
Ocean Wave Energy
9.6.4 WEC Economic Feasibility
The costs and cost of electricity (COE) estimates made by EPRI were for the first commercial-
scale (100-MW) wave plant. It is an established fact that learning through production experience
reduces costs. Cost reduction follows a logarithmic relationship such that for every doubling of
the cumulative production of new wave power units, a specific percentage reduction in
production costs would be expected. The specific percentage used in this study was 82%, which
is consistent with documented experience in the wind energy, photovoltaic, shipbuilding, and
offshore oil and gas industries.
As occurred with computers, flat-screen televisions, and wind turbines and PV panels, the costs
of WEC devices will decline as the industry moves toward larger-scale manufacturing and higher
cumulative production. The top line in Figure 9-22 is the industry-documented wind energy
learning curve. This curve was developed by EPRI based on data from a multitude of sources.
Figure 9-22 also shows the lower and higher bounds of the wave energy cost estimates as well.
The 82% learning curve is applied to the wave power plant installed cost but not to the
operations and management (O&M) component of the cost of electricity, which why the three
curves are not parallel.
Wave Low Bound Wave Upper Bound Wind
100.00
Actual
COE (cents/kWh)
Wind
COE
History
10.00
Projected Wave
Upper and Lower
COE
1.00
100 1000 10000 100000
Figure 9-22
Levelized COE Comparison to Wind: Oregon Example with Federal and State Financial
Incentives
The adoption of ocean power technologies will be based upon the value of electricity these
devices generate and supply to the grid as well as their ease of integration with the grid. In
addition to the capital cost of installation, O&M will make a significant contribution to the total
9-39
10581090
Ocean Wave Energy
cost. O&M costs related to unplanned maintenance is a major factor in the overall cost of
electricity.
EPRI’s economic assessments have been based on a wave power plant book lifetime of 20 years.
The cost flow profile for wave energy, much like many renewable energy technologies, is
heavily front-loaded; 90% or more of the COE from wave power plants will derive from initial
capital costs and installation costs. Conversely, typical fossil fuel power plants experience fuel
and ongoing operations that are about 80% of the plant's cost of electricity.
Levelized COE takes into account all fixed and recurring costs of a wave power plant as a
function of the electrical energy it generates. The costs, annual energy produced, and financial
assumptions upon which the Figure 9-22 estimates are based are documented in [12].
While there are some federal and state tax incentives to build renewable power systems, these
incentives are insufficient to finance early-adopter, small-scale projects. As a result, developers
will be put in the position of having to push for large commercial installations to drive cost
down, and in the process may be forced to assume the significant technical, economic and
environmental risks of deploying unproven technologies at large scale.
Dozens of institutional investors in U.S. renewable energy projects pulled out of the market
when the nation’s liquidity dried up in 2008. Some found more lucrative investments elsewhere
while others found themselves unable to take advantage of tax credits because they lacked the
profits to do so. The American Recovery and Reinvestment Act (ARRA) of 2009, approved
February 19, 2009, changed the investor ground rules; however, construction delays engendered
by permitting requirements precluded ocean energy projects from exploiting those funds.
Large insurance companies and investment banks that engage project developers provide the
majority of the renewable energy project financing.
Installed offshore wave capacity as of July 2010 was less than 0.1 MW in the United States and
less than 1 MW worldwide. The first shore-based grid-connected wave power unit was deployed
in Scotland in July 2000 and has since operated successfully. The first offshore grid-connected
wave power unit deployed was the one quarter-scale WaveDragon in Denmark in 2003, closely
followed by a full-scale Pelamis at the European Marine Energy Center (EMEC) in the Orkneys
in July 2004. Based on the successful testing of the one-quarter-scale WaveDragon, the company
moved to Wales with the expectation of funding to build a full-scale prototype. Pelamis
WavePower announced the first commercial sale of offshore wave power in May 2005 to Enersis
of Portugal.
Provided regulatory barriers are overcome and government support and incentives are provided,
it is anticipated that wave energy will grow rapidly and reach 10,000 MW rated capacity by
2025.
9-40
10581090
Ocean Wave Energy
9.8 R&D Needs
Hydropower Technologies Program, EPRI sponsored a waterpower workshop for the water
power industry in October 2008 [22]. The purpose of the workshop was to identify and prioritize
research, development, deployment and demonstration (RDD&D) needs which will further the
deployment of conventional hydro/pumped storage and emerging wave and marine hydrokinetic
(MHK) technologies. The priorities identified may be used to shape both EPRI’s and DOE’s
research agendas and will support R&D initiatives throughout the public and private sectors.
The U.S. RDD&D Needs Workshop used the 12 topics from the UK Marine Energy technology
Roadmap [23] as the starting point for developing the U.S. technology needs. These 12 are:
1. Resource Modeling
2. Device Modeling
3. Experimental Testing
4. Moorings and Seabed Attachments
5. Electrical Infrastructure
6. Power Take Off and Control
7. Engineering Design
8. Lifecycle and Manufacturing
9. Installation, O&M
10. Environmental
11. Standards
12. System Simulation.
Four other topics were identified by the Steering Committee prior to the EPRI workshop and two
other topics were identified by the participants during the workshop:
1. Materials–low cost, corrosion and biofouling
2. Storage
3. System configuration evaluations
4. Vision, Goals, Objectives and Roadmap
5. Master Generation and Transmission Plan
6. Education.

1. Testing, including experimental through pilot demonstration
2. Environmental, which will require device testing and deployed projects
3. Standards.
9-41
10581090
Ocean Wave Energy
Once funding is available, specific programs and projects for high-priority topics identified in
this workshop should be developed and implemented.
The RDD&D topics of "Environmental," "Standards," and "Testing" need to be addressed in a

comprehensive fashion. Also, these topics may warrant further consideration in follow-on
meetings to be explored in more detail with knowledgeable experts in the field.
Topics falling below the high-priority ranking, though less urgent, are still important and need to
be addressed.
9.9 Conclusions
Considerable potential exists for generating electrical power from wave energy off the coast of
the United States and many other places in the world. Wave energy climates are the most
favorable for west-facing coasts between 35 and 55 degrees latitude in the northern and southern
hemispheres. In the United States, the prime locations (i.e., those with a good wave climate, port
infrastructure, and/or coastal grid infrastructure) are:
• Northern California and Hawaii, which have excellent wave energy climates, good coastal
grid infrastructures, good ports, and high electricity prices.
• Oregon, which has excellent wave energy climate, good coastal grid infrastructure, good
ports, but low electricity prices.
• Washington, which has excellent wave energy climate, poor coastal grid infrastructure, good
ports, but low electricity prices. The system load is in the Seattle area and there is no
transmission infrastructure to transmit power across the Olympic peninsula.
• Alaska, which has excellent wave energy climate, poor coastal grid infrastructure, good
ports, and high electricity prices. The system load is relatively small in the Anchorage,
Fairbanks, and Juneau areas and there is no transmission infrastructure to move the power to
those areas.
The potential recoverable electricity from wave energy resources is estimated by EPRI to be
about 6.5% of today’s electricity consumption in the United States. Initial studies suggest that
given sufficient deployment scale, these technologies will be commercially competitive with
other forms of renewable power generation. However, significant technical, economic,
operational, environmental and regulatory barriers remain to be addressed in order to allow this
emerging industry to move progress to commercial development.
The experience related to ocean energy is limited to a few prototype installations that provide a
limited understanding of economic, operational, environmental, and regulatory issues. It will be
critically important to gain a full understanding of all life-cycle-related issues over the coming
years to pave the way for larger-scale commercial deployments and a successful wave energy
industry. Such understanding can only be gained from demonstration and commercial systems
deployed by early adopters. First-generation commercial systems will not only address
technology-related issues, but will also provide confidence to regulators, the general public and
investors. Both market push (R&D) and market pull (economic incentives) mechanisms will be
required to successfully move this technology sector forward and develop the capacity to harness
wave energy from the ocean.
9-42
10581090
Ocean Wave Energy
9.10 Resources and References
EPRI: http://oceanenergy.epri.com
European Wave Energy Thematic Network: www.wave-energy.com
Department of Transportation and Industry (UK): www.dti.gov.uk/renewable
Australian Renewables including Wave Energy: www.greenhouse.gov.au/renewable/index.html
Danish Wave Energy: www.waveenergy.dk
European Wave Energy Research Network (EWERN):www.ucc.ie/ucc/research/hmrc/ewern.htm
European Wave Energy Thematic Network: www.wave-energy.net
World Wave Atlas: www.oceanor.no/projects/wave_energy
World Energy: www.worldenergy.org/wec-geis/publications/reports/ser/wave/wave.asp
9.10.2 References
1. Zurkinden, A.S., F. Campanile, and L. Martinelli. 2007. Wave energy converter through
piezoelectric polymers. COMSOL Users Conference. Genoble, France.
2. OWET (Oregon Wave Energy Trust). 2009. Utility Market Initiative: Integrating Oregon
Wave Energy into the Northwest Power Grid. Prepared by Pacific Energy Ventures on behalf
of the Oregon Wave Energy Trust. December 2009.
3. OWET (Oregon Wave Energy Trust). 2009. Wave Energy Infrastructure Assessment in
Oregon. Prepared by Advanced Research Corporation. December 1, 2009.
4. Guidelines for Preliminary Estimation of Power Production by Offshore Wave Energy
Conversion Devices, EPRI, Palo Alto, CA: December 2003. E2I EPRI-WP-US-001.
5. Tolman, H.L. 1991. A third-generation model for wind waves in slowly varying, unsteady
and inhomogeneous depths and currents. Journal of Physical Oceanography 21: 782-797.
6. Tolman, H.L. 2009. User manual and system documentation of WAVEWATCH III version
3.14. Environmental Modeling Center, Marine Modeling and Analysis Branch, National
Oceanic and Atmospheric Administration. MMAB Contribution No. 276. May 2009.
7. Feasibility of Using Wavewatch III for Days-Ahead Output Forecasting for Grid Connected
Wave Energy Projects in Washington and Oregon. EPRI, Palo Alto, CA: February 2008.
EPRI-WP-012.
9-43
10581090
Ocean Wave Energy
8. Wave Energy Forecasting Accuracy as a Function of Forecast Time Horizon, EPRI, Palo
Alto, CA: October 2009. EPRI-WP-013.
9. E2I EPRI Assessment: Offshore Wave Energy Conversion Devices, EPRI, Palo Alto, CA:
June 2004. E2I-EPRI-WP-004-US (Revision 1).
10. California Wave Power Demonstration Project: Bridging the Gap Between the Completed
Phase 1 Project Definition Study and the nest Phase -- Phase 2 Detailed Design and
Permitting, EPRI, Palo Alto, CA: December 2007. EPRI-WP-011-CA.
11. Report to Congress on the Potential Environmental Effects of Marine and Hydrokinetic
Energy Technologies. U.S. Department of Energy, Energy Efficiency and Renewable
Energy, Wind and Hydropower Technologies Program, Washington, DC. December 2009.
12. System Level Design, Performance and Costs -- Oregon State Offshore Wave Power Plant,
EPRI, Palo Alto, CA: November 2004. E2I-EPRI Global-WP-006-OR (Revision 1).
13. E2I EPRI Survey and Characterization of Potential Offshore Wave Energy Sites in Hawaii,
EPRI, Palo Alto, CA: June 2004. E2I-EPRI-WP-003-HI (Revision 1).
14. E2I EPRI Survey and Characterization of Potential Offshore Wave Energy Sites in
Washington, EPRI, Palo Alto, CA: May 2004. E2I-EPRI-WP-WA-003.
15. E2I EPRI Survey and Characterization of Potential Offshore Wave Energy Sites in Oregon,
EPRI, Palo Alto, CA: May 2004. E2I-EPRI-WP-OR-003.
16. E2I EPRI Survey and Characterization of Potential Offshore Wave Energy Sites in Maine,
EPRI, Palo Alto, CA: June 2009. E2I-EPRI-WP-003-ME.
17. System Level Design, Performance and Costs—Hawaii State Offshore Wave Power Plant,
EPRI, Palo Alto, CA: January 2005. E2I-EPRI-WP-006-HI.
18. System Level Design, Performance and Costs—San Francisco California Energetech
Offshore Wave Power Plant, EPRI, Palo Alto, CA: December 2004. E2I-EPRI-006B-SF.
19. System Level Design, Performance and Costs for San Francisco California Pelamis Offshore
Wave Power Plant, EPRI, Palo Alto, CA: December 2004. E2I-EPRI Global-006A-SF.
20. System Level Design, Performance and Costs—Massachusetts State Offshore Wave Power
Plant, EPRI, Palo Alto, CA: November 2004. E2I Global-EPRI-WP-006-MA.
21. System Level Design, Performance and Costs—Maine State Offshore Wave Power Plant,
EPRI, Palo Alto, CA: December 2004. E2I-EPRI Global-WP-006-ME.
Marine and Other Hydrokinetic Renewable Energy, EPRI, Palo Alto, CA: December 2008.
23. UKERC (UK Energy Research Centre). 2008. UKERC Marine (Wave and Tidal Current)
Renewable Energy Technology Roadmap. Summary Report.
9-44
10581090
10
RIVER IN-STREAM ENERGY
Installed Capacity About 200 kW worldwide; 130 kW in the United States.

(as of July 2010) Estimated annual incremental U.S. capacity additions:
2012: Hundreds of megawatts are in the FERC preliminary permit process
pipeline; the number of projects which will actually be completed is
unknown.
Estimated cumulative capacity by 2025: 500 MW.
River In-Stream Energy RISEC is an emerging technology. Verdant Power (USA), Ocean
Conversion (RISEC) Renewable Power Corporation (USA), HydroGreen (USA), and New Energy
Technology Readiness (as Corporation (Canada) demonstrated their technologies in small tidal or river
of November 2010) applications in 2010.
Economic Status Results from the first EPRI economic feasibility studies evaluating remote
village applications in Alaska show a simple payback period of 4 to 5 years
for two remote villages (avoided cost of electricity is $0.65/kWh) and 9 to 11
years for one grid-connected village (avoided cost of electricity is
$0.18/kWh).
Environmental Impact In-stream river power projects are likely to be controversial due to
uncertainty regarding environmental impacts.
Given proper care in siting, installation, operation, and decommissioning,
RISEC technology may be one of the more environmentally benign
generation technologies.
Pilot demonstration testing is needed to understand the interactions
used to incorporate new information into decision-making processes that
will address project build out and cumulative effects
Regulatory Status The Federal Energy Regulatory Commission (FERC) has primary
jurisdiction for licensing river in-stream energy under the Federal Power Act
(FPA).
FERC has developed a six-month license application process for pilot
demonstration plants. Alaska Power and Telephone applied for the first pilot
license for its Yukon River at Eagle project. A license at an existing
hydroelectric project (Mississippi Lock and Dam No. 2, Hastings,
Minnesota) has been modified to allow installation of a hydrokinetic array in
the existing project’s tailrace.
There are 114 current, FERC-issued Preliminary Permits for river in-stream
hydrokinetic projects totaling 6,864 MW. Of the 114 active Preliminary
Permits, 12 expire at the end of 2010. There were 10 pending applications
totaling 102 MW as of October 4, 2010.
10-1
10581090
River in-Stream Energy
Government Support of DOE initiated a Waterpower R&D Program in FY 2008 with a

RISEC Technology Congressionally mandated $10 million, which was followed by additional
Congressionally mandated funding of $40 million in FY 2009 and $50
million in FY2010.
Trends to Watch Preliminary Permits give project developers an opportunity to evaluate the
permitted site and prepare a license application without encroachment by
other developers for a period of three years. An important indicator of
industry development will be the number and progress of applications for
FERC-issued pilot licenses and commercial licenses.
10.1 Introduction
The overall hydrokinetic power potential of U.S. rivers with discharge rates greater than 113 m3/s
and velocities greater than 1.3 m/s was estimated by New York University (NYU) in 1986 [1] to
be 12,500 MW, using conservative assumptions of turbine array deployment. The best river in-
stream hydrokinetic resources in the United States are in Alaska and the Pacific Northwest;
however, all states have some hydrokinetic resources assuming conversion technology is
economical at lower power densities than was assumed in the NYU study. Figure 10-1 shows the
major North America rivers and their yearly discharges in km3/year.
Western Hudson Bay

Nelson, Hayes, Churchill,
Kazan, and Thelon Rivers
~300
MacKenzie
~330
Yukon
~200
St. Lawrence
~440
Columbia
~260
Hudson
~19
Sacramento
~40 Mississippi
~600
Colorado
~22
Figure 10-1
Major North America Rivers and their Yearly Discharges in km3/year
10-2
10581090
In North America, the discharge of fresh water is dominated by the Mississippi (Q = 600 km3/yr)
3
and the St. Lawrence (Q = 440 km /yr), both of which receive large inputs from the high-
precipitation region east of the mountain ranges that dominate the western third of the continent.
Other large rivers in North America include the Mackenzie and the Yukon (which discharge into
Arctic marine waters), the Western Hudson Bay Group, the Columbia, the Sacramento, the
Colorado, and the Hudson rivers.
10.1.1 Main Components of River Hydrokinetic Resource
There are four major components of river in-stream or current flow at a given transect of a river:
discharge, velocity, cross-section area and stage. Discharge is the volume of water moving past a
given cross section of river per unit of time. The units generally used to express this
measurement are cubic meters per second (m3/s). Velocity is the speed of water movement in
2
meters per second (m/s). Cross-section area is the water area (in m ) in a plane perpendicular to
the flow at a given point along the river. Stage is the height (in meters) of the water level above
an arbitrary zero point. All of these components are interrelated: when water levels are up, so are
the discharge and speed; when the stage is up, the cross-sectional area increases.
Flow velocity is dependent on the amount of friction between the water and the stream channel.
Smoother channels will have less friction and, therefore, faster flow. Channel roughness
contributes to turbulence, which dissipates energy and reduces flow velocity. The Manning
Equation is used to calculate average flow velocities in open-channel systems:
V=
(
k R 2 / 3 S 1/ 2 )
n
In this equation, V is the average flow velocity in feet per second or m/s; R is the hydraulic radius
in feet or m, and is the ratio of the cross-sectional area of the stream divided by the length (feet
or m) of channel bottom along that cross section (wetted perimeter), S is the slope of the water
surface, k is a conversion constant that is 1.49 for English units and 1 for metric units, and n is
the Manning roughness coefficient. A larger value of the Manning coefficient yields a smaller
value for flow velocity. Typical values of n are shown below.
Table 10-1
Manning Coefficients for Representative River Substrates
Manning
Substrate Type
Coefficient (n)
Mountain streams with rocky beds 0.04-0.05
Winding natural streams with weeds 0.035
Natural streams with little vegetation 0.025
Straight, unlined earthen channels 0.020
Smoothed concrete 0.012
10-3
10581090
Current speed varies spatially within any river or channel. Current speed is influenced by
proximity to obstacles such as rocks and vegetation found along the river-bed, which reduce the
flow rate. Therefore, the current is usually highest in the middle and lowest along the edges and
bottom of the river; however, the current is usually higher at the outside of a river bend than on
the inside of the bend. Fast current may also scour out the outer channel if the substrate is
erodable, thereby deepening the channel there.
Erosion and water velocity have a direct relationship, as erosion and the maximum particle size
transported increases with flow velocity. Thus, substrate particle size is sorted by local water
velocity. Just as speed affects erosion, erosion affects speed. The speed of a river affects the
shape of the river itself. At a high gradient, the speed of a river is at its maximum, therefore the
shape of the river is very narrow—it stretches out as the slope gradient decreases.
10.1.2 The Conversion of River In-Stream Hydrokinetic Energy to Electricity
A river in-stream energy conversion (RISEC) device harnesses the unidirectional flow of a river
or man-made canal. The power of a river current, in watts, is given by:
Pwater = ½ ρ A V
3
where A is the cross-sectional area of flow intercepted by the device (in square meters), ρ is the
water density (in kilograms per cubic meter), and V is the water velocity (meters per second).
The energy recovery efficiency and performance of an in-stream energy conversion device can
be estimated using the simplified model of a generic RISEC device as described below. The
calculation addresses the energy collection and conversion efficiency of each step in the process,
beginning with the kinetic energy of the flowing water stream and proceeding through the
turbine, drive train, generator, and power conditioning steps as illustrated in Figure 10-2.
Water Turbine Drive Gen Power Grid

Kinetic Efficiency Train Eff. Cond. Ready
Energy Eff. Efficienc Electric
Energy
Figure 10-2
Steps Affecting Hydrokinetic Turbine Efficiency
The individual efficiencies (f) of the RISEC components determine the net usable power that can
be delivered to the grid as described in following equation:
Pdelivered = Pwater x fRISEC,
where
fRISEC = fTurbine x fDriveTrain x fGenerator x fPowerConditioning
10-4
10581090
Since all RISEC devices rely on moving water, and the water must retain enough kinetic energy
to exit the turbine, there is a limit to the portion of kinetic energy that can be removed by a
turbine. The limit on energy extraction and dissipation by a turbine immersed in a free
(unducted, unconstrained) stream is known as the Betz limit. The Betz limit of 59% is sometimes
used as a theoretical maximum efficiency of a turbine. Typical values for component efficiencies
are fTurbine = 40%, fDriveTrain = 95%, fGenerator = 95%, and fPowerConditioning = 97%. The resulting overall
efficiency in this example is 35%.
Figure 10-3 is an illustration of a typical open-rotor, horizontal-axis RISEC device. The

hydrokinetic power seen by such a water turbine per unit cross sectional area (in this case, the
area of the rotor blades) is equal to one half the density of water multiplied by the cube of the
water speed. This relationship is shown in Figure 10-4.
Figure 10-3
Example of a Hydrokinetic Turbine
(P/A)flow = 0.5 x density of water x V3
V
A
Figure 10-4
Water Power Density
10-5
10581090
10.2 U.S. In-Stream River Energy Highlights: 2009–2010
Interest in river in-stream renewable energy continues to grow in the United States. This section
presents a brief accounting of notable federal and state developments and device developer
activities within the sector during the second half of 2009 and through October 2010.
10.2.1 Federal and State Project Highlights
10.2.1.1 Federal-Level Projects
In FY 2010, the U.S. Congress appropriated $50 million to conduct research and development on
technologies (wave, tidal, ocean current, in-stream tidal, river hydrokinetic, and ocean thermal),
as well as conventional hydropower that uses a dam or diversionary structure. This represents a
significant increase in funding from the $10 million appropriated in 2008, the first year marine
hydrokinetic power research was supported by DOE.
DOE’s priorities for wave and tidal/river/ocean current hydrokinetic power include:
• Facilitating the deployment of prototypes and collecting data on their energy conversion
performance and their environmental and competing-use impacts;
• Determining the available, extractable and cost-effective resources in the United States;
• Characterizing and comparing the wide variety of existing marine and hydrokinetic
technologies;
costs; and
• Minimizing the cost, time and negative environmental impacts associated with project siting.
The DOE issued a solicitation in April 2009 which resulted in the selection of EPRI in late 2009
to conduct a national river in-stream hydrokinetic energy resource assessment. That assessment
is ongoing and will be concluded in 2011. The DOE issued another solicitation in April 2010 that
focused on advancing marine and hydrokinetic technology readiness. Awards totaling $37
million were announced in September 2010, $22 million of which were related to RISEC
technology development. Matching funds from non-federal sources will double that amount.
10.2.1.2 State-Level Projects
The first RISEC project in the United States was installed in August 2008 in the Yukon River at
Ruby, Alaska (Figure 10-5). A 5-kW EnCurrent turbine from New Energy Corporation of
Canada delivers electricity to a remote village grid. The project was coordinated by the Yukon
River Intertidal Watershed Council, and the design and installation was performed by ABS
Alaskan. This project is ongoing, with annual deployment of the turbine during the open-water
season.
10-6
10581090
Figure 10-5
Hydrokinetic Turbine Deployment from Barge at Ruby, Alaska, on the Yukon River
Another Alaska project was the first to seek a FERC pilot hydrokinetic license. Alaska Power &
Telephone (AP&T) submitted a draft pilot license application along with a notice of intent to test
a 100-kW hydrokinetic system on the Yukon River at Eagle. In 2007, the Denali Commission of
Alaska awarded $1.6 million in grant funds to assist in the development and deployment of the
company’s river hydrokinetic project. A 25-kW EnCurrent turbine, adaptable to a variety of
locations where sufficient tidal or river current flow is available, is being evaluated over a five-
year period.
Figure 10-6
Village of Eagle, Alaska
Unfortunately, in early May 2009, rapidly rising temperatures in excess of 70°F caused a
rapid "melt-out" of ice and snow along the Yukon River, resulting in unprecedented flooding that
nearly wiped out the small community of Eagle. An aerial view (Figure 10-7) shows at least two
dozen buildings submerged in a sea of car-sized ice chunks and 30 feet of muddy floodwater.
10-7
10581090
The Yukon rose 30 feet over its normal level when 4- to 7-foot-thick ice pans surged
downstream, choking the river and bulldozing islands and shorelines. Despite this setback, the
project continues with seasonal deployment of the 25-kW turbine.
Figure 10-7
Village of Eagle during Spring Break-Up of 2009
10.2.1.3 Other Projects
EPRI System Definition and Feasibility Study at Three Sites in Alaska. This study project was
conducted under the sponsorship of the Alaska Energy Authority (AEA), Chugach Electric, and
Anchorage Municipal Light & Power. EPRI first conducted a site survey for six locations in
close proximity to small villages that typically have power needs of a few hundred kilowatts. The
attributes needed for a good RISEC site were characterized. Of particular importance is the river
current velocity profile [2]. From this survey, the Alaskan sponsors selected three sites for
preliminary design of a RISEC power plant: Kvichak River at Igiugig, Eagle on the Yukon
River, and Whitestone on the Tanana River. EPRI then prepared the design and estimated the
cost to construct, deploy, operate and maintain this plant. Lastly, EPRI assessed the technical and
economic feasibility of RISEC technology applied at these sites. The simple payback periods
estimated for the remote villages of Iguigig and Eagle, both of which have an isolated grid, are
three to four years and four to five years, respectively. The simple payback period estimated for
Whitestone, a remote but grid-connected village, is eight to nine years [3].
Alaska Energy Authority Contract Awards. The AEA awarded two contracts to local
universities to characterize the river in-stream resource potential in Alaskan rivers:
• University of Alaska Anchorage (UAA): UAA will conduct resource assessments of
hydrokinetic energy potential in rural Alaska. This project is assessing the potential
hydrokinetic energy resources for many sites in rural Alaska by collecting velocity,
bathymetric, and water surface elevation/slope data. UAA will begin by creating a list of
about 24 sites/communities which appear to have the greatest potential. Alaska's larger rivers
(e.g., the Yukon, Koyukuk, Kuskokwim, and Susitna rivers) are obvious places to evaluate
for in-stream energy. UAA will examine available data to develop a list of river stretches and
10-8
10581090
community partners. The UAA Native Science and Engineering Program (ANSEP) has a
well-established relationship with many communities throughout Alaska and has agreed to
assist the faculty in establishing working relationships with the communities. Following site
selection, UAA will select student research assistants who will be trained in hydrographic
surveying and velocity measurement. Working in cooperation with the communities, UAA
will survey the selected river stretches to obtain bathymetric, current distribution, and water
surface elevation/slope data. UAA will also install water-level gauges at the various study
sites to obtain flow/velocity data through the open-water period. Next, using data from U.S.
Geological Survey (USGS) gauging stations and gauging stations set up at the various village
sites, UAA will estimate the long-term hydrologic (i.e., velocity and depth) conditions at the
selected rural sites.
• University of Alaska Fairbanks (UAF): UAF will deploy measurement devices and
characterize the long-term velocity profile at Nenana on the Tanana River. UAF will perform
environmental effects analysis, analyze the effects of cold temperatures and ice on the water
turbine, and plan a water turbine pilot demonstration project.
First FERC-Licensed Commercial Hydrokinetic Plant Installed in the Continental United

States. Hydro Green Energy installed a hydrokinetic turbine in the tailwater of the U.S. Army
Corps of Engineers Lock and Dam #2 on the Mississippi River at Hastings, Minnesota, in
January 2009. The 3.6-m diameter turbine, suspended under a barge in the tailwater (Figure 10-
8), had a rated power output of 38 kW at 2 m/s. Full commercial power operations commenced
on August 20, 2009 with a 100-kW nameplate-capacity hydrokinetic turbine.
Figure 10-8
Hydro Green’s hydrokinetic turbine and the deployment site at Lock and Dam #2 on the
Mississippi River at Hastings, Minnesota
RISEC Project Feasibility Study on the Detroit River. The Detroit/Wayne County Port
Authority (DWCPA), NextEnergy, the Detroit Riverfront Conservancy, and the University of
Michigan School of Naval Architecture and Marine Engineering are engaged in a collaborative
effort to develop and deploy a RISEC system on the Detroit River. Vortex Hydro Energy LLC, a
company set up by the University of Michigan, is developing the patented RISEC technology,
intended to be used at the DWCPA’s new Public Dock & Terminal project.
10-9
10581090
Free Flow Power Proposal to Place 180,000 Underwater Turbines along 500 Miles of the
Mississippi River. Free Flow Power, a Massachusetts-based company, has secured Preliminary
Permits from FERC for more than 100 proposed sites in seven states. Each turbine would
produce up to 40 kW, with a maximum collective output of 7,200 MW and an average
generation of 1,800 MW. Figure 10-9 shows a sampling of potential Free Flow Energy sites in
Arkansas.
In 2010, Free Flow Power tested interactions between its 3-meter turbine and fish in a flume at
the USGS Conte Anadromous Fish Laboratory in Turners Falls, Mass. (Figure 10-10).
Figure 10-9
Preliminary Permit Sites in Arkansas Held by Free Flow Power
10-10
10581090
Figure 10-10
Pre-test Inspection of Free Flow Power 3-meter Turbine in Flume at the USGS Conte
Anadromous Fish Laboratory
10.3 Canadian River In-Stream Energy Highlights: 2009–2010
In Canada, there is strong interest in both river in-stream renewable energy and tidal in-stream
generation, whereas the rest of the world seems to be primarily interested in large tidal sites. This
section presents a brief account of notable Canadian activities within the sector during the second
half of 2009 through October 2010.
RISEC Project Deployment in the St. Lawrence River at Cornwall, Ontario. Verdant Power
Canada is developing the Cornwall Ontario River Energy (CORE) Project on the St. Lawrence
River (Figures 10-11 and 10-12). The CORE project has received grant support through the
Ontario Ministry of Research and Innovation (MRI) for a three-year kinetic hydropower
demonstration project in the fast-flowing waters of the St. Lawrence, 3.5 km downstream of the
bi-national New York Power Authority-Ontario Power Generation Moses-Saunders hydroelectric
dam. The project, to be deployed in three stages, will primarily test the foundation and
mountings of Verdant’s proprietary Kinetic Hydro Power System (KHPS) turbine for fast-
flowing riverine systems. In addition, the demonstration project will continue to expand the
science and evaluation of RISEC technology impacts on aquatic resources through a testing
program undertaken with the St. Lawrence River Environmental Institute of Science. If
successful, the project could be expanded in phases to 5 MW and potentially to as much as 15
MW in the general vicinity (see www.verdantpower.com for more information). The Phase I
deployment will occur in July 2011.
10-11
10581090
Figure 10-11
Location of Verdant’s Cornwall Ontario Renewable Energy (CORE) project in the St.
Lawrence River.
Figure 10-12
Site of the CORE Project on the St. Lawrence River (left), and Verdant Turbine (right).
10-12
10581090
Deployment on a Microgrid at Fort Simpson. A 25-kW EnCurrent hydrokinetic turbine was

deployed at Fort Simpson on the Mackenzie River, Northwest Territories, Canada. The
microgrid-connected turbine was removed for the winter season and will be reinstalled following
ice-out in the spring of 2011.
10.4 River In-Stream Power and Energy Resources
EPRI has performed in-stream river resource power and energy assessments for six sites in the
state of Alaska. For the Alaska site survey, only USGS gauging station transects were used.
EPRI obtained the hand-written calibration sheets from the USGS Office in Anchorage, Alaska.
What follows is a general description of how the river speed profile was obtained for the Yukon
River site at Eagle, Alaska.
EPRI researched available data from USGS gauging stations and developed a methodology for
estimating in-stream hydrokinetic resources and predicting its performance. USGS calibrates
most of its 420 measurement stations in Alaska six times per year. To do this, the USGS
establishes a cross-sectional profile at the calibration cross-section, measures velocities at
various locations throughout the cross-section, calculates the water discharge volume, and
correlates the discharge or volumetric flow rate with the water level in the river.
The USGS maintains a stream gauging station on the Yukon River at Eagle and has recorded
daily discharge data for 57 years. To develop the river speed profile, a relationship was first
established between discharge rate and velocity. The relationship was then applied to the full
data set to determine the statistical parameters shown below in Figures 10-13 through 10-18.
Note that the velocity profiles and associated power densities are only valid for the transect that
the USGS used to calibrate the flow data and which EPRI used to calibrate the velocity data.
3.00
2.50
2.00
Velocity (m/s)
1.50
1.00
0.50
0.00
0 50 100 150 200 250 300 350 400 450 500
Distance from shore (m)
Figure 10-13
Stream Velocity Profile across the Yukon River at the USGS Gauging Station at Eagle,
Alaska for a Discharge Rate of 183,000ft3/s
10-13
10581090
0.0
-2.0
Water Depth (m)
-4.0
-6.0
-8.0
-10.0
-12.0
0 50 100 150 200 250 300 350 400 450 500
Distance from shore (m)
Figure 10-14
Channel Cross-section at the USGS Gauging Station at Eagle, Alaska, at a Discharge Rate
of 183,000ft3/s
8
y = 1.9989Ln(x) - 17.952
7
5
Velocity (ft/s)
0
0 50000 100000 150000 200000 250000 300000
Discharge (ft^3/s)
Velocity Fitted Log. (Velocity)
Figure 10-15
Velocity Versus Discharge at the USGS Gauging Station at Eagle, Alaska
10-14
10581090
35%
30%
25%
Frequency
20%
15%
10%
5%
0%
05
25
45
65
85
05
25
45
65
85
05
25
45
65
85
0.
0.
0.
0.
0.
1.
1.
1.
1.
1.
2.
2.
2.
2.
2.
Velocity (m/s)
Figure 10-16
Velocity Distribution at the USGS Gauging Station at Eagle, Alaska
250,000
200,000
150,000
ft^3/s
100,000
50,000
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Figure 10-17
Average Discharge by Month at the USGS Gauging Station at Eagle, Alaska
10-15
10581090
9.00
8.00
7.00
6.00
Velocity (ft/s)
5.00 Average Vel ft/s

4.00 Max Vel ft/s
3.00 Min Vel ft/s
2.00
1.00
0.00
-1.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Figure 10-18
Average Velocity by Month at the USGS Gauging Station at Eagle, Alaska
10.5 River In-Stream Energy Conversion (RISEC) Development

A RISEC device is similar to a tidal device except that it does not need to accommodate bi-
directional flow nor is it exposed to a salt water environment.
10.5.1 Harnessing In-Stream River Energy
In-stream river energy extraction is complex and many device designs have been proposed. It is
helpful to describe their physical arrangements and energy conversion mechanisms, which may
be classified as one of three types:
• Axial flow: the axis of rotation is parallel to the direction of water flow.
• Cross flow: the axis of rotation is perpendicular to the water stream and may be oriented at
any angle, from horizontal to vertical, with respect to the water surface.
• Non-turbine: oscillatory hydrofoil, vortex induced motion, or hydro Venturi device.
Figure 10-19 illustrates axial and cross-flow types of turbines.
10-16
10581090
Axial and Cross-flow Turbines Marine Current Turbines Axial Turbine

(Courtesy Marine Current Turbines)
Lucid Energy Cross-flow Vertical Turbine ORPC Cross-flow Horizontal Turbine

(Courtesy Lucid Energy) (Courtesy ORPC)
Figure 10-19
River In-Stream Energy Conversion Devices
The subsystems for a hydrokinetic turbine typically include a blade or rotor, which converts the
energy in the water to rotational shaft energy; a drive train, which usually comprises a gearbox
and a generator; a support structure for the rotor and drive train; and other equipment, including
controls, electrical cables, and interconnection equipment. Additional ways of classifying these
devices include:
• Support structures: These may be either gravity foundation based, attached to a monopile
foundation, or anchored and moored and allowed to “fly” in the stream using buoyancy
and/or dynamic pressure forces. RISEC devices may also be hung from a floating platform
that is anchored and moored to maintain its position in the flow.
• Open vs. ducted: Turbine rotors may be either open, much like wind turbines, or be enclosed
in a duct or shroud. Since the energy is a function of the cross-sectional area of the flow
directed through the swept area of the turbine, using a duct of a given cross-sectional area is
the equivalent of using an open rotor with the same cross-sectional area. The wind energy
industry has found that adding length to a rotor blade is more economical than adding a duct
to increase the cross-sectional area and power of a wind machine. However, the economics
may differ for water turbines.
10-17
10581090
• Fixed vs. variable pitch blades: Pitch control is used to limit power, maximize the efficiency
of the turbine in variable flows, and enable bi-directional operation. There are other ways to
accomplish these three functions with fixed blade turbines and many different design
concepts for implementing them.
• Closed-center vs. open-center hubs: Instead of a fixed hub and rotating blades, a design
variation uses an outer fixed rim and an inner rotating bladed disc. The potential benefit of an
open-center design is the elimination of the need for a gearbox by encapsulating the stator of
a generator on the rim of the machine.
• Savonius vs. Darrieus vertical-axis turbines: First invented in Finland, the Savonius turbine is
S-shaped when viewed from above. This drag-type turbine turns relatively slowly, but yields
a high torque. The Darrieus turbine was invented in France in the 1920s. Often described as
looking like an eggbeater, this turbine has vertical blades that rotate into and out of the flow.
Using hydrodynamic lift, these turbines can capture more energy than drag devices.
• Helical vs. cycloidal: Aerodynamic lift-type vertical-axis turbine blade configuration. A
helical blade traces a three-dimensional curve that lies on a cylinder or cone, such that its
angle to a plane perpendicular to the axis is constant. A cycloidal blade has the shape of a
curve traced by a point on the circumference of a circle that rolls on a straight line.
Non-turbine types are also being investigated, such as hydro Venturi and hydrofoil or vortex
induced oscillation. A few examples are described in the following subsection.
10.5.2 RISEC Technology Developers
Today, a number of entrepreneurial companies are leading the commercialization of in-stream

river energy conversion technologies. Any tidal in-stream energy conversion (TISEC)
technology can be applied to river in-stream energy conversion (RISEC). Table 10-2 lists known
RISEC/TISEC developers that are developing in-stream river applications as of October 2010.
10-18
10581090
Table 10-2
In-Stream River Energy Conversion Device Developers
Device Developer 1 Development

Device Name Type
Website Status 2
AeroHydro Research and Technology Unknown Oscillatory Laboratory
www.ahrta.com
Free Flow Power FFP Turbine Axial Horizontal Experimental
www.freeflowpower.com Generator Axis
Free Flow 69 Osprey Cross Flow Experimental
www.hi-spec.uk.co/page10.htm Vertical Axis
Lucid Energy Gorlov Helical Cross Flow Technology
www.licidenergy.com Turbine (GHT) Vertical Axis Demonstration
Hydro Green Energy Krouse Turbine Axial Horizontal Commercial
www.hgenergy.com Axis Demonstration
New Energy Corporation EnCurrent Turbine Cross Flow Commercial
www.newenergycorp.ca Vertical Axis Demonstration
Ocean Renewable Power Corp OCGen Cross Flow Commercial
www.oceanrenewablepower.com Horizontal Axis Demonstration
UEK Underwater Electric Axial Horizontal Commercial
www.uekus.com Kite Axis Demonstration
Verdant Power Free Flow Turbine Axial Horizontal Commercial
www.verdantpower.com Axis Demonstration
Vortex Hydro VIVACI Oscillatory Laboratory
www.vortexhhydro.com
1. This list excludes individual inventors with conceptual-level technology..

2. The following definitions of development status were used:
• Experimental: Sub-scale at sea testing
• Technology Demonstration: Large engineering prototype at sea testing whose purpose is to test for
function and performance
• Commercial Demonstration: Large manufacturing prototype at sea testing whose purpose is to test for
commercial viability
• Early Commercial: Offering many large units for purposes of generating and selling the electricity
produced.
10.5.2.1 Aero Hydro Research and Technology Associates
Aero Hydro Research and Technology Associates (AHRTA) is developing a hydropower

generator that uses oscillating wings to convert the flow energy of rivers and tidal streams into
electrical energy. AHRTA’s concept manipulates an airfoil to oscillate in both plunge (pure
translation) and pitch (rotation about some axis on the airfoil chord line) to extract energy from
air or water flow. The phase angle between the pitch and plunge oscillations must be close to 90
degrees. AHRTA has constructed an experimental model which enforces the plunge and pitch
oscillation with the proper phasing between the two motions, as shown in Figure 10-20. The
10-19
10581090
model has two wings arranged in a tandem configuration so that the two wings also operate with
a 90-degree phasing. Thus far, the model system has operated satisfactorily. AHRTA is now in
the process of developing a new model with a simpler mechanism to enforce the phasing
between the pitch and plunge motion.
Figure 10-20
Experimental configuration of the AHRTA Oscillating Turbine
10.5.2.2 Free Flow Power
Free Flow Power (FFP) is developing a RISEC turbine system that uses a rim-mounted,
permanent-magnet, direct-drive generator to maximize efficiency, with front and rear diffusers
and one moving part, the rotor. The generator uses a start-up bearing and a combination of
magnetic levitation and hydrodynamic bearings. At a flow of 9 feet per second (2.7 m/s), the
turbine can produce 20 kW.
Magnetic arrays, using rare-earth neodymium magnets, provide high field strength for greater
efficiency and lower harmonic content. This arrangement facilitates easier grid synchronization
than traditional bipolar magnet arrays. Meanwhile, the generator’s rotor is designed to operate
over a wider range of flow speeds (2 to 5 m/s). Figure 10-21 illustrates both the cross section and
component parts of the FFP turbine generator and an experimental rotor.
FFP is designing a prototype turbine in collaboration with Sigma Design Co. of Springfield, N.J.,
Advanced Energy Conversion of Malta, N.Y., and Turbo Solutions Engineering of Norwich,
Vermont. Looking ahead, the company plans to place six to 12 turbines in arrays on pilings,
25 feet off the bottom of a river and at least 40 feet below the surface to stay clear of ships and
boats.
10-20
10581090
Figure 10-21
Free Flow Power Turbine Generator
Free Flow Power expects to manufacture two versions of the Free Flow Turbine Generator: a 2-
meter turbine designed to generate 10 kW at a flow velocity of 2 m/s, and a 1-meter turbine
designed to generate 10 kW at a flow velocity of 3 m/s.
10.5.2.3 Free Flow 69
Free Flow 69, founded in 2005, is researching a tidal power concept called “the sea engine,”
invented in 1988. It has developed a vertical-axis turbine called the Osprey. Although the design
of the turbine is still confidential, the developer says that the key advantages of its turbine
include suitability for both river and tidal streams, high efficiency over a range of stream flows,
relatively simple design and manufacture, and easy maintenance, since most of the complex
components are above water level.
Figure 10-22 shows the Osprey Prototype Turbine test rig, a 30-foot aluminum catamaran
manufactured by Able Engineering. Initial pilot trials are now being conducted.
Figure 10-22
Free Flow 60 Osprey Turbine in Experimental Test Configuration
10-21
10581090
10.5.2.4 Hydro Green
Houston-based Hydro Green Energy, LLC has developed and patented a hydrokinetic turbine
array (HTA) system. The company intends to operate as an Independent Power Producer (IPP),
selling the power generated from its HTAs via long-term, wholesale power purchase agreements
(PPAs). Figure 10-23 illustrates a 2x2 hydrokinetic HTA configuration, and Figure 10-24 is an
underwater view of the patented hydrokinetic in-stream river current device array configuration.
Figure 10-23
Hydro Green Turbines
Figure 10-24
Hydro Green Turbine Array Configuration
10-22
10581090
10.5.2.5 Lucid Energy Technologies
Formed in March 2007, Lucid Energy Technologies is a joint venture between GCK
Technology, Inc. and Vigor Clean Tech, Inc. The company is focusing on designing and
commercializing complete hydrokinetic electricity generation systems based on the Gorlov
Helical Turbine (GHT). Figures 10-25 and 10-26 show a Lucid Energy turbine prototype and an
array configuration, respectively.
Figure 10-25
Lucid Energy Gorlov Helical Turbine
Figure 10-26
Lucid Energy Array Configuration
10-23
10581090
10.5.2.6 New Energy Corp.
New Energy Corp. is the Canadian manufacturer of the proprietary EnCurrent turbine, whose
design is based on the Darrieus wind turbine. When the turbine rotor is placed within a water
current, the hydrofoils generate a lift vector in the forward direction and the energy is captured as
shaft rotation. The hydrofoils experience their maximum forward torque at the top and bottom of
their rotation, when the water moving past them is tangential. The turbine rotates in the same
direction regardless of the direction of the water current and captures between 35% and 40% of
the energy in moving water. It rotates at a very low speed, with the blade speed between 2 and
2.5 times the speed of the water passing through the device.
One of the unique properties of the Darrieus turbine design is that it is able to capture the energy
from the water irrespective of the direction of the current. This feature enables the EnCurrent
turbine to harness the energy of both flood and ebb tides. A permanent magnet generator is
mounted on the turbine shaft to convert the torque generated by the rotor into electricity. The
output from the permanent magnet generator is a variable-frequency AC current which is
rectified to DC and then converted to constant-frequency AC by an inverter.
New Energy currently manufactures 5-, 10-, and 25-kW models of the EnCurrent Power
Generation System, and is developing 125- and 250-kW models. New Energy also provides a set
of ancillary products that support the installation of the EnCurrent Power Generation System.
Figure 10-27 illustrates such a device mounted on a double-hull pontoon boat, and Figure 10-28
illustrates the turbine design. The generator is mounted on the shaft above the water surface.
Figure 10-27
New Energy System Concept
10-24
10581090
Figure 10-28
New Energy EnCurrent Turbine
10.5.2.7 Ocean Renewable Power Corp.
Founded in 2004, Ocean Renewable Power Co. has developed a proprietary RISEC turbine
called the Ocean Current Generation Turbine Generating Unit (OCGen TGU). The TGU turbine
rotates in one direction regardless of the stream flow direction. Two cross-flow turbines drive a
permanent magnet generator on a single shaft. TGUs are “stacked” (horizontally or vertically)
and combined into OCGen modules that contain the ballast/buoyancy tanks and power
electronics/control system (Figure 10-29).
Assembled OCGen modules are deployed in arrays consisting of tens to hundreds of modules
and are held in position underwater by deep-sea mooring systems. A power and control cable
connects each OCGen module to an underwater transmission line, which transmits the energy to
an on-shore substation. Power output up to 250 kW is achievable in a 6-knot current.
Figure 10-29
Ocean Renewable Power Corp OCGen Module
10-25
10581090
In mid-May 2007, ORPC initiated an OCGen TGU demonstration project in tidal currents
in the Western Passage (Passamaquoddy Bay) near Eastport, Maine. The demo was completed in
early 2008, and it successfully proved the basic design and technical feasibility of the TGU. Data
were also collected for use in the subsequent TGU commercial designs. ORPC has continued to
test improved designs at Eastport and was the recipient of a $10 million grant from DOE to
facilitate commercialization of its technology.
10.5.2.8 UEK Corp.
UEK Corp. was founded in 1981 by Philippe Vauthier, an early visionary and pioneer in TISEC
technology. Since his death in 2008, the company has been led by his widow Denise Vauthier.
“UEK” stands for “Underwater Electric Kite,” Vauthier’s description of a system that consists of
two joined, free-flow turbines tethered to the river or sea bed.
Figure 10-30
UEK Prototype Demonstrated in 2000 (Source: UEK)
10.5.2.9 Verdant Power
New York-based Verdant Power, founded in 2000, has fabricated, tested, and deployed four
working marine energy system prototypes. Dubbed the Free Flow, the system consists of arrays
of three-bladed, horizontal-axis turbines that resemble and operate like wind turbines (Figures
10-31 and 10-32). The turbine rotor spins slowly and steadily at approximately 32 rpm, and is
driven by the natural currents of tides and rivers. The rotor drives a gearbox to increase rotational
speed and power the grid-connected generator. The gearbox and generator are encased
in a waterproof, streamlined nacelle mounted on a streamlined pylon.
10-26
10581090
Figure 10-31
Verdant Power Free Flow Turbine
Figure 10-32
Verdant Power Turbine Lowered into the East River Prior to Mounting on a Monopile
Verdant Power’s Free Flow turbines can operate in both tidal and river settings. Turbines
deployed in tidal settings are assembled with internal yaw bearings, which allow the turbines to
pivot with the changing tide and capture energy for the majority of the day. Turbines deployed in
rivers are fixed and generate power throughout the day. Depending on the site, various types of
devices can be used to anchor the turbines underwater.
10-27
10581090
10.5.2.10 Vortex Hydro
Founded in 2004, Vortex Hydro Energy LLC is a Michigan company that has developed a
technology it calls Vortex Induced Vibrations Aquatic Clean Energy (VIVACE). As its name
suggests, VIVACE uses vortex induced vibrations (VIV) to extract energy from ocean, river,
tidal, and other water currents. For decades, engineers have been trying to prevent such
vibrations from damaging offshore equipment and structures. VIVACE works by maximizing
and exploiting VIV rather than preventing it.
As shown in Figure 10-33, VIV results from vortices forming and shedding on the downstream
side of a bluff body in a current. Vortex shedding alternates from one side to the other, thereby
creating a vibration or oscillation. The VIV phenomenon is nonlinear and self-regulated, which
means it can produce useful energy at high efficiency over a wide range of current speeds.
Figure 10-33
Vortex Induced Vibrations Oscillates Objects in Fluid Currents
VIVACE devices can be positioned below the surface, thereby avoiding interference with other
river uses such as fishing, shipping, and tourism. In addition, VIVACE utilizes vortex formation
and shedding, the same mechanism fish use to propel themselves through the water, which
makes it compatible with marine life.
A VIVACE prototype is currently operating in the Marine Hydrodynamics Laboratory at the

University of Michigan. Testing is being funded by DOE and the Office of Naval Research.
10.5.3 Survival in Storms and Hostile River Environments
In-stream river energy conversion systems that are submerged below the surface do not bear
the full brunt of storms. For long-term survival in the marine environment, the key issues are
debris and ice, particularly at sites where ice break-up generates large pieces.
In Alaska, rivers normally begin to freeze in October, and freezing to solid ice with a thickness
of 4 to 8 feet is not uncommon. There is also a frazil ice layer below the solid ice. Ice breakup
normally occurs in April and clears by May. Ice breakup is potentially destructive, with large
pieces of ice scouring the river bottom and edges. Figure 10-34 depicts an ice jam during a
spring breakup in the Yukon River at Eagle, Alaska.
10-28
10581090
Figure 10-34
Ice Jam during Spring Breakup: Upstream View of the Yukon River at Eagle, Alaska
10.5.4 Environmental Impacts of In-Stream River Power Plants
Given proper care in siting and scaling projects, river in-stream power promises to be one of the
most environmentally benign electrical generation technologies. Early demonstration and
commercial river in-stream power plants should include rigorous monitoring to record both
environmental and natural impacts at nearby undeveloped sites.
No definitive in situ monitoring studies have been conducted to date due to the newness
of RISEC technology and lack of deployed systems. Two TISEC test programs with thorough
environmental assessments of sea life and environmental effects were started in 2008: the
Verdant RITE project in New York’s East River and the MCT SeaGen project in Strangford,
UK. The results from these TISEC programs should be relevant to RISEC applications.
EPRI is currently testing interaction between hydrokinetic turbines and fish in experimental
flumes at Alden Research Laboratories in Holden, Mass., and at the USGS Conte Anadromous
Fish Research Laboratory in Turners Falls, Mass. The results of these studies, sponsored by
industry and the DOE, will be available in early- to mid-2011.
10.5.5 Permits for In-Stream River Power Plants
The novelty of RISEC technology has to date triggered conservative evaluations and an
extensive environmental review process at the federal, state, and local levels. As of October 4,
2010, there were 114 active FERC-issued Preliminary Permits and 10 pending Preliminary
Permits. There were no Preliminary Permits for inland hydrokinetic projects issued before 2008.
A list of hydrokinetic Preliminary Permits and licenses is posted on the FERC website at:
www.ferc.gov/industries/hydropower/indus-act/hydrokinetics/permits.asp.
10-29
10581090
10.5.6 Overview of Regulatory Status for River In-Stream Energy Conversion

Projects
The regulatory permitting and licensing processes associated with U.S. river in-stream projects
can be quite involved, complex, lengthy, and costly. Indeed, regulatory issues represent a
significant impediment to capital acquisition and in-stream river energy development in the
United States.
Since 1920, constructing and operating a non-federal hydroelectric project in the United States
result of legal interpretation from FERC, unconventional tidal and wave energy hydroelectric
projects were placed under FERC jurisdiction. In October, 2007, FERC announced a new
streamlined permitting approach for marine and hydrokinetic energy projects intended to reduce
the duration of the approval process from five years to six months for five-year licenses for pilot
projects under 5 MW. FERC issued a white paper in April 2008 explaining the new pilot project
licensing process, posted online at http://www.ferc.gov/industries/hydropower/indus-
act/hydrokinetics/pdf/white_paper.pdf.
10.6 Design, Performance, Cost, and Economic Feasibility
10.6.1 RISEC Sites
There are many factors to consider when evaluating potential sites for an in-stream river energy
plant, including:
• First and foremost, a high annual current flow, in terms of both volume and speed.
• A transmission and distribution system that can transport the power from the river plant into
the grid and a substation interconnection point close to shore.
• Sufficient depth for the device to be deployed off the river bottom and completely
submerged, and which provides navigation clearance if required.
• Minimum conflicts with other river uses.
10.6.2 RISEC Devices Studied
In 2008, EPRI completed a site characterization study of the six sites in Alaska shown in Figure
10-35. After reviewing the data for those six sites, three sites were selected for conceptual
feasibility design studies: Tanana River at Whitestone, Yukon River at Eagle, and Kvichack
River at Igiugig. The Igiugig and Eagle sites are connected to small isolated village grids, while
the Whitestone site is located near a 26-kV transmission line.
For each of the three sites, point designs for both a single-unit demonstration and a commercial
tidal power plant were used to estimate cost and performance. Performance estimates were
developed using river current data predictions obtained from USGS.
10-30
10581090
Tanana at Whitestone
Tanana at Manley Hot
Springs
Yukon at Eagle
Taku at Juneau
Yukon at Pilot Station
Kvichak at Igiugig
Figure 10-35
Site Location Overview, with Chosen Sites Indicated in Yellow
Figure 10-36 shows the cross-sectional water depth profiles at the USGS measurement stations
of the three sites of interest.
0
0 50 100 150 200 250 300 350 400 450 500
-2
-4
Water Depth (m)
-6
-8
-10
Kvichack @ Igiugig
-12 Yukon @ Eagle
Tanana @ Big Delta
-14
Distance from Shore (m)
Figure 10-36
Water Depth Profiles at Three Alaska Sites during Typical River Discharges
Figures 10-37 and 10-38 show the monthly average velocities and monthly average cross-
sectional power densities at the three sites. Rotor power output is directly related to the
hydrokinetic power density at a river site and significantly affects the project economics. As
noted earlier, the hydrokinetic power density of a free-flowing stream is proportional to the cube
of fluid velocity. Although the power densities for all three sites are much higher during summer
than winter, the Iguigig site exhibits much less summer/winter variability than the other two
sites. This is a direct result of the storage provided by Lake Llama upstream of the Iguigig site.
The higher river discharge rates during summer and associated higher velocities are a direct
result of the annual snow melt. \
10-31
10581090
2.50
Yukon River @ Eagle
Tanana River @ BigDelta
Monthly Average Velocity (m/s)
2.00 Kvichak River @ Igiugig
1.50
1.00
0.50
0.00
Month
Figure 10-37
Average Water Velocities by Month at Three Sites in Alaska
4.50
Monthly Average Power Density (kW/m^2)
Yukon River @ Eagle

4.00 Tanana River @ BigDelta
3.50 Kvichak River @ Igiugig
3.00
2.50
2.00
1.50
1.00
0.50
0.00
Month
Figure 10-38
Average Power Densities by Month at Three Sites in Alaska
10.6.3 RISEC Design, Performance, and Cost
For each of the three Alaskan sites described in Section 10.6.2, EPRI prepared a point design,
performance, and cost estimate for both a single-unit demonstration and a commercial RISEC
power plant. The performance and cost estimates were used to assess the economic feasibility
in terms of a simple payback period. Table 10-3 presents the cost estimates for the three
commercial plants. These cost estimates are in 2008 dollars and do not include permitting
and engineering costs.
10-32
10581090
Table 10-3
Cost and Performance Estimates for Alaska In-Stream River Power Plants
Kvichak at Iguigig Yukon at Eagle Tanana at Whitestone

Number of Units 3 1 30
Annual Electrical Energy 207 107 1,325
at Busbar (MWh/yr)
Nameplate Rating (kW) 41 47 593
Capital Cost $/kW
Month/Year Dollars December 20092 December 20092 December 20092
On-shore Transmission
and Grid Interconnection $2,575 $2,246 $178
Subsea Cables $1,025 $894 $71
Powertrain Mounting $1,100 $589 $627
Power Conversion
Modules $1,325 $1,505 $1,168
Structural Sections $925 $327 $565
Assembly $225 $131 $150
Transport to Site $150 $65 $105
Onsite Assembly $150 $44 $105
Device Deployment $38 $11 $26
Contingencies $350 $196 $232
Total Plant Cost (TPC) $7,863 $6,008 $3,229
AFUDC (interest during
construction) $786 $601 $326
Due Diligence,
Permitting, Fees and
Taxes $393 $300 $161
Total Capital
Requirement $9,042 $6,909 $3,717
Yearly O&M Costs
$/kW-yr $302 $140 $224
Unit Availability (%) 90 38 90
Cost Estimate Accuracy +30% Capital Cost +30% Capital Cost +30% Capital Cost
Rating
-30 to +80% O&M -30 to +80% O&M -30 to +80% O&M
Costs Costs Costs
10-33
10581090
10.6.4 RISEC Economic Feasibility
Table 10-4 summarizes the economic analyses for the three commercial-scale river power plants
in Alaska. The cost estimates are in 2008 constant dollars and do not include permitting and
engineering costs. Iguigig and Eagle are remote villages and their RISEC plants were sized to
meet the summer low daily load (40 kW for Iguigig and 70 kW for Eagle).
Table 10-4
Cost Estimates for Three Feasibility Evaluation Sites
Site Parameters Iguigig Eagle Whitestone

Ice Freeze-over? No Yes No
2 2
Annual Average Power Density 1.48 kW/m 1.5 kW/m 0.67 kW/m2
Mid-channel Average Power density 3.24 kW/m2 3.2 kW/m2 1.48 kW/m2
Average Total Kinetic Power 719 kW 4,601 kW 762 kW
Summer/Winter Power Density 1:4 1:20 1:10
Variability
Site Distance from Shore 60 m 150 m 50 m
Grid Feed-In Limit 40 kW 70 kW > 5 MW
RISEC Plant Parameters
# of RISEC Devices 3 2 30
# Rotors per Machine 4 4 4
Rotor Diameter 1.5 m 2m 2m
Rated Capacity 42 kW 61 kW 593 kW
Plant Annual Generation 220 MWh/year 113 MWh/year 1,356 MWh/year
Availability 90% 38% 90%
Cost and Economic Parameters
Installed Cost $269,000 $308,000 $1,821,000
Installed Cost per kW $5,800/kW $7,500/kW $3,100/kW
Avoided Cost (selling price) 0.65 $/kWh 0.65 $/kWh 0.18 $/kWh
Simple Payback Period 3 Years 4 Years 8 Years
Notes:
1. The value of electricity revenues is the avoided cost. For a rural Alaskan utility running on diesel, the avoided
cost is essentially the fuel cost. With fuel costs of $8/gallon delivered and efficiencies of 13 kWh/gallon, the
avoided cost is typically $0.65/kWh. The O&M cost of a diesel genset is $0.02-$0.05/kWh, but we
conservatively assume that a genset idling in the background has no O&M savings. For the grid-connected
Whitestone case, a value of $0.18/kWh was used.
2. The simple payback period (SPP) is the year when the yearly revenues (calculated as the product of the annual
electricity produced multiplied by the avoided electricity cost) equals the sum of the capital cost plus O&M cost.
Escalation of non-fuel cost was assumed to be 3% per year and escalation of fuel costs was assumed to be 8%
per year.
10-34
10581090
Table 10-5 summarizes the key environmental issues, possible impacts, and potential mitigation
approaches surrounding RISEC project implementation.
Table 10-5
Major Environmental Issues and Mitigation Recommendations
River Issue Impact(s) Mitigation
Withdrawal of Retardation of river discharge rates. Limit the withdrawal of river energy
River Energy to a level that does not result in any
noticeable ecological effects.
Interactions with Fish mortality seems to be the most Limit the rotation speed of blades to
Marine Life, significant issue. Impacts on the benthic a level which precludes fish
Seabirds and ecosystems occur primarily during mortality.
Benthic installation; impacts on sediment
Ecosystems transport processes are also of concern.
Emissions Applies to devices with closed-circuit Use only biodegradable fluids.

hydraulic systems where working fluid
(which may be biodegradable) may leak
or spill during transfers, and use of fluids
for installation (i.e., drilling sockets in
hard rock seabed).
Visual The aesthetic effect of visually impacting Fully submerge the energy
Appearance and a pristine river scene may be conversion devices, land the cable
Noise unacceptable in some situations. under the shore.
Conflicts with Potential conflicts with recreational uses, Hold siting, design and installation,
Other Uses of commercial shipping, commercial fishing, operation, and procedure
River Space dredging, and other activities. discussions with all local
stakeholders prior to making final
plant detail design decisions.
Debris and Ice Potential damage to turbine rotor and the Install trash racks to reduce debris
(including ice entire system. impacts and remove devices during
break-up) spring ice break-up to avoid damage.
As of October 2010, there are three installed RISEC devices in the United States: a 100-kW
Hydro Green turbine at Hastings, Minn.; a 25-kW EnCurrent system on the Yukon River at
Eagle, Alaska; and a 5-kW EnCurrent system on the Yukon River at Ruby, Alaska. Table 10-6
presents the EPRI estimate of river in-stream MW capacity that is expected to come on line in
the United States—meaning commissioned and delivering electricity to the grid—during 2010
though 2013.
10-35
10581090
Table 10-6
Installed and Planned U.S. River In-Stream Power Capacity (MW)
Developer Project Name- Site 2010 2011 2012 2013
ABS Alaskan Yukon River at Ruby, Alaska 0.005
Hydro Green Hastings, Minnesota 0.100
Alaska Power & Yukon River at Eagle, Alaska 0.025

Telephone
Free Flow Power Many on the Mississippi River 0.1 100s of 100s of
Projects 1 MW MW
Other Hydro Green Many in Alaska, Mississippi 0.1 100s of 100s of

Projects 1 River MW MW
Other 2008-Issued Various 0.1 100s of 100s of

FERC Prel. Permits MW MW
TOTAL NEW YEARLY CAPACITY 0.130 0.3 ? ?
CUMULATIVE 0.2 0.5 ? ?

1. If Free Flow Power and Hydro Green complete their license applications within the three-year time frame of the
FERC preliminary permit, they should have their construction and operation licenses in 2011 and construct and
begin operation in 2012 or shortly thereafter.
In general, EPRI expects that river in-stream energy will experience a growth rate that is limited
by regulatory barriers and government support for the emerging river energy industry. EPRI
estimates that there could be 500 MW of river in-stream energy plant capacity by 2025. This
assumes that FERC and other regulatory agencies change their practices to enable river in-stream
plants to be permitted at about the same cost as onshore wind plants, and that Congress enacts
the same financial incentives for river in-stream energy as for wind energy. There is a potential
for river in-stream capacity to significantly increase if Free Flow Power and Hydro Green build
their plants under the existing Preliminary Permits issued by FERC. Of course, it is unknown at
this time whether and how many of the RISEC projects in the FERC queue will actually be built.
10.9 RISEC R&D Needs
Hydropower Technologies Program, EPRI sponsored a workshop for the water power industry in
October 2008 [4]. The purpose of the workshop was to identify and prioritize research,
development, deployment and demonstration (RDD&D) needs that will further the deployment
of conventional hydro/pumped storage and emerging wave and marine hydrokinetic (MHK)
technologies to increase domestic, low-carbon energy production. The priorities identified are
intended to shape both EPRI’s and DOE’s research agendas, and will support R&D initiatives
throughout both the public and private sectors.
10-36
10581090
The U.S. RDD&D Needs Workshop used the 12 topics from the U.K. Marine Energy
Technology Roadmap as the starting point for developing the U.S. technology needs:
• Resource Modeling
• Device Modeling
• Experimental Testing
• Moorings and Sea Bed Attachments
• Electrical Infrastructure
• Power Take Off and Control
• Engineering Design
• Lifecycle and Manufacturing
• Installation, O&M
• Environmental
• Standards
• System Simulation.
Four other topics identified by the Steering Committee prior to the workshop are not on the U.K.
Roadmap list:
• Materials—low cost, corrosion and biofouling
• Energy Storage
• System Configuration Evaluations
• Vision, Goals, Objectives and Roadmap.
Two other topics identified by workshop participants were:

• Master Generation and Transmission Plan
• Education.
• Testing (development including experimental through pilot demonstration)
• Environmental (which will require device testing and deployed projects)
• Performance Standards.
Once funding is available, specific programs and projects for high-priority topics identified in
this workshop should be developed and implemented. Clearly, the RDD&D topics of
“Environmental,” “Standards,” and “Testing” need to be addressed in a comprehensive fashion.
Also, these topics may warrant further consideration in follow-on meetings to be explored in
more detail with knowledgeable experts in the field.
10-37
10581090
Though less urgent, topics falling below the high-priority ranking with less consensus are still
important and need to be addressed.
10.10 Conclusions
Considerable potential exists for generating electric power from river energy in the United States
and many other places in the world. Most of the high-power-density U.S. river hydrokinetic
energy potential exists in the Pacific Northwest and particularly in the state of Alaska.
Many Preliminary Permits have been issued for the Mississippi, Missouri, and Ohio rivers.
EPRI has not studied these river systems; however, we suspect that the power densities are lower
than for rivers in Alaska and the Pacific Northwest. It remains to be seen if Free Flow Power,
Hydro Green or other technology developers are able to develop a water turbine for lower power
densities that could be economically competitive with existing sources of energy without
government incentives and with a potentially costly licensing environment.
10.11 Resources and References
EPRI River Power (RP) Reports are available from the River page at
http://oceanenergy.epri.com/.
10.11.2 References
1. Miller, G., J. Franceschi, W. Lese and J. Rico. 1986. The Allocation of Kinetic Hydro Energy
Conversion SYSTEMS(KHECS) in USA Drainage Basins: Regional Resource and Potential
Power. NYUDAS 86-151.
2. River In-Stream Energy Conversion (RISEC) Characterization of Alaska Sites, EPRI, Palo
Alto, CA: 2008. EPRI-RP-003-AK.
3. System Level Design, Performance, Cost and Economic Assessment -- Alaska River In-
Stream Power Plants, EPRI, Palo Alto, CA: 2008. EPRI-RP-006-AK.
Marine and Other Hydrokinetic Renewable Energy. EPRI, Palo Alto, CA: 2008.
10-38
10581090
11
RENEWABLE ENERGY GRID INTEGRATION
TECHNOLOGIES
Non-hydro based renewable generation has played a relatively small role in meeting electricity
demand in the United States, reaching roughly 3.0% in 2009, with approximately 70% of this
coming from wind power (EIA Annual Energy Outlook 2010). Looking ahead, the role of
renewables is clearly expanding, and renewable resources will account for about 30% of the
expected new generation capacity additions during the next 10 to 15 years. Over a longer period
of 20–25 years, wind, biomass, geothermal, incremental hydro, and solar power are expected to
increase by more than ten-fold compared to today’s deployments. With this rate of growth, the
renewable component of electric generation is likely to reach 30% of the total electric sector by
2030.
Managing renewable generation, particularly variable generation (VG) such as wind and solar
PV, will require changes in the power delivery system at both the transmission and distribution
levels. At the transmission level, the variability and uncertainty of certain renewable resources
will lead systems operators to redefine ancillary requirements, such as reserve requirements and
ramp rates, in order to operate the grid reliably. Operational changes may affect many tools and
processes currently in use, and will increase in importance for operators working to meet
reliability standards. Systems will need to be planned to ensure the system can integrate large
amounts of variable generation, both in terms of the transmission infrastructure and the
generation portfolio.
At the distribution level, variability of renewable generation will also require changes. The issues
are different but equally important for enabling penetration of distributed generation into existing
and future systems. These include evaluating interface devices, analytics, studies, applications
with end-use resources, and assessment of new technology for effective interconnection and
integration of renewable and other distributed generation. Recognizing the need for change and
adapting the electric grid can enable higher penetration levels without reducing safety, reliability,
or asset utilization effectiveness.
This chapter addresses technologies and strategies that will ease the integration of renewable
energy into the electricity grid. Energy storage, power electronics, resource forecasting, grid
planning and operating tools, increased flexibility from conventional plant, demand response and
increased transmission throughput can all be used to mitigate the impacts of adding large
amounts or wind, solar, and other variable generation to the electricity grid.
11-1
10581090
Renewable Energy Grid Integration Technologies
11.1 Renewable Generation Integration Challenges
The grid integration challenges related to higher penetration of renewable resources are
dependent on the location, technology and penetration level. For example, rapid expansion of
wind generation capacity may create more challenges to the grid than additions of solar, wave,
tidal, biomass and geothermal plants, due to its greater variability and unpredictability. All have
unique generation characteristics that will affect their integration into the electric grid. The
primary challenges for the non-thermal renewable generation technologies are variability and
predictability of generation output, coupled with the fact that many of these technologies do not
behave in a similar way to conventional electrical generators (Table 11-1).
Table 11-1
Comparison of Non-Thermal Renewable Generation Technologies
Technology Wind Solar CSP Solar PV Wave Tidal Current
Development Emerging Pre- Pre-

Commercial Commercial
Status Commercial Commercial Commercial
Energy Uneven solar Wind over Gravity of

Sun Sun
Source heating water moon & sun
Power Density 0.25–0.35 0.18–0.3 0.15–0.24

20–25 kW/m2
(Annual 8760 kW/m2 kW/m2 kW/m2 2–10 kW/m2
(US West
Average per (US Great (Southwest (Southern & (Northern US)
Coast)
unit area) Plains) US) Western US)
Day-night; Day-night;
Weather Diurnal and
Variability clouds, haze, clouds and Sea states
patterns semi-diurnal
and humidity haze
Predictability Hours Minutes Minutes Days Years
Thermal renewable generation resources such as biomass, geothermal and biogas will behave
much as thermal plants currently do. Therefore, the issues likely to be seen on the distribution
and transmission systems will not be different than those seen when installing conventional
sources of electricity. This chapter does not examine these in any great detail. When certain
characteristics of these plants are relevant they are mentioned, but such plants are not the focus
of this chapter.
About 50% of currently deployed renewable generation cannot be dispatched and their available
power output depends on the weather. This number is likely to grow with increased variable
generation. Thus, grid-operating flexibility is reduced and reserve requirements increased to
accommodate these less controllable resources.
Wind in particular, but also run-of-river and geothermal, are typically remote and not necessarily
near primary transmission rights-of-way. The economics of adding transmission or distribution
for renewable energy can be challenging if capacity factors are low or output is variable. Tidal
11-2
10581090
and wave energy devices are currently not near commercial deployment, so they are not focused
on in this discussion. In the case of run-of-river hydropower, not shown in the table, output may
be more predicable with changes related to wet and dry seasons and longer-term weather
patterns.
Power delivery contracts, as well as environment constraints, can also affect renewable
generation flexibility. For example, many hydro facilities must meet river flow requirements
regardless of grid power needs, geothermal is often purchased as “must take” energy, and
municipal solid waste facilities are driven to dispose of their accumulating fuel. In contrast, solar
thermal electric technology includes about 30 minutes of natural thermal inertia and is very
conducive to the addition of additional thermal storage without significantly increasing cost. In
many areas of the West, two to six hours of thermal storage can provide solar plant flexibility
and carry afternoon solar peaks into early evening load peaks.
This chapter first examines the current system operation paradigms (i.e. before high penetration
of VG is seen), then outlines relevant characteristics of wind and solar PV power that will impact
those paradigms. It then examines transmission and distribution networks separately: first
examining the impact VG will have on the system, and then suggesting solutions to these issues.
11.1.1 Integration Technologies for Transmission and Distribution

Requirements for integration technologies are expected to evolve from location-specific
interconnection requirements to larger area grid codes, intelligent distribution grid devices,
interface hardware purchase specifications and energy purchase power agreements (PPAs).
Technologies and strategies to integrate transmission-connected renewable generation such as

wind and solar thermal fall into four main categories, as listed below and shown in Figure 11-1:
• Better renewable conversion technology for interconnection and control.
• Utility-scale energy storage to address energy output intermittency.
• Visualization of grid operations to address increased generation variability, and tools to
better utilize this visualization.
• Transmission infrastructure to address more dispersed and/or remote resources, and reduce
variability from these resources.
11-3
10581090
Figure 11-1
Different Integration Technologies can Combine to Ease Integration of Renewable Energy
Resources
For increased distribution-connected renewable generation, the categories are similar but the
specific analytical methods, integration tools and needed hardware are different. Distribution
technologies include advanced voltage control devices, inverters with reactive power control,
and communications to determine the status of distributed generators for operation and safety.
Storage technologies are more distributed, with higher levels of integration with other end-use
energy resources, including load control. New tools for distribution planning and for aggregation
and operation will be needed with a higher level of renewable penetration. Distribution-
connected renewable technologies will need to be designed to effectively interact with both the
distribution grid and neighboring loads and generators.
For these technologies to be effective, utility experience is needed in their deployment and
application. Applying new analytical methods and tools to the grid infrastructure and operations
is expected to enable higher penetration of intermittent and less controllable renewable
generation resources. Deployment will also need to involve new research to address life-cycle
issues including consideration in planning and operating as well as maintainability, reliability
and durability. To best integrate large amounts of VG, strategies which best use the above
technologies will be needed.
11.2 Basics of Grid Operating Systems
This section outlines the operation of the electrical grid, particularly highlighting those areas to
be discussed later which will be most affected by VG integration.
11.2.1 Functions of System Generation
The main function of generation in an electric power system is to produce electric energy to
serve the power system load. This function may include regeneration from stored energy. The
Authoritative Dictionary of IEEE Standards Terms, ANSI/IEEE Std. 100, defines generation as
producing or storing electric energy with the intent of enabling practical use or commercial sale
11-4
10581090
of the available energy. A generating station is also defined as “a plant wherein electric energy is
produced from some other form of energy by means of suitable apparatus (for example,
chemical, mechanical or hydraulic).”
The operation of electric power systems is fundamentally different from services provided by
other types of utilities. Electric systems have two unique physical operating constraints:
• Electric energy is not commercially stored23 like natural gas and water. Production and
consumption (generation and load) must be balanced in near real-time. This requires
continuous monitoring of load, generation, and the voltages and energy flows throughout the
power system, as well as adjusting generation output to match consumption.
• The transmission and distribution network is primarily passive, with few “control valves” or
“booster pumps” to regulate electrical energy flows on individual lines. Flow-control actions
are limited primarily to adjusting generation output and to opening and closing switches to
add, remove, or reroute transmission and distribution lines and equipment from service.
These two operating constraints lead to four reliability consequences with practical implications
that dominate power system design and operation:
• Every action can potentially affect all other activities on the power system. Therefore, the
operations of all bulk-power participants must be coordinated.
• Cascading problems that quickly escalate in severity are a real threat. Failure of a single
element can, if not managed properly, cause the subsequent rapid failure of many additional
elements, potentially disrupting the entire power system.
• Contingency readiness may limit current operations. For example, likely power flows that
occur if another element fails could limit the allowable power transfers.
• Maintaining system stability and reliability often requires that actions be taken
instantaneously (within fractions of a second), requiring automatic computations,
communications, and controls.
These consequences bear directly on how wind and solar power interacts with the power system.
11.2.2 Expectations Based on Traditional Generator Performance
Reliable and low-cost operation of the interconnected electric utility is not a “natural” state but
relies on operational scheduling, planning, and coordination of a multitude of electric utility
control systems. A traditional electric utility generator is not just a source of energy. Operators
rely on an array of performance capabilities and specific services from traditional generators.
Some of these services are simply provided by generator operators as part of their participation
in an energy market. Others are obtained from bidding into separate ancillary services markets
that have evolved, particularly in light of deregulation. The exact definition of these services
varies in different markets but, in general, they fall into the categories described in Table 11-2.
23
Electricity is not “stored” directly. Rather, it is converted to another form of energy and re-converted later.
Pumped storage hydro converts electricity to mechanical potential energy by lifting water. Batteries convert electric
energy to chemical potential energy. Their-conversion to electricity uses conventional generators or inverters.
11-5
10581090
Table 11-2
Functions and Services Provided by Generation
Functions and Services Description Time Frame

Baseload Units Energy (firm) scheduled well in advance, based on Long-term
(non-regulating) availability, price, and long-term contracts. commitments
Committed Units (usually Energy (firm) scheduled based on availability and Day before plan,
with regulation capacity) price to meet block load, with LOLE1 and load hourly resolution
forecasts considered.
Load-following or Energy ramping to follow the load, met by adjusting Hourly plan with 5- to
Energy-balancing Units generation schedules and the imbalance energy 10-minute resolution
market.
Frequency Regulation Service provides capacity based on a signal from Every few minutes,
(regulating reserves) dispatcher, with AGC2 to meet CPS1 and CPS23 and minute-to-minute
no net energy4. resolution
Reactive Supply and Service of injecting or absorbing of reactive power Continuous with
Voltage Control to control local transmission voltages (usually response in seconds
provided with energy).
Spinning Operating Service to provide energy in response to Begin within 10 sec
Reserves contingencies and frequency deviations. full power in 10 min
Non-spinning operating Service to provide load/generation balance in Respond within 10
reserves response to contingencies, not frequency response. minutes
Replacement Reserves Service to restore contingency capacity to prepare Respond within 60
for the next generation or transmission contingency. minutes, run up to 2
hours
System Black Start Service to restore all or a major portion of the power As required
system without outside energy after a total collapse.
1. Loss of load expectation (LOLE) is the probably measure that a load cannot be served with available
generation.
2. Automatic generation control (AGC) is a method for adjusting generation to minimize frequency deviations and
regulate tie-line flows.
3. Control performance standards (CPS1 and CPS2) are minute-to-minute and 10-minute average criteria for
load frequency control in each control area. These criteria require control areas to maintain their area control
errors (ACE) within tight limits. ACE is measured in MW and is defined as the instantaneous difference
between the actual and scheduled interchanges plus frequency bias (imbalances that bias the system toward
maintaining 60 Hz).
4. Frequency regulation service is usually provided by generation that is on-line and delivering some level of
base energy power on a full-time basis or scheduled. The service increases and decreases power output so
that the average output over the scheduled period does not change—that is, there is no net change in
delivered energy attributed to the frequency regulation service. Consequently, energy-storage devices could
provide this service.
Response time is one of the critical factors in defining generation capabilities and services. The
reaction time is critical because of the nature of the electric utility system, which operates with
very little electrical storage as described above. Furthermore, the degree to which variable wind
and solar power can be absorbed without significant integration issues, as well as the value
attributed to wind power, is determined by how well the traditional generation functions are
fulfilled, as well as how much they need to be fulfilled.
11-6
10581090
11.2.3 Typical Power System-Operating Strategy
With or without variable renewable generation, the system operating strategy will likely include
energy balancing from both committed and reserve units, unit commitment block scheduling, and
an economic dispatch procedure. This strategy starts with a forecast of system load with time-
varying characteristics and expected minimums and maximums. Inventories of available
generation and the estimated cost of energy as well as other services are maintained. Desired
reserve margins are set. Baseloaded units and firm energy blocks can be scheduled so that, in the
short term, the operator is only dealing with the differences between the predicted and actual
load and/or available generation.
Thus, the system operator is effectively managing supply and demand over three time scales:
1) days- and hours-ahead scheduling, 2) intra-hour load following and balancing, and 3) fast
regulation to maintain system frequency and voltage. The impact that wind generation may have
on this strategy is discussed later. All three time frames are also important to effectively
integrating any generation resource, including wind.
11.2.3.1 Energy Balancing
System operators run load-following and regulation markets to ensure that there is adequate on-
line generation capacity for ramping up or down to follow the load and regulate the faster and
more random changes in load.
• For frequency regulation: Load-following units must be on-line and provide fast response to
meet minute-by-minute fluctuation in the system energy balance. The North American
Electric Reliability Council (NERC) has set guidelines on response-time performance for
plants providing frequency regulation.
• For intra-hour load following: These resources are dispatched to follow within-the-hour load
changes in the frequency consistent with the economic dispatch cycle (5 to 10 minutes per
cycle).
• For reliability: This is primarily contingency reserves (use of spinning and fast-start
generators) that can be deployed within 10 minutes in order to replace any MW loss
following line or generator outage. Typically, one-half of system operating reserves is
spinning, so that a sudden loss of generation will not result in a loss of load, with the balance
available to serve the load within 10 minutes. This includes replacement reserves.
Both regulation and load following are dealt with on a balancing or control area basis. It is not
necessary to compensate for each and every load variation directly, but the aggregate change in
the area must be balanced; this can include scheduled flows on tie-lines. The fast, random
fluctuations associated with regulation are typically uncorrelated. Consequently, the total
regulation requirement is not the sum of all the regulation requirements of the individual loads
and uncontrolled generators, but is instead the sum of the correlated components. Various plants
in the generation mix will bid in or be scheduled to provide balancing (energy regulation) to the
system. Table 11-3 below shows the typical controllability characteristics of different types of
generation.
11-7
10581090
Table 11-3
Comparison of Output Controllability for Various Generation Technologies
(EPRI, T. Key, November 2010)
Controlled Un- Minimum Typical

Start-Up Minimum
Generation Source Ramp Rate Controlled Output % of Unit Size
Time Cycling Time
(Up or Down) Variability Rated** [MW]
Conventional Minutes Minutes 100% in 30 sec Seasonal 10% 1-200

Hydro* to 2 min
Steam – Coal Hours-Days Once/day 1 to 2 % per Rarely 40 to 50% 10-750

minute
Steam – Coal Hours-Days Once/day 0.2 to 2 % per Rarely 40 to 50% 10-750

Critical Pressure minute
Combined Cycle Hours Twice/day 2 to 3% per min Rarely 45 to 55% 40-400

Natural Gas
Combustion Minutes More than 3 to 5% per Rarely 55 to 65% 20-250

Turbine Simple twice/day minute
Cycle
Diesel Internal Minutes More than 100% in 15 to Rarely 10% 0.1 – 20

Combustion twice/day 30 seconds
Steam – Nuclear Hours-Days Days 0.2 to 2 % per Rarely 90 to 100% 750 - 1500
minute
Wind* Minutes* Minutes* Seconds* 10%/min 2 to 5% 0.1 – 4.5
Solar PV* Seconds* Minutes* Seconds* 10%/sec 2 to 5% <0.1 – 1
Solar Thermal 60-90 More than ~2% per min Daily site- 10% 50-150
6-hr Storage* Minutes twice/day (faster down) dependent
Compressed Air Minutes More than 30 to 40% per Rarely 25 to 30% 20-250
Energy Storage twice/day minute
Battery Energy Seconds to Minutes Seconds Rarely 2 to 5% <0.1 – 1

Storage Minutes
* Controlled output changes depend on available resource (water, wind, sun), while uncontrolled output variability
such as output ramp rates vary by technology and weather.
** Minimum output is dictated by economics as well as technical limits, and market incentives could act to lower
minimum outputs
It should be noted that controlling loads is another way to balance supply and demand. Load
control or load as a resource is being considered by most ISOs. It can be relatively fast and may
be the least-cost option. With the advent of the smart grid and associated technologies,
controllable or responsive loads may well increase significantly in the next decade, and such
resources may prove very useful.
11-8
10581090
Load-following requirements tend to be more highly correlated; most area profiles rise in the
morning and drop off in the evening. Still, because load-following requirements are not perfectly
correlated, the total system load-following requirement is less than the sum of the load-following
requirements of individual end users. This aggregation of loads has a powerful effect on system
planning and operation because the system must respond to total variations, not the sum of
individual variations.
11.2.3.2 Unit Commitment
Unit commitment (UC) is part of a set of programs for scheduling generation on an hourly to
weekly basis. Vertically integrated utilities typically run their UC optimization computer
programs the day before the operation is needed. These complicated programs accept as inputs
detailed information on the characteristics of the individual generating units that are available to
produce electricity on the following day, including unit status, minimum and maximum output
levels, ramp-rate limits, startup and shutdown costs, durations and lead times, minimum
runtimes, and unit fuel costs at various output levels. Unit commitment can be run for both
markets and system operation; markets may not include the security-constrained aspect seen in
commitment decision made by system operators. Security-constrained unit commitment
optimizes for production cost, while including flows on the transmission network and ensures
that the system can recover from contingencies.
11.2.3.3 Economic Dispatch
Once generators are committed (turned on and synchronized to the grid), they are available to
deliver power to meet customer loads and reliability requirements. Utilities will typically run
their least-cost dispatch model every five minutes or so. This model forecasts load for the next
five-minute interval and decides how much more or less generation is needed during the next
interval for regulation to meet the system load. The model then selects the least-cost combination
of units that meet the need during the next intra-hour interval.
11.2.3.4 Role of Ancillary Services
In addition to committing and dispatching units, system operators contract for ancillary services
to support operation of the grid. FERC has defined ancillary services as those necessary to
support the transmission of electric power from seller to purchaser, given the obligations of
control areas and transmitting utilities within those control areas to maintain reliable operations
of the interconnected transmission system. This statement recognizes the importance of ancillary
services for bulk-power reliability and to support commercial transactions. Though the resources
necessary to create these services are generators, the services themselves must be deployed and
controlled by the same system operator that controls the transmission system.
Ancillary services are conceptually well defined. They have existed throughout the history of the
power system. However, the details of obtaining them from markets are still evolving. Ancillary
services normally provided by generation include regulation, contingency reserves, voltage
control, and black start capability.
11-9
10581090
11.2.3.5 Regulation and Regulating Reserves
Power system operators need rapid, automatic generation control (AGC) over some resources to
compensate for normal short-term fluctuations in the aggregated load and generation and meet
NERC control performance criteria. AGC is related to maintaining the real power balance and
stable frequency in the power system. Time scales are minute to minutes.
The power system operator could eliminate the need for regulation by insisting that all end-use
consumption, such as for large industrial or commercial customers, conforms precisely (minute-
by-minute) to its power purchase schedule. Alternatively, each end-use consumer could be
required to contract with a power generator (independent power producer) to precisely match its
individual minute-to-minute load fluctuations. These alternatives are both somewhat impractical
and could be wasteful from a viewpoint of minimizing needed resources.
Instead of achieving energy regulation with load, system operators accomplish this requirement
by providing an AGC signal to increase or decrease output to on-line generators that have agreed
to provide this service. Operators can also call on reserve generation to provide regulation.
11.2.3.6 Reserve Generation
Contingency generation reserves are needed in case of sudden loss of generation, off-system
purchases, unexpected load fluctuations, and/or unexpected transmission line outages. These
reserves may be spinning or non-spinning reserve, or simply serve as replacement. Reserve
generation units may be categorized as follows:
• Contingency reserves. The power system must always be prepared to survive the
unexpected loss of a generator or a transmission line. To accomplish this, the power system
operator is required to maintain contingency reserves consisting of spinning reserve and non-
spinning reserve sufficient to meet the largest credible contingency.
• Spinning operating reserve. At least half the contingency reserve is typically spinning
reserve. Spinning reserve comes from generation that is on-line, not fully loaded, capable of
responding fully within 10 minutes, and able to maintain that output for at least two hours.
Spinning reserve units are also required to be responsive to frequency deviations.
• Non-spinning reserve. The definition of non-spinning reserve is similar to that of spinning
reserve except that the reserve is not required to be on-line or frequency responsive. In
addition, non-spinning reserve does not have to be provided by generation; instead, it can be
provided by dispatchable demand, interruptible exports, certified off-line generation, or
external imports.
• Replacement reserve. Contingency reserves must be restored so that the system is prepared
for a subsequent unexpected outage. Some regions specifically identify replacement reserves,
which are capable of responding within one hour and sustaining that response for an
additional two hours. They can be generators, loads, or resources from outside the
independent system operator’s (ISO) control area.
Together, spinning reserve, non-spinning reserve, and replacement reserve provide resources that
begin responding immediately to an unexpected event, are fully deployed within 10 minutes, are
capable of responding to a second event within one hour, and can sustain the total response for
11-10
10581090
three hours. This coordinated set of resources is designed to provide sufficient time for markets
to begin functioning again and return the system to normal operations. All of these services and
regulation are typically procured through day-ahead and hour-ahead markets run by the ISO or
balancing authority.
Figure 11-2 provides a summary of deployment times for various ancillary services. Reserves are
deployed only during contingency operations, while regulation and voltage control are required
during both normal and contingency operations.
Replacement Reserve
C ontinge ncy Oper atio ns
Non-Spinning Operating Reserve
Spinning Operating Reserve
Norma l & Co ntin gen cy Op era tions

Regulation
Voltage Control
0.1 1 10 100 1000
98041
TI ME (MINUTE S)
Figure 11-2
Ancillary Services are Distinguished by their Deployment Times and Durations
11.2.3.7 Voltage Control
Because of relatively strict geographic requirements for voltage control, it is not normally
procured through markets. Instead, all generators are required to provide limited reactive support
and voltage control without monetary compensation. Reactive support provided by generation or
absorption of reactive power (MVAR) is used to control voltage in the vicinity of the generator.
All generators are required to be capable of following a system-operator-supplied voltage
schedule under automatic voltage control within a power factor range of 0.90 lagging and 0.95
leading. The system operator will typically compensate generators if they require them to
provide additional reactive support and voltage control.
11-11
10581090
11.2.3.8 Black Start
Because of relatively strict geographic requirements for black-start capability, it is also not
normally procured through markets. System operators determine the system’s black-start
requirements to ensure that the system could be restored to service expeditiously if it should ever
fail completely. Long-term contracts are established with selected black-start units. Each black-
start unit must be capable of starting without external assistance within 10 minutes. It must be
capable of supplying the reactive power requirements and controlling the voltage of the
energized transmission system and operating for a minimum of 12 hours.
11.3 Integration Issues of Variable Generation
When wind, solar, and other variable generation resources are part of the generation mix,
operators must consider their unique characteristics when planning and operating the power
system. The following characteristics of variable generation are likely to require changes in
system operation and planning:
• Variability
• Uncertainty
• Limited reactive power control
• Distributed generation
• Remote location of some renewable resources.
11.3.1 Variability
The output of most traditional generators can be planned, as well as controlled, even minute-to-
minute to meet the expected daily load and to balance or regulate load variations. Variable
generation output depends on the weather and is determined by characteristics such as time-
varying wind speeds, solar insolation or tidal current. Its output varies day to day as well as from
hour to hour and even minute to minute. Figure 11-3 shows the annual wind output at a site in
the Midwestern United States with hourly resolution, while Figure 11-4 shows the hourly output
of a solar PV plant over one week.
Although two variable generation facilities may have the same capacity rating and annual energy
production, different wind regimes or solar insolation patterns may result in very different
hourly, daily, and seasonal operating schedules. Consequently, aggregating output over a large
area can reduce variability because the output of different widespread wind farms or solar plants
is not perfectly correlated. Additionally, variability is reduced when examining wind and solar
PV output together.
11-12
10581090
Hourly Wind Production
250
200
150
MW
100
50
0
1 2 3 4 5 6 7 8 9 10 11 12
Month
Figure 11-3
Wind Output Varies Diurnally, Seasonally, and with Weather Changes
Figure 11-4
Hourly Variation of Solar PV System Output Over a Few Days (Recorded Data for a 100-kW
Site Near Albany, NY, 1998)
Figure 11-5 compares typical variations in system load (electricity demand) with variations in
wind power generation. The difference is the amount of required generation scheduling and
energy regulation for non-wind power plants. Experience to date has shown that wind output is
not likely to coincide with peak load requirements and that there are interesting successes and
significant challenges to absorbing higher levels of deployment. As the footprint covered by the
installed variable generation increases, the total variability decreases because variability is
smoothed as outputs from wind plants are not fully correlated. Therefore, when considering the
effect that wind plants have on the system, it should be noted that each does not need to be
balanced individually, but rather the impact on the system net load as a whole should be
considered.
11-13
10581090
Figure 11-5
Hourly Load Shapes with and without Wind Generation (from DOE 20% Wind Energy by
2030 (Report DOE/GO-102008-2567, May 2008)
In addition to balancing supply and demand on an hourly basis, maintaining grid stability usually
involves moment-to-moment balancing of the load and generation by holding a constant
frequency. Rapid changes in system loads are matched by quickly ramping generation up or
down to balance load and generation in a 5- to 10-minute timeframe. Some units are required to
ramp more quickly, following an AGC signal to maintain system balance. Operators refer to this
process as regulation, as described previously, and the time frame is in minutes.
A substantial load change may cause sudden changes in frequency to which speed governors on
individual generator units must respond. The number of plants and their capacity to provide load
balancing and regulation are limited in any power system. The challenge is to have sufficient
energy balancing and regulating generation with ramp rates to meet the worst-case net load
fluctuations. Introducing variable generation usually means that additional flexibility will be
required on the part of other conventional generation to accommodate sub-hourly ramps. Figure
11-6 from NREL shows the distribution of 5-minute changes in generator loading requirements
with and without wind generation.
Wind power outputs have significant spatial variations. Outputs from nearby wind plants are
correlated, but outputs from distant wind plants are not. This spatial variation can smooth output
in show time frames. As shown in Figure 11-7, short-time fluctuations of wind power are
independent regardless of distance between wind plants. However, longer-time wind power
fluctuations are correlated for nearby wind power plants.
11-14
10581090
Figure 11-6
Distribution of Five-Minute Load Requirement Changes with and without Wind (from GE
Consulting 2008).
Figure 11-7
Example of Wind Power Plant Output Correlation with Distance (from NREL)
Rapid changes in power system loading or generation can also cause small voltage fluctuations
that manifest themselves as light flicker. Voltage changes of 1% to 2% can cause visible flicker.
Wind turbines or photovoltaic arrays can cause voltage fluctuations due to a number of natural
weather and machine conditions, including fluctuations in wind speed, cut-in or cut-out
transients, tower shadow effects, wind shear, and pitching errors of blades for wind plants, and
cloud cover passing overhead for solar PV. Another factor in flicker production is the type of
11-15
10581090
wind generator deployed: induction with fixed or variable poles, variable-speed through
electronic power conversion with or without pitch control, directly connected synchronous
generators, or synchronous generators connected through a converter. While these issues may not
be extremely significant when examined from a system perspective, at a local level they can
cause substantial voltage fluctuations.
11.3.2 Uncertainty
As well as being variable, most variable generation (with the exception of tidal power) is
unpredictable. The degree of predictability varies with technology; wind power can be well
forecast for the next few hours, but beyond that is difficult to predict. Solar PV can be difficult to
forecast over one hour for smaller arrays. This uncertainty of output increases the need for
regulating reserves, and should also be considered when making unit commitment decisions. The
likely forecast error, often measured as mean absolute error (MAE) or root mean square error,
needs to be accounted for by including additional flexibility in the commitment, so that forecast
errors can be covered by plants in an economical and reliable way.
The prediction error for VG increases with time–forecasts are less accurate a day ahead than an
hour ahead. As with variability, uncertainty is decreased when VG is spread out over larger
areas. The main prediction errors tend to be for large ramps, which are decreased when a larger
area is considered since there is not perfect correlation between the output of plants in different
areas.
11.3.3 Limited Reactive Power Control
Depending on the type of generator, reactive power supplied or absorbed by a renewable

generation facility may vary along with the real power output. For example, most wind turbines
use induction rather than synchronous generators and therefore require external capacitors to
provide reactive power, usually switched in discrete steps. Even with switched capacitors,
reactive control to match wind power generation may still be an issue. Externally switched
capacitors will probably satisfy the steady-state reactive power balance requirement for the
system but will not address any dynamic VAR control needs. Dispatching reactive power to
maintain transmission network voltage could place new requirements on controlled generating
facilities or necessitate capital investment for reactive compensation equipment. The addition of
power electronics into the wind system can provide both dynamic and static reactive power
control.
In the case of inverter-connected solar PV, many systems do not have reactive power control.
The reactive power output varies depending on line voltage and solar output. As shown in Figure
11-8, this variation can be quite different than the real power output and is expected to require
some form of utility control as penetration levels continue to increase.
11-16
10581090
Figure 11-8
Example of Interaction between Real and Reactive Power from a PV Inverter
11.3.4 Distributed Generation
Wind, solar, and other renewable generation are deployed both as large projects interconnected
to the transmission system and as small distributed facilities connected to the distribution system.
Even though wind is considered primarily as a bulk-generation resource, a growing number of
installations are being deployed on distribution systems at or near smaller load centers. Because
distributed systems are designed for one-way power flow, distributed generators require special
arrangements for connection and protection.
Two types of distributed wind and solar applications are common. Some facilities use local
distribution to collect energy from individual units and deliver it to the distribution system. Other
facilities are more dispersed and are individually connected to the system. Integration issues that
may surface in these distribution scenarios include voltage regulation, protection coordination,
and interferences or harmonic flows along rights-of-way. In general, these are the same
integration concerns that occur when generation is added into the distribution system.
11.3.5 Remote Location of Some Renewable Resources
Although good quality renewable energy sites exist throughout the United States, the best wind
and solar resources are located relatively far from large load centers. In addition, sites selected
for building large wind farms are likely to be away from population centers. This is not unique to
wind. For example, other generation types such as hydro and coal-fired stations may also be
located at significant distances from load centers. However, wind developments have tended to
not consider the available transmission capacity to these sites nor include the transmission-
related costs of taking the wind to market.
11-17
10581090
11.3.6 Integration of Wind Power
In some regions, such as Northern Europe and the upper Midwest and California in the United
States, relatively high penetration levels are managed via a strong existing transmission
backbone and by relying on other controllable generation, such as gas turbines and hydropower.
Other regions with less flexibility have experienced forced curtailment of wind generation, such
as transmission congestion in west Texas, transmission and distribution line voltage control
issues along the Buffalo Ridge in Minnesota and northwest Iowa, and challenges for grid
operators to follow load and manage ramp rates in territories such as the California ISO, New
Mexico and the Big Island of Hawaii (EPRI Wind Integration Technology Assessment and Case
Studies, 1004806, March 2004). Curtailment will generally happen for one of two reasons: lack
of flexibility in the generation fleet (plant cannot be ramped down or turned off quick enough to
accommodate wind), or congestion in the transmission network.
As penetration levels of VG increase, so does the challenge of managing this resource. Wind
penetration levels are already 10% to 15% of capacity in some areas of New Mexico and
California. Even higher penetrations have occurred on islands and in European grids. With
higher penetration, the response of wind turbines to weather and power system transients have
created operating problems. Over-speed tripping due to storm fronts and under-voltage tripping
due to power system faults have led to some significant tripping events, as illustrated by Figures
11-9 and 11-10.
Figure 11-9
Loss of Wind Turbine Output Due to Transmission Fault and Reconnect
11-18
10581090
Figure 11-10
Example of Wind Plant Output Variation on Area Control Error (ACE) (Courtesy of Public
Service of New Mexico)
11.3.7 Integration of Solar Power
Compared to wind, solar energy may be easier to integrate into the electric grid. Good solar
resources are more often close to load centers and are more likely to be available during the daily
load peak. One exception is the rapid rate of change in output that may occur as clouds pass over
solar arrays.
The energy resource potential for photovoltaic electric generation is very well distributed.
Although some areas like the Southwestern United States, Spain, Northern Africa and Australia
have excellent solar resources, the range of the best to the worst locations for solar is much
narrower than that of most of the other renewables. In the United States, the best locations are
less than twice as sunny as the worst. This point is illustrated in Figure 11-11, which shows the
relative cost of solar electric energy for a number of U.S. cities based on resource availability.
By its nature, solar energy is more available during the summer than winter, as shown in Figure
11-12. For most areas with summer peaking of demand, this profile is a good match with load
requirement. Even so, seasonal and daily variations of available output affect capacity factor and
overall system economics.
11-19
10581090
Figure 11-11
Relative Cost of Photovoltaic Electricity Due Only to Resource Variability
Figure 11-12
Monthly Solar Energy Variation Over One Year
As shown, solar photovoltaic power output can be quite variable and unpredictable due to partly
cloudy weather and storm fronts. This has been observed in desert systems as well as those
located in more temperate climates. Significant changes in solar system output can occur as
clouds pass by because these systems do not normally have any means to store energy. The ramp
rates of solar PV are likely higher than those of wind generation, as shown in Figure 11-13.
11-20
10581090
Figure 11-13
Daily Variation of Solar PV Output over One Month
Good resource locations for central-station solar thermal energy are not as well distributed as
those for solar photovoltaic. Solar thermal technology works best in dry desert climates such as
the U.S. Southwest or southern Spain. These areas are generally remote from populations,
requiring transmission right-of-way. Load factor can be an issue, in that the plant’s rated output
is available only about 20% of the time. However, an advantage of solar thermal over both solar
photovoltaic and wind power is the potential to add thermal energy storage (TES) without
significant cost. Solar thermal plants feature a built-in thermal inertia of around 30 minutes and
can be configured to add two to six hours of TES, providing better coincidence with late
afternoon load peaks. Another way to improve the load factor on transmission lines is to combine
wind and solar on the same line. Generally, hot sunny weather is not coincident with windy
weather in dry desert regions.
Together, distributed and central-station solar offer a practically unlimited energy resource.
Although it is not clear when the cost and efficiency of solar will become competitive with those
of conventional resources, the trend in that direction is well established. Venture capital
investment in solar has reached all-time highs in the past few years and a number of potential
breakthrough solar technologies are being actively pursued.
Widespread deployment of solar would certainly challenge both retail and wholesale grid
operations. Some studies estimate that such challenges could manifest as early as 2020 in some
parts of the country. (See EPRI white paper Distributed Photovoltaics: Utility Integration Issues
and Opportunities, August 2008, 1018096, as well the prior Western Governors’ Association
Clean and Diversified Energy Initiative Solar Task Force Report, completed January 2006.)
11-21
10581090
11.4 Variable Generation Impacts and Challenges
The technical issues and opportunities for integrating variable renewable generation differ
depending on its point of connection to the power system. The key question is: What grid
changes are needed to enable large-scale deployment of distributed renewable generation and to
integrate with other load and generation resources? These changes will vary depending on low-,
medium- and high-voltage points of connection to the power system. The appropriate strategy
also depends on the penetration level relative to the power system capacity at the point of
connection. The rules, concerns, and potential paybacks all vary at different system levels and
have been treated separately in the past. This point is illustrated in Table 11-4.
Table 11-4
Distributed Power System Performance Expectations at Various Connection Points in the
Electric System
DG Expectations System Integration Local and System

Interconnection Rules
and Connection Concerns Values or Payoff
Connection at Power, heat, load

End Use Feeder-level issues such control, quality and
Local connection
(Low Voltage) as power flows, protection, reliability.
requirements, e.g.,
and voltage impacts, e.g.,
IEEE 1547 and
Connection at issues related to high
derivatives. Ancillary service support
Distribution penetration levels.
(Medium Voltage) to utility T&D, e.g.,
reserve capacity,
Connection at Understanding system demand response,
Special grid rules for
Transmission response for planning and expansion deferral, etc.
generators <20 MW.
(High Voltage) analysis of scenarios.
11.4.1 Transmission Connected Generation Impacts
This section addresses renewable generation integration issues and costs with regard to power
system planning and operation. It is based on information presented in the EPRI reports, Wind
Power Integration Assessment and Case Studies (EPRI 1004806, March 2004), Wind Power
Integration: Smoothing Short-Term Power Fluctuations (EPRI 1008389, March 2005), Wind
Power Integration: Energy Storage for Firming and Shaping (EPRI 1008388, March 2005) and
Survey of Wind Integration Study Results (EPRI 1011883, March 2006).
Transmission is normally subject to the control of system operations and balancing authorities,
which in deregulated markets is the Independent System Operator (ISO). In vertically integrated
utility markets it is usually the utility that owns most of the transmission and/or generation in the
region.
11-22
10581090
Transmission systems are designed and built for two-way power flow. Therefore, if capacity is
available at the location of connection, renewable generation variability can be absorbed by the
grid. However, with traditional generation, system operators assume that most of the available
energy resources are “dispatchable” and therefore expect to be able to schedule generator output
with a high probability of delivery of the specified output. For wind, solar and other intermittent
generation, this assumption is less valid.
Depending on the location and the relative size of renewable output, absorbing variable
generation ultimately leads to increased cost related to the impacts of variability, uncertainty,
fluctuating output, lack of voltage control, and remote location on transmission scheduling,
system dispatch, network stability, load following, and load balancing.
The impacts of wind generation on both technical operating limits and ancillary costs are
summarized in the EPRI Report, Survey of Wind Integration Study Results (EPRI 1011883,
March 2006), covering North American sites. A related report also covers European sites (EPRI
1012734, March 2007).
The North American report reviews wind integration studies conducted for nine mainland
systems:
• NYSERDA-NYISO
• Xcel North-280 MW
• Xcel North-1500 MW
• Alberta Electric System Operator (AESO)
• Bonneville Power Administration (BPA)
• Southwest Public Service Company
• WE Energies
• Great River Energy
• PacifiCorp.
Wind generation penetration levels considered in these studies range from 0.3% to 20% of
system peak load. Table 11-5 presents the wind integration technical impacts determined in each
of these studies, and presents the wind integration cost impacts. These summary tables provide a
high-level view of the individual study results and allow for comparisons among the system
studies requirements. The data show that most of these studies have not assessed the impacts of
wind generation on contingency reserves. For the two studies that did make this assessment, the
results are completely different. The NYSERDA study found no impact and the AESO study
showed an approximately 134% increase in contingency reserves. There are certainly not
sufficient data to suggest any trends in this particular area. However, this is not the case for
regulating reserve impacts, as many of the studies assessed the impact of wind power on
regulating reserve requirements. The majority of the study results show that wind penetrations of
up to 20% lead to regulating reserve requirements of 30% or less.
11-23
10581090
Table 11-5
Summary of Integration Technical Impacts
Penetration Operating Reserve Load Following Scheduling Reliability

Study
% Regulating ACE Contingency Intra-hour Inter-hour Uncertainty Cap. Credit
NYSERDA-NYISO 10 +17% Minimal No impact +3% +6% +19% to +44% 15%
Xcel-280 0.3 +4% Minimal n/a n/a +0% n/a n/a
Xcel-1500 15 +12.9% Minimal n/a +15.6% +5% n/a 27%
AESO 13 +49% n/a +134% +49% +134% n/a n/a
BPA 11 n/a n/a n/a 0 to +3.5% n/a +17.2% n/a
SPS 20 n/a n/a n/a n/a n/a +49.6% n/a
WE 14 +9% n/a n/a +9% n/a +38% n/a
GRE 16.6 +28% n/a n/a n/a n/a n/a n/a
Pacificorp 20 n/a n/a n/a n/a +17% to +22% n/a n/a
Table 11-5 also illustrates that the increase in intra-hour and inter-hour variability is relatively
low for most of the studies. The potential increase in intra-hour load-following burdens range
from 3% to 15% with the exception of the one 49% study result. Likewise, the potential increase
in inter-hour load-following burden was found to range from approximately 5% to 22%. Also,
the capacity credit allocated to wind generation is only addressed in two of the nine studies
presented here, with the credit ranging from 15% to 27% of rated capacity. The majority of study
results show that wind penetrations up to 20% lead to regulating reserve requirements of 30% or
less.
Table 11-6 shows that the associated intra-hour load-following cost impacts are also relatively
low, ranging from 0 to 40 cents/MWh of wind. The inter-hour load-following costs are generally
more significant. For the studies where the inter-hour load-following costs were determined in
isolation, the cost impacts were $2.50–$3.50/MWh of wind generation. In other studies, the
inter-hour load following cost is embedded with the uncertainty costs determined from
scheduling/unit commitment assessments, but the costs appear to be in the same general range.
The associated cost impacts of the increased regulating reserve burdens vary from 20 cents to a
couple of dollars per MWh of wind. Much of the variation is associated with the assumed cost of
providing (or obtaining) regulating reserves.
Table 11-6 also shows that the increase in the standard deviation for scheduling uncertainty
ranges from 20% to 50%, again with the AESO study yielding the highest increase. The
associated costs shown in the table are in the $1.00 to $4.50/MWh range. Only one study yielded
costs in excess of $10/MWh. It should be noted that not all cost components were calculated for
all studies. Nonetheless, the studies that did evaluate cost impacts for most time frames yielded
costs in the same range.
11-24
10581090
Table 11-6
Summary of Wind Integration Cost Impacts ($/MWh)
Intra-Hour Inter-Hour
Penetration Scheduling/Unit
Study Regulation Load Load Total
Level (%) Commitment
Following Following
NYSERDA-NYISO 10 - - - -
Xcel-280 0.3 - 0.41 1.44 1.85
Xcel-1500 15 0.23 0.00 4.37 4.60
AESO 13 7.37 - 3.64 - 11.01
BPA 11 0.19 0.28 - 1.00 1.47
SPS 20 1.00 - 2.25 0.01 - - 1.01 - 2.26
WE 14 1.08 0.14 - 1.61 2.83
GRE 16.6 1.28 0.18 - 3.08 4.54
Pacificorp 20 - - 2.50 3.00 5.50
Figure 11-14 shows wind integration cost data as a function of penetration level. As expected
there is a spread in cost data relative to penetration which depends on location specific
circumstances and the mix of other generation resources in the area.
Figure 11-14
Wind Integration Costs
11-25
10581090
Since that report was issued, numerous other studies have been released, either studying new
areas or expanding the scope of previous efforts. The principal studies which add to the
knowledge of the above studies are the Eastern Wind Integration and Transmission Study
(EWITS) (NREL 2010) and the Western Wind and Solar Integration Study (WWSIS) (NREL
2010).
These reports were produced to examine the impact of high wind generation (and solar in the
case of WWSIS) on the operation of the two largest interconnections in the United States, the
western and the eastern interconnects. Using similar methods as before, these examined the
impact of large amounts of VG on the system as a whole. These studies showed the benefit of
increased balancing area cooperation, facilitated by an increase in high-voltage transmission
between areas of high VG resources (the Midwest for wind, the Southwest for solar) and areas of
high demand. Again, additional reserves were required, and there was a cost put on integration of
VG.
Figure 11-15 shows the impact on reserves in the EWITS study; integration costs for this study
were estimated at between $3.10 and $5.13/MWh, which places it within the values established
by previous works. WWSIS, while not explicitly specifying costs, estimates the impact to be
about the same as in the EWITS study.
Figure 11-15
Regulating Reserves from Eastern Wind Integration Study (Source: EWITS)
11-26
10581090
11.4.2 Distribution Connected Generation Impacts
Today’s electric distribution systems have evolved over many years, in response to load growth
and changes in technology. The largest single investment of the electric utility industry is in the
distribution system. Most common are radial circuits fed from distribution substations designed
to supply load based on customer demand requirements while maintaining an adequate level of
power quality and reliability, as shown in Figure 11-16.
Figure 11-16
Today’s Typical Distribution Feeder Topology [1]
Figure 11-16 shows how the system is designed to be fed from a single source. Protection is
based on time-overcurrent relays and fuses that use nested time delays to clear faults by opening
the closest protective device to a fault and minimizing interruptions. It is designed to safely clear
faults and get customers back in service as quickly as possible. In areas of high load density,
11-27
10581090
network systems are common. These systems are fed by multiple transmission sources, and
thereby provide high reliability. Both radial and network distribution systems have been
designed to serve load, with little planning for generation connected at these levels.
Circuit sectionalizing switches are manually controlled to restore load in unfaulted sections
downstream from a failure. The system voltage is maintained in compliance with American
National Standard Institute (ANSI) Std. C84-1, which specifies that service voltage be delivered
within 5% of the system rated voltage. These systems are generally considered to be ready to
support small photovoltaic installations without change, when the PV inverters meet appropriate
IEEE, UL, and FCC standards, and the overall penetration levels are very low.
The designs and technologies associated with today’s distribution systems impose important
limits on the ability to accommodate rooftop solar and other distributed generation, end-user load
management, distributed system controls, automation, and future technologies like plug-in
hybrid electric vehicles (PHEV). The system characteristics that lead to these limitations include:
• Voltage control is achieved with devices (voltage regulators and capacitor banks) that have
localized controls. These schemes work well for today’s radial circuits but they do not handle
circuit reconfigurations and voltage impacts of local generation well. This results in limits on
the ways circuits can be configured and important limits on the penetration of distributed
resources. It also limits the ability to control the voltage on distribution circuits for
optimizing customer equipment energy efficiency.
• There is little communication and metering infrastructure to aid in restoration following
faults on the system.
• There is no communication infrastructure to facilitate control and management of distributed
resources that could include renewables, other distributed generation, and storage. Without
communication and control, the penetration of distributed generation on most circuits will be
limited. In addition, the distributed generation must disconnect in the event of any circuit
problem, limiting reliability benefits that can be achieved with the distributed generators.
• There is no communication to customer facilities to allow customers and customer loads to
react to electricity price changes and/or emergency conditions. Customer-owned and
distributed resources cannot participate in electricity markets, limiting their economic
payback in many cases. Communications to the customer would also provide feedback on
energy use, which has been shown to help customers improve energy efficiency.
• The infrastructure is limited in its capacity to support new electrical demand such as home
electronics and PHEVs. These new loads have the potential to seriously impact distribution
system energy delivery profiles. Communication and coordinated control will be needed to
effectively serve this new demand.
At the same time, the distribution system infrastructure is aging, resulting in concerns for
ongoing reliability. Utilities are struggling to find the required investment just to maintain the
existing reliability, much less achieve higher levels of performance and reliability. New
automation schemes are being implemented that can reconfigure circuits to improve reliability,
but these schemes do not achieve the coordinated control needed to improve energy efficiency,
manage demand, and reduce circuit losses.
11-28
10581090
The bottom line is that today’s power distribution system has not been designed for distribution-
connected PV or other high penetration of distributed generation. In the past this was not an
issue, but today, with larger amounts of PV connecting to the electric system, we can expect new
challenges in how distributed and variable generation can be safely and reliably interconnected.
The following impacts will be seen when large amounts of PV are integrated at the distribution
level:
Steady-State Voltage Regulation: Since DER raises voltage levels when they inject power into
the grid, they may cause high voltage conditions at high penetration levels. This will have an
impact on the stability of the grid, and needs to be mitigated.
Voltage Flicker: Voltage flicker is a sudden change in voltage that occurs in seconds or
fractions of a second that can cause objectionable changes in the visible output of lighting
systems. The existing standards are likely adequate for high penetration of distribution-system-
connected PV.
Harmonics: Harmonics are distortions in the regular 60-Hz sine wave in North American power
systems. Too much harmonic distortion can cause adverse operation of customer and utility
equipment. Improvements over the last 10 years in the quality of inverter output have drastically
reduced issues with PV system harmonics. The existing standards are likely adequate for high
penetration of distribution system connected PV.
Unintentional Islanding: Utilities are regularly required to isolate a section of the power system
by disconnecting the section with network protectors or switches. Unintentional islands can be
established when a section or the grid is isolated from the substation supply while the load
continues to be maintained by an energy source within the isolated section that continues to
provide power. Unintentional islands pose a threat to proper utility system operation due to their
unsynchronized nature, hazard to public safety, and inadequacy to meet customer loads. The
IEEE 1547 standard and utility interconnection guidelines require PV systems connecting to the
network to have anti-islanding protection set to disconnect in the event that the network voltage
or frequency goes outside of predefined limits. As a result, unintentional islands are not currently
considered a significant concern. However, current anti-islanding techniques require PV systems
to drop offline rather than ride-through temporary faults, contributing to voltage drop or
frequency problems. Large amounts of PV could be prone to tripping during severe transmission
system disturbances that typically affect a wide geographical area.
In order to directly address the issues related to connecting large amounts of PV in the
distribution system, four key areas will need to be addressed:
1. Voltage regulation practices
2. Overcurrent protection practices
3. Grounding practices
4. Switching and service restoration practices.
11-29
10581090
Fortunately, due to the robustness of the existing design practices, the distribution system can
handle some level of distributed renewable generation without modification. The basis for
simplified interconnect rules such as California Rule 21 is that some level of robustness in
existing distribution design allows connection without detailed engineering studies. Standards
such as IEEE 1547-2003 and U.L. Inverter Test Standard 1741 evolved to enable connection
without major design changes to the electric system.
In most areas today, distributed renewable generation is treated as negative load and the usual
functions of generation are not expected or required. However, as PV deployments grow, feeder
cases that cross the threshold are expected, requiring changes in operation rules such that
distributed generation will need to provide voltage support and eventually energy balancing. This
is anticipated to occur initially on individual feeders with high penetration and later on at the
substation and sub-transmission level. Table 11-7 shows how the role and the rules for behavior
of distributed generation need to need to change with increasing penetration levels.
Table 11-7
Grid Penetration Scenarios and Changing Role of Distribution Generation
% of Generation • 2% • 10% • 30% 100%
Grid Penetration I. Low-numbers II. Moderate-level III. High-level of IV. PV operates

Scenarios and level of PV of PV with PV with capacity part time as an
with relatively stiff relatively soft grid of grid less than island or microgrid
grid connection connection the load demand
PV Impact and Very low, not Non critical, can Critical to power Primary power
its role in the significant to grid affect distribution delivery and source for
Grid operation voltage near PV meeting demand standalone
operation
Interconnection Non interference, Manage any local Engage PV for Rely on PV for
and Integration good citizen and distribution system stability and
Objectives compatible impacts operations and regulation
control
Rules / Standard IEEE 1547-2003 Modified 1547, New rules include Standalone rules
Operating current practice add network and operation and that are system
Procedures radial feeders penetration limits grid support dependent
requirement
Main Concerns -Voltage and -Interfere with -Availability -Availability

with Respect to current trip limits regulation
System Dynamic -Regulation -Load following
Grid Impacts -Response to -Recovery times provided
faults -Voltage control
-Islanding -Ramping
-Synchronization response -Normal and
-Coordination reserve capacity
-Interactions of
machine controls
… Transitions On- and Off-Grid…
11-30
10581090
11.5 Integration Technologies and strategies for Transmission-Connected

Renewable Generation
Renewable generation integration technologies include generator interface and output controls,
fault ride-through technology, auxiliary reactive power compensation equipment, integration
with hydro and pumped storage, other energy-storage technologies, and renewable generation
forecasting tools. Together with these technology advances that allow integration, multiple
strategies related to how the system is planned and operated can enable integration. As seen in
previous section, the primary issues for integration on the transmission system relate to balancing
supply and demand, which becomes more difficult with the additional variability and uncertainty
of VG.
The overall trend to more low- and non-CO2-emitting generation resources will challenge grid
operations. Some of the resources are less controllable. Wind power is the most striking
example. In the case of wind, solutions include better wind turbines, improved fault tolerances,
more accurate wind forecasting, application of power electronics for stabilization and
compensation, increased flexibility from conventional plant, reduction of variability through
diversity, demand-side response and electric energy storage. Some of these solutions are already
being applied in wind farm applications, while others are only in the design and development
stage. The role of various solutions stems from the basic planning and operating objectives of
utility power systems that were addressed in the previous section. The solutions are intended to
complement wind power capabilities and to fill in some of the gaps of wind power relative to
traditional generation.
11.5.1 Variable Generator Interface and Output Control
The generator type, interface, and output controls vary between renewable generation
technologies. The generator system for biomass, geothermal, and other renewable technologies
that generate steam to drive a steam turbine generator are similar to those for conventional fossil
and nuclear power plants. Those that use biogas to drive a gas turbine generator are similar to
conventional combustion turbines. Solar photovoltaic generators use inverters to convert dc
electricity to 50- or 60-Hz ac electricity. Wind turbines use various generator and interface types,
including asynchronous induction generators that are either connected directly to the grid or
through power electronics that convert ac to dc and back to 50- or 60-Hz ac electricity. Each
generator type affects the electricity system and can impact ramping burden, fluctuating output
power, and reactive power control performance in different ways. An understanding of the
differences in the renewable generator output control is essential for effective integration into
the electricity grid.
Except for solar photovoltaic power generation, which uses inverters to convert dc to ac
electricity, virtually all variable renewable generators are based on one of three generator-
connection types: 1) constant-speed induction generators with direct grid connection; 2) variable-
speed induction or asynchronous generators with power electronic interface to grid; and 3)
variable-speed synchronous generators with inverter interface to the grid. Each is described
further below.
11-31
10581090
Apart from these mainstream systems, many other designs and technologies have been
developed and tried over time. Some have survived, others have come and gone, and still
others are yet to be installed and proven in the field.
11.5.1.1 Constant-Speed and Semi-Variable Speed Induction Generator with Direct

Grid Connection
A constant-speed or fixed-speed wind turbine consists of a rotor and a squirrel-cage induction

generator connected to the rotor by a gearbox as shown in Figure 11-17. The generator stator
winding is connected to the grid. The generator slip varies with the generated power, so the
speed is not in fact constant. However, because the speed variations are very small (just 1% to
2%), this machine is commonly referred to as a “constant-speed” turbine. A squirrel-cage
generator always draws reactive power from the grid, which is undesirable, especially for weak
networks. For the megawatt-class wind turbines discussed here, the reactive power consumption
of squirrel-cage generators is therefore always compensated by capacitors, which are switched in
stages depending on VAR consumption of the generator.
Utility Grid
Gearbox
Shaft Speed ω Blades
Squirrel Cage Induction Machine
Switched
Capacitor
Banks
Figure 11-17
Constant-Speed Induction Machine Wind Turbine Generator with Gearbox
The 30-MW Peetz Table wind power plant in Northeast Colorado is a good example of a large
constant-speed wind turbine installation. This plant consists of 33 NEG Micon (now Vestas)
turbines rated at 900 kW and uses capacitor-compensated induction generators. The plant is
connected to the grid via a 34.5-kV to 115-kV substation (see Figure 11-18). The power from the
entire field is collected via underground direct-burial 35-kV primary cable. A 35-kV switched
capacitor bank is installed at the 50-MW substation and controlled by the Western Electric
Coordinating Council (WECC) SCADA system.
11-32
10581090
A variation of this type of turbine technology is the semi-variable-speed induction generator.

This turbine utilizes a wound-rotor induction machine with an external resistive rotor current
control that actively compensates for small variations in the machine’s speed. Similar to the
fixed-speed wind turbine, switched shunt compensation at each turbine is used to compensate for
the reactive power consumed by the generator.
Figure 11-18
Grid Interface for Peetz Table 30-MW Wind Farm
11.5.1.2 Variable-Speed Induction or Asynchronous Generator with Power Electronic

Interface to Grid
Variable-speed wind turbines have progressed dramatically in recent years. Variable-speed

operation can only be achieved by decoupling electrical grid frequency and mechanical rotor
frequency. To this end, power electronic converters are used, such as an ac-dc-ac converter
combined with advanced control systems.
11.5.1.2.1 Doubly-Fed Induction Generator
In a variable-speed turbine with a doubly-fed induction generator, the converter feeds the rotor
winding, while the stator winding is connected directly to the grid (Figure 11-19). The electrical
rotor frequency can be varied by this converter, thus decoupling mechanical and electrical
frequency and making variable-speed operation possible.
The power electronics are typically rated at 30% of the generator rating, thus leveraging the
investment in electronics. Control advantages include better wind power tracking, smoother
output power, more constant torque, and reduction in torque surges.
11-33
10581090
This type of wind turbine generator (WTG) offers the simplicity of the induction generator and
the advantages of electronic control. The power electronics can benefit wind integration in rural
and more isolated grids by maintaining system stability, reducing the risk of voltage collapse,
and minimizing the impact of grid disruptions. The design is currently very popular and
machines are available in sizes up to 4.5 MW from a half dozen manufacturers. Competition is
steep in Europe, where all six manufactures are engaged. However, in North America, GE Wind
(formerly U.S. Windpower, Kenetech, Zond, and Enron) has dominated sales by threat of U.S.
patent infringement lawsuits. At least four patents were issued on this variable-speed wind
turbine design from January 1991 through July 2003. The original patent by Richardson and
Erdman [10] was assigned to U.S. Windpower, Inc. Within the last few years, however, other
manufacturers such as Enercon, Vestas, Gamesa, and RePower have executed either cross-
licensing agreements or patent agreements with GE Wind. In addition to GE, Gamesa and others
are currently offering this technology in the United States.
P, Q (stator) Gearbox
750 kW wound-
rotor induction
generator
Shaft Speed ω Blades
P (rotor/converter)
Power Power
Converter Converter
(line side) (machine side)
iabc(rotor)
Switch
Control
i*abc(rotor)
Pgen ,Qgen
ω
Rotor Current Torque
Computation Computation
T*
Lookup Table
(T vs. ω)
Figure 11-19
Variable-Speed Doubly-Fed Induction Machine-Connected WTG with Gear Box
11-34
10581090
11.5.1.2.2 Full-Converter Synchronous or Induction Generator
A full converter means that all the energy from the turbine is processed via power electronics.
Advances in power electronic technology have made the full converter a practical reality. The
approach completely decouples the turbine by means of a power electronic converter and allows
a wide range of variable-speed operation. In some cases, a direct-drive multi-pole generator is
used, which eliminates the need for a gearbox (Figure 11-20). The direct-drive approach results
in obvious maintenance and cost savings. However, generator speed is also reduced, requiring
additional electrical poles in the generator that leads to a relatively large machine resembling a
water wheel in a hydroelectric plant. Alternatively, another U.S. manufacturer, Clipper Wind
Power, offers a slightly different approach. The Clipper WTG is based on a distributed drive
train approach, consisting of four permanent-magnet synchronous generators connected to the
grid via a four-quadrant IGBT switched inverter.
dc ac
to
Grid
ac dc
Figure 11-20
Variable-Speed, Direct-Drive Synchronous Machine Converter-Connected WTG (No
Gearbox)
In a full-converter machine, power electronics are sized for the full and transient overload
rating. The topology usually includes two stages: first to rectify the variable-speed output of
the generator, and second to produce ac power for interfacing with the electric grid, as shown
in Figure 11-21. These turbines are available in sizes up to 6.0 MW from three manufacturers
and provide a full range of output control, including both real and reactive power. In a
variable-speed turbine with direct-drive, the generator and the grid are dynamically isolated.
Overall, variable-speed wind turbine generators experience less mechanical stress. In addition,
rapid power fluctuations occur only rarely because the rotor acts as a flywheel or buffer by
temporarily storing energy. In general, no flicker problems occur with variable-speed turbines.
The electronic interface allows for fast reactive power control and compensation for grid voltage
11-35
10581090
variations. With VAR control, it is possible to decouple wind fluctuations and other system
events and the prime mover, thus improving overall performance. Although power electronics
make the generator more expensive, variable-speed operation can lead to major savings
elsewhere, such as lighter-weight mechanical components and lighter foundations in off-shore
applications.
Permanent Rectifier Inverter

magnet or stage stage
Synchronous
generator
Utility grid
Wind Turbine
Figure 11-21
Direct-Drive Synchronous Generator with Simple Rectifier at the WTG and Voltage-Source
Converter Interface with the Utility Grid
11.5.1.2.3 Variable-Speed HVDC-Link to Grid
With the advent of off-shore and potentially large-scale wind farms, the high-voltage dc
(HVDC) link is being considered as a grid interface. Advances in power electronics technology
have led to the development of HVDC systems at lower power ratings than were previously cost-
effective. For instance, one manufacturer offers a system in the range of 2 to 200 MW based on
IGBT voltage source converters, as shown in Figure 11-22. One advantage of voltage-source
self-commutated converters over traditional thyristor-based HVDC current-source converters is
that synchronous rotating machines are not required at each end of the link. The ac connection
voltages at both ends do not have to be the same, possibly saving a site transformer at the shore
[11].
11-36
10581090
Figure 11-22
One Leg, Single-Line Diagram of 12-Pulse HVDC System (Source: Siemens)
Figure 11-23 shows three alternative configurations for off-shore connection of wind plants to
substations on shore.
Figure 11-23
Alternative Configurations for Connecting Off-Shore Wind via DC-Link (courtesy of Barrow
Off-shore Wind)
Although HVDC systems are available from at least three manufacturers, to date only a
demonstration system has been installed, and there has yet to be an off-shore installation. The
technology is in its infancy and further advances are likely. In the longer term, it is possible that
off-shore turbines will be directly connected to high-voltage dc without conversion to ac. In this
scheme, a platform is required to house the off-shore converter and switchgear, and the whole
wind farm would be lost if the cable fails. So far, studies indicate that the HVDC connection is
currently too expensive for distances closer to shore than 25 km and for power levels less than
11-37
10581090
200 MW [12]. There is also a risk associated with applying new HVDC technology off-shore
before it has been proven in on-shore applications.
11.5.2 Ride-Through During Faults and Other Grid Disturbances
Weather events and accidents that result in switching events, lightning strikes, equipment
failures, and downed power lines are a fact of life on utility grids around the world. Protective
devices at the grid interface of wind turbines are intended to protect both the turbine from grid
disturbances and the grid from either poor-quality generation or from unwanted islanding fed
by the wind generator. A practical problem has emerged where turbine protection trips the unit
off-line during momentary system faults but the operating utility prefers that the generation
remain on-line and ride through the brief disturbance. Better coordination of the protection
settings and more immunity in some power electronic-connected turbines are needed.
Several techniques have been developed to ride through momentary voltage dips caused by faults
and/or switching events. These techniques can reduce wind power intermittency and the potential
rapid loss of generation (ramping burdens) due to tripping. In addition, they help with remote
location exposure to power system disturbances and issues related to the protection and tripping
of wind generators.
11.5.2.1 Adaptive Protective Relaying
Adaptive relaying is an application of the concept that protection timing and trip limits for
generators and feeders should vary depending on power system conditions. In this way, the
cutoff of a distributed generator may be advanced or delayed to minimize circuit and generator
disruption. Because of the very short decision time, relays are either preprogrammed
to sense conditions and act accordingly, or a local intelligent agent that supervises and controls
the response is needed.
11.5.2.2 Low-Voltage Ride-Through Circuits
One drawback of variable-speed wind systems is that their built-in power electronics are
relatively sensitive to grid disturbances compared to an induction or synchronous generator
interface without electronics. On the other hand, the power electronic interface offers a faster
and more controllable connection than conventional machines. The result is that a perceived
weakness has been turned into an advantage by the addition of low-voltage ride-through
capability to electronically interfaced wind machines. The ride-through is feasible for both full
converter and double-fed machines and for momentary low voltages that last less than 15 cycles.
This can be a very practical solution because 80% to 90% of low-voltage events are momentary.
The key is to limit the high current in the rotor to protect the converter and provide
a bypass for the current via a set of resistors that are connected to the rotor windings. This
system makes it possible to ride through neighboring faults or faults cleared downstream without
disconnecting the turbine from the grid. Because the generator and converter stay connected,
synchronization is maintained during and after the fault, and normal operation can be continued
immediately after the fault has been cleared. As an additional feature, reactive power can be
11-38
10581090
supplied to the grid during long dips in order to facilitate voltage restoration. A control strategy
has been developed that takes care of the transition back to normal operation. Without this
control strategy, large transients would occur.
GE Wind was the first to offer low-voltage ride-through (LVRT) capability on its 1.5-MW
doubly-fed induction generator. The LVRT option allows the machine to stay connected to the
grid during grid faults, thus keeping wind farms on-line by feeding reactive power during system
events. It can be programmed to align with the operational/control parameters of the host
transmission system. The ride-through option can be effective even below 30% voltage at the
point of common coupling, as shown in Figure 11-24. The system remains engaged until after
the fault is cleared, providing support to bring the system back to normal operating conditions.
Figure 11-24
A Severe Voltage Sag to Less Than 30% Voltage Remaining Voltage for 100 ms
(Upper Trace) and Wind Turbine with LVRT Recovering After the Event (Lower Trace)
The reported performance of the GE system is consistent with laboratory and simulation studies
by several researchers. Torbjörn Thiringer [13] at Chalmers University of Technology, Göteborg,
Sweden found that a full-power electronic converter can be set up to ride through a voltage
disturbance with virtually no dynamic interaction and a preprogrammed response. In his results,
a doubly-fed induction machine responded like a fully electronic interface for small voltage sags
and like an induction machine for large sags. Figure 11-25 shows typical machine response to a
large voltage sag for three types of commercially available wind turbine generators, including
fixed-speed, semi-variable-speed, and variable-speed. The response of each turbine to the voltage
sag differs considerably. The oscillations damp out within seconds after the event.
11-39
10581090
Over the years, utilities have begun requiring wind plants to remain online during transmission
system events. As a result, wind turbine manufacturers have begun integrating low voltage ride-
through technologies. With the recent FERC Order 661a, wind plants larger than 20 MW are
required to remain online during voltages down to 0% voltage and for up to nine cycles.
Variable-speed wind turbines are capable of meeting such requirements without additional
equipment. Other technologies, such as constant-speed induction generators, require additional
compensation to sustain voltages within the wind plant to acceptable levels.
Figure 11-25
Response of Four Commercially Available Wind Turbine Generators to a Large
Voltage Sag
11.5.3 Reactive Compensation Equipment
As with any generation technology, wind-driven generators need reactive power to excite the
generator field and control the voltage at the output terminals. Induction machines, which have
been most popular for wind generation to date, require an external reactive power source. This
source of reactive compensation is usually provided by lumped capacitor banks that are
connected when the generator operates. Compensating capacitor banks can also be switched in
stages relative to the power and system voltage levels. Electronic compensation provides the
added benefit of being infinitely variable and usually more responsive than permanent and
switched capacitor banks.
Benefits that can be expected from auxiliary compensation equipment include providing needed
levels of reactive power control, supporting local voltage regulation, and helping with control
of distributed collection. In addition, fast-acting electronic-controlled compensators may be
responsive enough to mitigate fluctuating wind power output and related flicker.
11-40
10581090
There are four basic types of reactive compensation/voltage control options that can be utilized
to meet specific power factor/voltage regulation requirements:
• Distributed Constant Power Factor
• Distributed Variable Power Factor
• Centralized Switched Capacitor Banks
• Static VAR (STATCOM, DSVC, etc.)
Figure 11-26 provides an illustration of each of the four approaches.
Figure 11-26
Illustration of Reactive Compensation and Voltage Control Mechanisms for Wind Farm
Applications
11.5.3.1 Distributed Constant Power Factor
This type of compensation refers to the mechanically-switched capacitor banks utilized at each
wind turbine generator, set to operate at a pre-defined constant power factor over the full range
of generator output. It is worthwhile to note that the effectiveness of this approach depends upon
the collection system (lines, cables, and transformers). A WTG power factor set point of 0.99
leading (generator convention) does not result in a 0.99 leading power factor at the point of
common coupling (PCC). The reactive power drawn by the substation and WTG transformers
and collection lines must be taken into account as well. Depending upon the X/R ratio of these
components, the impacts could vary significantly.
11-41
10581090
If necessary, switched capacitor banks located at the substation are commonly utilized to provide
unity power factor at the PCC.
11.5.3.2 Distributed Variable Power Factor
Some turbine vendors offer Distributed Variable Power Factor, which provides plant-wide
control of each WTG’s reactive power production/consumption. This approach takes advantage
of the reactive power capabilities of each turbine to control plant-wide voltages and/or power
factor. The advantages to this approach include voltage regulation throughout the wind plant and
the PCC, and controllable power factor at the PCC.
The effectiveness of this approach to regulating the power factor and/or voltage at the point of
common coupling is plant-specific. The power losses of the electricity collection system
determine how many VARs can be injected into the PCC. The maximum amount of reactive
power injection at the PCC is equal to the total reactive compensation minus all collection
system reactive power losses. This is illustrated in Figure 11-27, which shows the effective
impact of collection system losses on the aggregate power factor at the substation for an example
wind farm.
0.5
0.3
Reactive Power (per-unit)
0.1
-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
-0.3
-0.5
Wind Plant (w ith collection system)

-0.7
Wind Plant (no collection system)

-0.9
Active Pow er (per-unit)
Figure 11-27
Power Factor Curves for Example Wind Plant
This limit may be further reduced due to the voltage profile seen along the collection system
feeders. Turbines inject reactive power into the system up to the normal operating range
of the terminal voltage. The same applies for absorbing reactive power under low-voltage
conditions. Therefore, the amount of reactive power injection or support at the PCC is
11-42
10581090
determined by both the reactive power losses and the WTG voltage profiles. Transformer tap
settings also play a significant role in the reactive power curve for a particular plant. Detailed
analysis is necessary to determine a plant’s full reactive power capability.
11.5.3.3 Substation Switched Capacitor Banks
Large, switched capacitor banks installed at the interconnect substation are typically utilized
in conjunction with switched banks at the WTG terminals. The advantage of this approach is
its relative cost effectiveness and ability to regulate the voltage and power factor at the point
of common coupling. These capacitor banks are usually sized to maintain unity power factor on
the primary side of the substation transformer. Optionally, capacitor banks can be controlled by
the utility system operator to switch in or out based upon local system voltages as well.
Note that switched capacitor banks are less effective in extremely low-voltage conditions due
to the much lower reactive power injection resulting from the lower voltage (reactive power is
decreased by the square of the voltage: Q=V2/X).
Switching can be either mechanical, using contactors or circuit breakers, or electronic

using thyristors. Figure 11-28 shows an example application in India. This reactive power
compensation system includes 600 kVAR of low-voltage thyristor-switched capacitors and 300
kVAR of medium-voltage fixed capacitors for a 2.8-MW (7 x 400-kW WTGs) wind farm. In this
case, thyristor switching provides for a smoother transition of the low-voltage units.
Figure 11-28
Fixed and Switched Capacitor Bank Supporting a 2.8-MW Wind Farm in India
(Source: Crompton Greaves Ltd.)
Switched capacitor banks can have practical drawbacks. One is that their discrete size
with limited steps and deployment in one location make compensation less than optimal.
Frequent mechanical switching of the bank may create transient disturbances on the system. The
control and response of the switching cannot effectively respond to rapid demands for reactive
power during system transients.
11-43
10581090
Switched reactive compensation has become a standard component of wind power plants that
employ induction generators. In conventional generation plants (fossil, hydro, and nuclear), the
reactive compensation problem is solved by using a synchronous machine, which has built-in
reactive power control provided via the generator field current control. The synchronous
generator is more difficult to apply in wind power generation due to the variability of wind
turbine output power and speed.
11.5.3.4 Power Electronic Custom Power Devices
When power electronics are a part of the wind turbine generator, the issue of local reactive
compensation is usually resolved. This is the case for both doubly-fed and full-converter wind
generators. Both of these converter systems have fast-acting VAR control capability because
they employ power electronic conversion to interface the variable-speed turbine with the
constant-frequency grid. And reactive compensation is one of the key advantages of these
variable-speed conversion systems. However, in most existing wind-farm applications, the
WTGs are not variable-speed machines, and this built-in reactive compensation may not be
present.
Auxiliary equipment that combines power electronics with capacitors is another solution to
reactive compensation and VAR control in these plants. This technology involves devices and
circuit configurations of power electronic equipment used in distribution systems rated 1 kV
through 38 kV. The IEEE has identified this technology area as custom power and recently
published a standard to define the different configuration and applications [14].
Custom power devices include static switches, inverters, converters, injection transformers,
master control modules, and/or energy-storage modules that have the ability to perform current-
interruption and voltage-regulation functions in a distribution system to improve power quality.
As shown in Table 11-8, these individual devices may have more than one function, including
sag/swell, interruption, reactive power, and harmonic compensation. Voltage regulation is also
a direct result of reactive or harmonic compensation.
For wind applications, sag/swell and reactive compensation devices may be used for
plant-level reactive compensation. Improved voltage regulation can be obtained from several
of these custom power technologies, and a fast response capability may also enhance system
stability. The most likely devices for wind plant application are the static shunt compensator
(DSTATCOM) and the distribution static VAR compensator (DSVC). Devices that provide fast
voltage regulation and voltage-sag compensation such as the static voltage regulator (SVR, also
called dynamic voltage restorer or DVR) and the static series compensator (SSC) are also
possible solutions.
11-44
10581090
Table 11-8
Types and Functions of Custom Power Devices per Draft IEEE Standards Project 1409
Sag/Swell Interruption Reactive (Flicker) Harmonic

Mitigation Device
Compensation Compensation Compensation Compensation
Static Series Compensator (SSC) X
Static Voltage Regulator (SVR) X
Static Transfer Switch (STS) X X
Backup Stored Energy System Devices (BSES) X X X
Static Circuit Breaker (SSB) X X
Static Shunt Compensator (DSTATCOM) X X X
Distributed Static Var Compensator X
Distributed Series Capacitor X
Series/Shunt Capacitor X X X
11.5.3.4.1 Static Shunt Compensator (DSTATCOM)
DSTATCOM is a distribution-class static VAR compensator that can address the reactive
power management requirements of both wind systems and nearby dynamic loads. This
microprocessor-controlled device utilizes power electronic switches to instantaneously insert the
appropriate number of multi-stage power capacitors, providing reactive compensation on a cycle-
by-cycle basis. The device is often applied to weak systems to compensate for (relatively) large
loads (or wind generators) that cycle on and off and may create irritating flicker. A DSTATCOM
can control only voltage or a combination of voltage and power factor. In some cases, the
theoretical in-rush of an induction motor started on the system can be greater than the available
fault current at the point of common coupling. In this case, it is difficult to distinguish between
a turbine-start and actual fault. Therefore, fault duty and coordination of downstream breakers
must be considered in this application.
Figure 11-29 shows a DSTATCOM device that is utilized at a wind farm at Foote Creek,
Wyoming. The manufacturer calls it dynamic VAR (or D-VAR), and the technology is based on
a superconducting magnet. This system continuously monitors the collector bus voltage of a
wind farm to ensure that the voltage remains within the utility’s predetermined range.
Continuous voltage regulation is accomplished by a combination of VAR injection or absorption
from the dynamic VAR system and controlled switching of distribution capacitor banks. In
addition, the system mitigates voltage transients at the collector bus that typically originate on
the transmission grid and may otherwise cause the wind turbine generators to trip off-line.
Another type of DSTATCOM uses a voltage-source converter with a dc-link capacitor and
protects the distribution system from voltage flicker caused by the rapidly varying reactive
current draw of nonlinear dynamic loads. It thus enables such problem loads (or generators) to
co-exist on the same feeder with more sensitive loads. In effect, this device instantaneously
exchanges real and reactive power with the distribution system without the use of bulky
capacitors or reactors. ABB, Mitsubishi Electric, S & C Electric, Siemens, and others offer
DSTATCOM products. A shippable module installation example at a wind plant is shown in
Figure 11-30.
11-45
10581090
Figure 11-29
Conceptual Layout of a Dynamic VAR Device and Switched Capacitors Supporting a Wind
Farm (Source: American Superconductor)
Figure 11-30
DSTATCOM Installation (Source: S&C Electric)
11.5.3.4.2 Distribution Static VAR Compensator (DSVC)
A DSVC is a shunt-connected device with thyristor switching of passive-reactive components

(capacitors and reactors) in continuous and/or discrete steps. The device exchanges a
continuously-controlled reactive current with the power system to control electrical parameters
of the line at the point of common coupling. Typically, the DSVC equipment for distribution
applications is used for voltage control, reactive power compensation, and power factor
correction.
11-46
10581090
The DSVC provides ac voltage control by generating and absorbing reactive power using
passive elements. It can also be applied in balancing or unbalanced conditions. As shown in
Figure 11-31, it normally consists of one thyristor-controllable reactor (TCR) and a number
of thyristor-switched capacitor (TSC) branches. The value of the reactance of the inductor is
changed continuously by controlling the firing angle of the thyristors, while each capacitor is
switched on and off at the instants of the current zero crossings, thus avoiding in-rush currents
as well as undesirable switching transients. With this arrangement, the SVC can generate
continuously-variable reactive power in a specified range, and the size of the TCR is limited to
the rating of one TSC branch. Obviously, the size of the reactor limits the power that can be
absorbed in the inductive range [15]. There are various configurations that are classified as
DSVCs.
Figure 11-31
Simplified Schematic of a Static VAR (SVC) or Distribution Static VAR Compensator
(DSVC)
Figure 11-32 shows one example of such a device: the PureWave AVC, a DSVC made by S&C
Electric. Although it is typically applied for heavy commercial and industrial processes with
varying electric loads, it may also be effective for a small wind installation with rapidly varying
reactive power demand. This DSVC provides real-time voltage support for voltage sags and dips,
flicker mitigation, and power-factor correction. By reducing voltage fluctuations and managing
reactive power flow, the electric power system stability, reliability, and capacity are increased.
The DSVC can enable problem loads to co-exist on the same feeder with varying generation and
more sensitive loads, obviating the need for separate feeders. DSVC manufacturers include
ABB, Mitsubishi Electric, S&C Electric, Siemens, and Square D.
Figure 11-33 shows a field installation of a small DSVC at a wind site. This static VAR
compensator was installed in 2005 near a 50-MW wind farm in California. The plant includes 25
x 2-MW WTGs and the collection voltage is 34.5 kV. The SVC rating is -8 MVAR to +18
MVAR over a continuous range. The reactive compensation consists of two 4-MVar reactors,
11-47
10581090
two 6-MVar cap banks, and a 6-MVar adaptive VAR compensator (AVC). The equipment
regulates the voltage output on the 69-kV line.
Figure 11-32
Configuration of a Commercially-Available DSVC (Source: S&C Electric)
Figure 11-33
Distribution Static VAR Compensator (DSVC). The Reactive Compensation Equipment on
the Right Side of the Crestwood Substation Supports the 50-MW Kumeyaay Wind Energy
Project in California. (Courtesy of Alliant Energy)
11.5.4 Wind Energy Forecasting Tools
Wind energy forecasting is becoming increasingly important as installed capacity continues

to grow. Development of wind forecasting tools is focusing on improving forecast accuracy,
predicting high ramp rates of wind generation in response to rapid changes in wind speed, and
integrating the forecasts into grid operations.
Various contractual provisions have been devised to handle deviations between forecast and
actual power delivered. Better forecasting tools will directly help with both the intermittency and
ramping characteristics of wind power, with positive financial consequences. It is clear that more
11-48
10581090
accurate wind forecasts help both the buyer and seller of wind energy. From the point of view of
the seller, better forecasts minimize contractual penalties, and greater confidence maximizes the
value of the wind power. From the buyer’s perspective, accurate wind power forecasts reduce
risk and increase power system reliability.
One of the main costs of wind power intermittency is reserve requirements. This additional
reserve is required to maintain the same level of control performance in minute-by-minute
regulation and minimize any additional operating cost due to the balancing of intra-hour slow
variation. Expected limits in control and variability for wind power increase reserve
requirements. In the worst case, every unscheduled megawatt must be offset by some other
resource, generally at high cost.
Better forecasts can reduce the need for reserves and increase the value of wind energy. Many
different components go into developing an accurate wind power forecast. They include
numerical weather predictions, real-time and recent wind speed, direction, and energy
generation, physical models of wind flow, wind plant power curves, and artificial neural
networks and other sophisticated statistical techniques [16].
EPRI has been monitoring and evaluating the latest forecasting technologies for many years,
and worked with DOE and the California Energy Commission (CEC) to develop and test wind
forecasting systems in Texas and California [17] [18] [19] [20] [21] [22] [23] [24] [25]. Six wind
energy forecasting systems are of particular interest in the United States: the Prediktor Wind
Energy Forecasting system developed by Risoe National Laboratory; the eWind system
developed by AWS Truewind, LLC; the 3TIER Environmental Forecast Group forecast system;
the WindLogics system; the Garrad Hassan system; and the ISET Wind Forecast system from
Germany.
These forecasting systems are similar in some respects, but they use different techniques to
forecast wind speed, direction, and corresponding wind-energy generation. The model structure
typically blends physical and statistical models, with the blend dependent on the forecast
window, e.g. the forecast interval into the future.
Short forecast time intervals up to about six hours typically rely on real-time wind speed,
direction, and power generation data; rapid-update numerical physical model data; simple
persistence, and adjustments using self-learning statistical methods such as Screening Multiple
Linear Regression (SMLR), artificial neural network (ANN), and Support Vector Machines
(SVM) (Figure 11-34).
Longer forecast time intervals greater than four to six hours typically use numerical weather
prediction, physical wind flow models, and statistical techniques to predict weather conditions,
wind flow patterns near the wind turbines, and wind energy generation. The technology has
advanced to the point that the mean absolute error of same-day wind energy forecasts has
decreased to 6% to 8% and the error of next-day forecasts has decreased to about 15%.
11-49
10581090
Predict ors Predict and
P1 ,P2 ,... F
SMLR
ANN
Training
SVM Algorithm
F = f ( P1 ,P2 ,...)
Figure 11-34
Wind Predictions are Adjusted Using Self-Learning Statistical Methods (Source: AWS
Truepower [26])
Figure 11-35 is a schematic diagram of an example longer-term forecast model used by AWS
Truepower to generate 48-hour forecasts [26]. The longer-term forecast model typically involves
the following steps:
1. Download numerical weather forecast data from the National Center for Environmental
Prediction (NCEP) web site.
2. Calculate wind flow around wind plant using a meso-scale or wind-flow model.
3. Apply model operating statistics (MOS) to estimate adjusted wind speed and direction at
wind plant (Figure 11-36).
4. Estimate hourly power generation at 100% availability using wind plant model.
5. Apply Model Operating Statistics to adjust the hourly power generation forecast.
6. Adjust forecast to account for curtailment and forced outages of individual wind turbines (if
available).
7. Issue the 48-hour forecast of hourly wind speed and energy generation
During 2008, a consortium of the Xcel Energy Renewable Development Fund, the Minnesota
Department of Commerce, WindLogics, and the Utility Wind Integration Group completed a
major project to develop and test a state-of-the-art wind energy forecasting system for Minnesota
and address integration of the forecasts into grid operations [27]. The resulting forecast system
generates both day-ahead and hour-ahead forecasts at each of 17 nodes of the electricity grid.
The forecasts are then aggregated to provide the total forecast of wind energy generation for the
state.
11-50
10581090
Figure 11-35
Schematic of Typical Wind Energy Forecasting System (Source: AWS Truewind [26])
Figure 11-36
Graphical Presentation of Wind Flow Data from a Numerical Weather Prediction Model
(Source: AWS Truewind [26])
11-51
10581090
Although forecast accuracy has improved and commercial forecasting systems are being applied
throughout the world, there are still several issues that stand in the way of further reductions of
wind forecast error. They include 1) the general unavailability of robust real-time wind speed,
direction, and generation data for individual wind plants and for regional wind generation
resources; 2) the unavailability of operational status data for individual wind turbines required to
adjust forecasts for machines that are shut down or curtailed; and 3) methodologies to accurately
forecast rapid changes of wind speed and generation and that result in high ramp rates of regional
wind generation.
The need for more accurate forecasting of high ramp rates up and down has become especially
important because the combined output of large concentrations of wind power plants can change
quickly and affect operation of the grid. Due to the complexity of weather impacts on wind plant
output, ramp rate forecasting is particularly challenging. The factors affecting ramp forecasting
include passage of weather fronts and storms that cause sudden changes in wind speed and
direction, large-scale eddy currents that can’t be predicted, and variations of atmospheric
stability that affects local wind patterns.
As the installed wind capacity continues to grow, the value of more accurate wind energy
forecasts will increase. Thus, there will be a large incentive to further improve forecast accuracy
in future years.
11.5.5 Increased Diversity through Increased Transmission and BA Cooperation
As mentioned earlier, many integration studies show the benefit of diversity when integrating
VG. Increased diversity in the area in which VG is installed reduces variability and uncertainty,
making VG easier to manage. For example, while wind power can be very variable at one wind
plant, the variability of 10 wind plants will be significantly less. Considering a larger area
increases the flexible resources that can be used to deal with wind power variability, as well as
provide inertia. Therefore, strengthening the transmission network allows VG to be accessed and
its total impact reduced. This would also require increased balancing area (BA) cooperation to
ensure the flexible resources on the system as a whole are well utilized to minimize the effect of
variability. Instead of each balancing area needing to manage its own variability, increased
cooperation would allow a sharing of reserves needed to manage variability. The total reserves
needed over all regions would thus be decreased, and the flexible resources needed could be
reduced. A good example of this can be found in Europe, where Denmark, with very high wind
penetration, is able to access flexible resources—particularly hydro—in neighboring Norway to
mitigate variability effects.
11.5.6 Flexibility from Conventional (Non-VG) Plant
Increased demand for flexibility from the conventional fleet can be met by building new
technology, such as energy storage, and can also be managed, to a certain extent, by existing
resources. Dispatchable plants—which can respond to commands from the system operator to
change output—are the dominant source of flexibility today, and will remain of crucial
importance in the future. All types of dispatchable plants can change their output, but certain
types are technically better able to respond, both in the extent to which they can change their
11-52
10581090
output as well as the time it takes to do so. The critical characteristics of conventional plant
which can be used to maintain the balance of supply and demand are their ramp rates, startup
times, and minimum turn-down level.
11.5.7 Energy-Storage Technology
Of all the solutions discussed thus far, energy storage has the greatest versatility to address
renewable and load variability challenges. The addition of storage can put wind energy and solar
generators more on par with energy controllability inherent in traditional generation. Properly
sized energy storage can address renewable energy variability and ramping concerns. Energy
storage with fast output power control can also act to meet the power system’s reactive power
needs and reduce concerns about fluctuating wind output. Even remote wind locations, with
constrained transmission lines, can be better utilized when large-scale energy storage such as
compressed air energy storage (CAES) or pumped hydro enables scheduling of power flows
when transmission capacity is available.
The main issues with energy storage are capital cost and long-term durability. Also, there are
application challenges for optimizing the power and energy requirements for each specific
situation. Each of the viable storage technology options has advantages and disadvantages. For
example, the larger-scale solutions such as pumped hydro and CAES have shown the best
economics but may be potentially difficult to site. More modular technologies such as batteries
and ultracapacitors are much easier to site but may not store sufficient energy at an affordable
price. Higher-energy-capacity technologies such as flow batteries may not provide sufficient
power response to meet system ramping requirements. The following sections cover energy-
storage value and performance attributes, and compare costs of wind power applications.
Detailed information on the status, commercial and demonstration experience, performance and
cost of energy storage technology is presented in the 2003 EPRI report, EPRI-DOE Handbook
of Energy Storage for Transmission and Distribution Applications (EPRI 1001834, December
2003). A 2004 Wind Application supplement to the EPRI-DOE Energy Storage Handbook
presented a uniform approach to defining applications and estimating the benefits of energy
storage to wind energy. The 2010 EPRI report Energy Storage Technology Options: Primer on
Applications, Costs and Benefits and the 2009 research report Energy Storage market
Opportunities: Application Value and Technology Gap Assessment (EPRI 1017813) provide
added research findings on the value of energy storage. Also, EPRI completed a study in 2009
assessing the value and cost effectiveness of energy storage in ERCOT in a post-Competitive
Renewable Energy Zone (CREZ) Scenario 2 (CREZ-2) situation (see EPRI 1017824).
11.5.7.1 Energy Storage Technology Descriptions
The following descriptions are taken from the EPRI report Wind Power Integration: Energy
Storage for Firming and Shaping (EPRI 1008388, March 2005).
Pumped Hydroelectric Storage: Pumped hydroelectric storage, usually called pumped hydro
storage, is the most mature and most common form of bulk energy storage used at the utility
scale. Pumped hydro storage stores energy by pumping water from a reservoir at lower elevation
to a reservoir at higher elevation. The energy is stored in the higher potential energy of the water
11-53
10581090
at the higher elevation, and is recovered when the water is allowed to flow back to the lower
reservoir through hydroelectric turbines, regenerating the power. A cross-section of a typical
pumped hydro plant is shown in Figure 11-37.
Figure 11-37
Cross-section of a Pumped Hydro Storage Plant (Courtesy Tennessee Valley Authority)
Compressed Air Energy Storage: CAES plants store electrical energy by compressing and
storing a large volume of air. The compressed air is stored in underground caverns or aquifers, or
in aboveground piping or vessel systems. The energy is recovered by using the air as an input for
a turbine electric generator. Effectively, CAES plants operate by separating in time the
compression and expansion stages of the turbine. Figure 11-38 shows an example of a CAES
plant designed for electricity arbitrage.
Figure 11-38
Layout and Components of Typical CAES Plant (Source: Ridge Energy Storage)
11-54
10581090
Electrochemical Batteries: Electrochemical batteries are the oldest, best known, and most
widely used form of electrical energy storage. They were first developed in the early Nineteenth
Century, and played an important role in early investigations into electricity. Batteries store
energy in chemical form. An electrochemical cell is composed of two halves, a positive half and
a negative half. The positive half contains a positive electrode and a positive electrolyte; the
negative half contains a negative electrode and a negative electrolyte. The positive electrode and
electrolyte can react with each other to release energy, but only in the presence of a source of
electrons that can be supplied to the electrode. Similarly, the negative electrode and electrolyte
can react to release energy, but only in the presence of a sink that draws away the excess
electrons produced at the electrode.
If the positive electrode and the negative electrode are connected by a conducting wire, excess
electrons created by the reaction at the negative electrode can be drawn away to supply the
reaction at the positive electrode. This allows both reactions to proceed. The energy released can
be harnessed electrically by placing a load in the conduction path between the two electrodes.
The batteries most commonly used in the utility industry are secondary, or rechargeable,
batteries. In secondary batteries, the chemical reactions at the electrode can be run backwards by
applying a current in the reverse direction to the discharge current, with the battery absorbing
energy in the process. The absorption and release of energy can be done a number of times, until
the battery wears out due to chemical or mechanical wear. Secondary batteries are also often
called storage batteries.
Batteries have the advantage of being familiar and well-understood in application. Because they
store energy chemically, they are much more energy-dense than devices based on physical
processes such as pumped hydro, flywheels, and ultracapacitors. Batteries tend to wear out more
quickly than other energy storage devices, however, and are somewhat sensitive to the conditions
of use, particularly temperature. In addition, many batteries use environmentally toxic materials
such as lead and cadmium, which must be properly disposed of at the end of the project.
The battery technologies of most interest to the integration of VG, as described in the EPRI
report mentioned, are:
• Lead-acid batteries, which represent the oldest and most widely used type of commercial
storage battery. The lead-acid cell is composed of a lead negative electrode and a lead oxide
positive electrode in a common sulfuric acid electrolyte. They have been in commercial use
for well over a century, in applications throughout the industrial economy.
• Nickel-cadmium batteries, which are similar in operating principal to lead-acid batteries but
are constructed slightly differently, with nickel oxyhydroxide and cadmium electrodes in a
common potassium hydroxide electrolyte. Like lead-acid batteries, they are a relatively
mature technology, and can easily be built into relatively large systems. They boast
somewhat better cycle life and power performance than lead-acid, but at a considerably
higher expense.
• Sodium-sulfur batteries, which are based on a high-temperature electrochemical reaction
between sodium and sulfur, separated by a beta alumina ceramic electrolyte. While originally
developed for electric vehicle applications, they were adapted for the utility market by the
Tokyo Electric Power Company (TEPCO) and NGK Insulators, Ltd., both based in Japan.
11-55
10581090
• Flow batteries, which are electrochemical batteries in which the active materials are
contained in the electrolyte rather than in the solid electrodes. These electrolytes are stored in
tanks sized in accordance with application requirements, and are pumped through reaction
stacks which convert the chemical energy to electrical energy during discharge, and vice-
versa during charge. Figure 11-39 illustrates the principles of operation for a typical kind of
flow battery, the vanadium redox flow battery. There are several types of flow batteries of
which two types are available commercially: vanadium redox flow batteries and zinc-
bromine batteries.
Figure 11-39
Principles of Operation for a Vanadium Redox Flow Battery (Courtesy Sumitomo Electric
Industries)
Flywheels: Flywheels store energy in the angular momentum of a spinning mass. During charge,
the flywheel is spun up by a motor with the input of electrical energy; during discharge, the same
motor acts as a generator, producing electricity from the rotational energy of the flywheel.
Flywheels have several advantages over electrochemical batteries. Most products are capable of
several hundred thousand full charge-discharge cycles and so enjoy much better cycle life than
batteries. They are capable of very high cycle efficiencies of over 90%, and can be recharged as
quickly as they are discharged. Since the energy sizing of a flywheel system is dependent on the
size and speed of the rotor, and the power rating is dependent on the motor-generator, power and
energy can be sized independently. The down side to flywheels comes from their relatively poor
energy density and large standby losses.
Flywheels have predominantly been used in power quality applications, in which flywheels
provide ride-through capability for momentary voltage sags and interruptions. Many systems
incorporate both flywheels and generators. The flywheel allows ride-through of interruptions up
to about 15 seconds duration, a span that also allows enough time for a generator to come on-line
for longer interruptions.
11-56
10581090
Ultracapacitors: Ultracapacitors—also known as supercapacitors, electrochemical capacitors,

and electric double layer capacitors (EDLC)—are devices that resemble very large capacitors in
the way they perform electrically. This technology allows capacitors with very high capacitance,
measured in farads or even thousands of farads, but at relatively low voltages, between one and
three volts. High-voltage ultracapacitor systems consist of multiple individual cells connected in
series to produce the desired voltage.
Superconducting Magnetic Energy Storage (SMES): SMES devices store energy in magnetic
fields produced by continuously circulating current in a superconducting coil. Although it is
theoretically possible to use SMES systems for large-scale load-shifting applications, such an
application would require very large systems and would not be viable unless the cost of the
technology substantially decreases from its present cost. For this reason, SMES technology has
been implemented mostly for very short-duration discharges on the order of seconds.
Regenerative Hydrogen Storage: Regenerative hydrogen storage has also been proposed as an
energy storage mechanism. In this scheme, excess wind power is used to generate hydrogen via
the electrolysis of water. The water decomposes into oxygen, which is released into the
atmosphere, and hydrogen, which is compressed and stored. When electricity output is required,
the hydrogen is run through either an internal combustion engine or a fuel cell system to
regenerate electricity. The main objection to such a scheme is that the efficiencies associated
with hydrogen generation, storage, and conversion back to electricity are relatively poor, at least
at the present level of technology.
11.5.7.2 Energy Storage Applications and Benefits
The intermittent nature of wind power generation creates unique economic challenges for
developers when plants are operating in a deregulated environment. One of the basic principles
of deregulation is that buyers and sellers of electricity are able to enter into freely negotiated
contracts, months or even years ahead of time. This puts wind generators at a significant
disadvantage, due to the high risk associated with guaranteeing “firm” power at the contracted
delivery time. From an economic point of view, the intermittency of wind power has become
the primary issue for wind integration.
One way to minimize risk and increase wind value is to operate an energy-storage system
in parallel with the wind turbine. Appropriate operation of the resulting hybrid system will
potentially decrease trading risk by allowing firm power levels to be achieved. In addition,
storage enables optimization and will help maximize the possible revenue. Such operational
capability potentially allows the wind generator to take advantage of price fluctuations and to
avoid penalties for missing scheduled deliveries.
This potential for energy storage to directly augment the value of wind energy has been
recognized [28] [29]. Even so, it is likely that more than one benefit will be needed to make an
energy-storage investment cost effective. Wind energy arbitrage, ancillary service, firming and
shaping, spillage or curtailment reduction, and T&D investment deferral all potentially augment
the value of energy storage. Benefits and opportunities are very site- and market-specific, and
may also change over time. The full spectrum of potential benefits should be considered when
assessing the merits of combining energy storage with wind power. The choice to participate in
11-57
10581090
one market may preclude others. Economic dispatch of stored energy can be a critical and rather
complex problem, requiring both speculation and real-time decision making.
The economic value of energy storage systems to support high levels of wind penetration was
addressed in a 2009 research study for ERCOT. The objective of the study was to analyze the
economic benefits of installing various energy storage technologies in the ERCOT region.
Specific objectives were:
• Assess how CAES could increase wind utilization and penetration in ERCOT.
• Estimate the emissions impact of deploying CAES and battery systems.
• Estimate the impacts and interplay of CAES investments on wind curtailment, transmission
congestion and societal/system benefits.
• Assess similar impacts and the costs and benefits of deploying bulk battery energy storage
system and distributed battery systems near Texas load centers.
• Assess the value of emerging liquid air energy storage (LAES) systems.
This study examined energy storage deployment for a time frame assuming realization of the
CREZ Scenario 2 conditions, which assume an ambitious 18.5 GW of new wind energy
supported by ERCOT’s approved $4.9 billion investment in transmission system infrastructure.
The assessment was conducted using the UPLAN market simulation model and its underlying
suite of databases as the analytical platform. The ERCOT system was chosen for these energy
storage system assessments because of its significant wind deployments and because the system
details were already well documented in the model. EPRI provided estimates of cost and
performance of the energy storage options which included first- and second-generation CAES,
bulk battery and distributed battery systems, and a LAES system.
The resulting ERCOT study report summarizes market-based analysis of the impacts of three
energy or electricity storage technologies. It includes the impacts of these electricity storage
systems on locational marginal prices (LMP), CO2 emissions and wind curtailment. It calculates
the impacts of these systems on both the investors (power producers) and to society.
Natural-gas-based second-generation CAES systems were found to provide attractive rates of

return using installed capital costs of $ 750/kW. Bulk battery systems assumed to be installed for
$1,650/kW were not found to be as economical due to their current high capital costs. Advanced
2-MW distributed energy storage systems, assumed to be installed for $1,600/kW strategically
located around key load centers, were also found to be uneconomical. However these distributed
systems were found to provide very large societal system benefits in ERCOT. In the one case
studied, liquid air storage systems co-located near an existing combustion turbine were not found
to be economical due to their current high capital costs.
The present value of net income and societal benefit over a period of 15 years is presented in
Figure 11-40 below. The annual societal benefit of each storage option has also been included.
11-58
10581090
Present Value & Societal Benefit

1200 180
Present Value ($/ kW)
Societal Benefits ($ M) 160
Societal Benefits ($M)

Present Value ($/kW)
1000
140
800 120
100
600
80
400 60
40
200
20
0 0
CAESI CAESII Bulk Batt Dist Batt LAES
Storage Technology
Figure 11-40
Present Value and Societal Benefits of Energy Storage Technologies
A key recommendation from this research is that more regional, market-based studies, like the
ERCOT/UPLAN study, should be undertaken to better understand the roles, the locations, and
the trade-offs for energy storage and T&D investments to support wind integration.
Arbitrage Benefits: Prior studies have found that wind energy sold into bulk electricity markets
often receives lower-than-average energy value. One reason is that errors in forecasting wind can
lead to substantial energy imbalance penalties. Because energy storage allows for shifting wind
output over time, it mitigates wind forecasting errors or delivers energy at a time when electricity
prices are higher. The latter provides an energy arbitrage value to the wind storage system. As
reported by the Tennessee Valley Authority (TVA) [30], arbitrage benefits increase when there is
daily volatility in energy market prices. Higher peak energy prices increase storage benefits, but
higher off-peak energy prices decrease storage benefits. TVA also found that if storage is used
only for arbitrage, there is no optimal ratio of storage capacity to wind-production capacity. The
desired sizes of wind power and storage capacity can be chosen based on individual merits,
separately determined.
Though somewhat dated, these results are believed to be generally true. Changes in markets,
forecasting technology and the use of load as a resource may help to mitigate the effects of
forecasting errors and variability. Also, future improvement in market rules and operating
strategies may enable more effective collaboration of wind, storage, load control and other load
leveling resources.
Ancillary Services Benefits: With power system deregulation, regional markets and prices for
ancillary services have varied significantly. Spot market prices for regulation and reserves have
often exceeded the coincident prices for energy. Potentially lucrative service markets have not
been available to wind generators because of the intermittency of wind resources. However,
energy storage will enable wind energy to participate in the ancillary services market. A storage
11-59
10581090
subsystem, which can change output immediately and has very low operating cost when idle, is
especially well-suited to supply reserves and regulation in addition to providing energy arbitrage.
One requirement for using a wind and storage system for ancillary services is that the system is
on-line full time and available to respond quickly and reliably to orders for these services.
Several different types of storage batteries can charge and discharge fast enough to provide
frequency regulation. On-line storage can also provide spinning reserve and black start services.
Inverter-connected energy-storage systems may be capable of providing voltage regulation
services by making use of the kVAR rating of the inverter. The potential exists for a wind and
storage system to provide useful full-time ancillary support service during charging and
discharging, providing regulation, contingency reserve, or simply providing green
energy to the grid.
In another TVA study [31], the potential for using energy storage in different ancillary service
markets was evaluated. Figure 11-41 shows the annual total operating benefits for three ISOs:
New York ISO (NYISO), PJM Interconnection (PJM), and ISO New England (ISO-NE). Results
reflect the prevailing and varying prices in each market.
350
300
Annual Benefit, $/Installed KW
250 R eserv es (A ll T yp es)

R eg u latio n
200 E n erg y A rb itrag e
150
100
50
0
N Y IS O L o n g Islan d P JM E asto n Z o n e IS O -N E IS O -N E (P aym en ts)
Zone
Figure 11-41
Storage Operating Benefits in Different Ancillary Services Markets
In this case, prices in New England were significantly different from those in the other two
markets. Regulation provides the dominant benefit in PJM and NYISO. For ISO-NE, the
regulation benefit is much smaller than in the other markets, although it is still substantial. The
benefit available by providing reserves was noticeable only in the ISO-NE case. The need for
reserves varies significantly in different areas of the country.
TVA found that, if other benefits besides the clearing price of ancillary services are taken into
account, then the value can be significantly affected. ISO-NE was investigated because necessary
data were available and because other benefits in ISO-NE were much lower than in the other two
ISOs. The results for this case are labeled “ISO-NE (Payments).” In this case, accounting for the
other payments increased the overall benefits by 30%. A test case was run to estimate the impact
of the decline in regulation prices in the NYISO during the period of analysis. As might be
11-60
10581090
expected, the sensitivity of ancillary values to changing market prices was found to be high.
Overall, ancillary service markets for energy storage offer excellent value-added potential.
However, financing based on ancillary services alone is unlikely due to potential price volatility.
Other Energy Storage Benefits: Energy storage used for firming and shaping will allow more
of the wind-generated energy to be sold as firm energy by better matching wind output with daily
load variations. The value of this application of energy storage is the difference in market price
between firm and non-firm energy.
Curtailment of wind generation and related spillage of wind resources in constrained power
systems can be avoided when energy storage is applied. The condition that causes wind
generation to be curtailed during certain periods may be related to time-of-day or contingency
limits on the transmission, or related to the required margins for load following and regulating
generation. The value provided by energy storage in these cases is the capture of wind energy
that would otherwise be lost, as well as any related green power and production tax credits which
would otherwise be lost.
Closely related to curtailment avoidance is the value associated with substation or transmission
investment deferral. Here, energy storage allows power to be scheduled for periods when assets
are not heavily loaded, thus improving asset utilization and deferring the need for upgrades or
expansions.
Finally, in some cases, energy storage has been considered a way to stabilize real and reactive
power flow variations related to wind power. Energy storage must have fast response time, using
a power electronic interconnection combined with relatively high power capacity storage, to
mitigate voltage fluctuations that may be associated with fluctuating wind power output. Such
energy storage applications may have high value in solving localized stability and light flicker
problems.
11.5.7.3 Energy Storage Attributes and Costs
Despite the various possibilities for added value from energy storage in wind power, application
experience has been very limited. There are two main reasons for this. One is that electric energy
storage in a convenient form, such as batteries, is relatively expensive, typically tens to hundreds
of dollars per unit kilowatt-hour of storage capacity, or even thousands of dollars in short-
duration power quality applications. The second is that the electric utility grid serves as a large-
scale energy balancing and redistribution system for intermittent wind energy. The grid also
provides some level of damping for fluctuating wind power. Although energy storage is
effective, balancing, redistribution, and damping capacity of the grid are limited, and it incurs a
cost in operating the power system. As the penetration of wind generation grows, or in cases
where the grid is relatively weak, energy storage will find new applications. Both the longer-term
high-energy capacity storage for daily energy shaping and the short-term, high-power capacity
storage may be applicable for system stabilization.
A growing number of energy storage technologies are becoming available for these applications
and many new developments are in the R&D pipeline. In particular, new types of enclosed
batteries, being developed for the electric vehicle and portable electronics industries, may be
11-61
10581090
applicable to wind power. Also, because wind plants are considered stationary applications and
do not restrict size and weight, several larger-scale energy storage technologies are good
candidates, including pumped storage, compressed air, advanced lead-acid, and large flow
batteries. These larger systems may provide a needed economy of scale and better serve large
wind farm. In cases requiring both long-term energy and short-term stabilization, hybrids may be
considered.
Table 11-9 lists the candidate energy storage technologies and their use in supporting
applications of interest to the electric enterprise.
Table 11-10 compares the key performance and cost factors of these candidate technologies. The
comparison includes both long-term and short-term wind power applications. Long-term storage
typically provides 8-hour storage capacity with daily cycling to follow the system load. Short-
term storage provides up to 2-second charging and discharging to meet wind power stabilization
requirement. The initial cost per kilowatt-hour of each technology is based on existing plant
costs and ratings, or is projected for new technologies based on their expected relative cost and
performance. Several conclusions can be drawn from this table:
• Currently, of the enclosed batteries, advanced lead-acid, Zn/Br flow batteries and sodium-
sulfur (NaS) batteries are possible cost-effective solutions for firming and shaping wind
energy. The specific choice of system depends on the application and use. The key issues for
other battery types are limited field operating experience, depth of discharge (DOD), short
life expectancy and high initial cost.
• Among the flow battery technologies, Zn/Br and vanadium redox batteries (VRB) look
promising because of their relatively high-cycle life and scalability. In the pipeline, Zn/Cl
and Fe/Cr flow batteries may also be future options.
• Compressed air energy storage (CAES) and pumped hydro offer an excellent match with
wind energy by providing lower initial cost and higher storage capacities to meet the needs of
variable generation operating at 30% to 40% capacity factor. Also, their larger scale and long
life provide a good match for today’s large-scale wind farms. Second-generation CAES
plants offer potentially lower costs and higher efficiency than earlier first-generation CAES
options. Two second-generation energy storage plants will be built and demonstrated within
the next two to three years supported by DOE ARRA funding. One is in Norton, Ohio (First
Energy), the other is the Iowa Stored Energy Park.
11-62
10581090
Table 11-9
Advantages and Disadvantages of Candidate Energy Storage Technologies to Support
Wind Power Energy Variability
Independent
Operating Electric Advantages Disadvantages Manufacturers
Storage Technology
Vented Lead-Acid Mature and Coup de fouet (voltage drop Enersys, GNB (Exide)
(PbA Default) well-known which occurs at the beginning of
Low initial cost the discharge of lead-acid
Long life batteries)
Relatively intolerant of
temperature extremes
Valve-Regulated Low maintenance Coup de fouet C&D Technologies Hawker
Lead-Acid Low initial cost Intolerant of temperature Energy (Enersys)
extremes, Short life
Vented Nickel- Mature and well-known Low cell voltage Saft, Alcad
Cadmium (Ni-Cd) Long life, Relatively Float effect makes, capacity
tolerant to temperature testing difficult
extremes
Lithium-ion High energy density High initial cost (at present), Valence, Johnson Controls
Batteries Long cycle life Requires balancing and charge (formerly Varta), A123,
Potential for cost control electronics BYD, Altair
reduction due to markets Nanotechnologies, EnerDel,
for PHEV and EV NEC many others.
Lithium Metal High energy density Relatively untested, High initial Avestor, Seeo
Polymer Batteries Tolerant to temperature cost (at present), Requires
extremes balancing and charge control
electronics, one source
Nickel-Metal Hydride High energy density, Low cell voltage Intolerant of Johnson Controls (formerly
(Ni-MH) possible cycle life temperature extremes, Float Varta), Electro Energy
advantage over PbA effect makes capacity testing
difficult
Vanadium Redox Relatively high energy for Single manufacture and limited Prudent Power Systems
(VRB) large system sources for Vanadium, Not yet
Power and energy rating proven for cycle life or
is independent maintenance costs
Compressed Air Mature technology, long Requires suitable site geology, For main turbo- expander:
(CAES) life, Low cost for large may have ramp rate limit Large Alstom, Dresser-Rand,
scale Power and energy scale requires large capital Sulzer Energy Storage and
independent investment and collaborations Power LLC
Pumped Hydro Mature technology, long Requires suitable site geology, For Systems: Gugler
life, Low cost for large difficult to site, large scale GmbH, Sulzer, North Am.
scale Power and energy requires large capital investment Hydro, Water Alchemy,
independent and collaborations Harris
Sodium-Sulfur High energy density Relatively new and untested, NGK Insulator
Batteries (NAS) High initial cost (at present),
Single manufacture
Zn/Br Low cost; Long cycle life Limited field trials – however Premium Power, ZBB
some underway
Advanced Lead-Acid Low cost; long cycle life Limited field trials Extreme Power, Ecoult /
Batteries East Penn, Hitachi
11-63
10581090
Table 11-10
Advantages and Disadvantages of Candidate Energy Storage Technologies to Support
Wind Power Energy Intermittency and Output Fluctuations (Continued)
Independent
Operating Electric
Advantages Disadvantages Manufacturers
Storage
Technology
Limited field experience.
Zinc-bromine High energy density Mechanical parts require ZBB Energy;
batteries, Flat voltage profile maintenance May require
(needs power Premium Power
Hybrid Pairs
Low cost occasional stripping cycles,

partner)
Single manufacture
Still in R&D stage with technical
Regenerative uncertainties. Fluidic
zinc-air,
High energy density Voltage drop at start of Grid Storage LLC
(needs power
partner) discharge, Limited shelf and Revolt
cycle life, Single manufacture
High current density, High initial cost (at present), Maxwell, NESS
Ultracapacitor
Short Term
for short-term power, Relatively low energy density, Capacitor, EPCOS,

Stabilizer
s
high cycle life stacking limitation TBD ESMA, NEC, etc
High current density, Relatively high initial cost per Active Power, Beacon,
Flywheels for short-term power, kWh, Hitec, Piller, Pentadyne,
High cycle life Low energy density Teledyne
Updated cost comparisons of current energy storage technologies for various applications are
presented in the Table 11-11 below. Emerging or advanced technologies in the R&D pipeline are
not shown.
First-of-a-kind system costs will be higher than shown. Future system costs may be lower than
shown after early demonstrations are proven and products become standardized. Short-term
energy storage technologies that fit wind stabilizing applications exhibit relatively low initial
cost. However, in this application, energy storage must compete with other solutions such as
improved turbine output controls, fast response of turbines employing power electronics, and a
natural mitigation of fluctuations that comes from the diversity of large numbers of wind turbines
operating simultaneously.
11-64
10581090
Table 11-11
Energy Storage Characteristics by Application (Megawatt-scale)
Technology Maturity Capacity Power Duration % Efficiency Total Cost Cost

Option (MWh) (MW) (hrs) (total cycles) ($/kW) ($/kW-h)
Bulk Energy Storage to Support System and Renewables Integration
Pumped Mature 1680-5300 280-530 6-10 80-82 (>13,000) 2500-4300 420-430
Hydro
5400-14,000 900-1400 6-10 1500-2700 250-270
CT-CAES Demo 1440-3600 180 8 See note 1 960 120

(underground) (>13,000)
20 1150 60
CAES Commercial 1080 135 8 See note 1 1000 125
(underground) (>13000)
2700 20 1250 60
Sodium-Sulfur Commercial 300 50 6 75 (4500) 3100-3300 520-550
Advanced Commercial 200 50 4 85-90 (2200) 1700-1900 425-475
Lead-Acid
Commercial 250 20-50 5 85-90 (4500) 4600-4900 920-980
Demo 400 100 4 85-90 (4500) 2700 675
Vanadium Demo 250 50 5 65-75 (>10000) 3100-3700 620-740
Redox
Zn/Br Redox Demo 250 50 5 60 (>10000) 1450-1750 290-350
Fe/Cr Redox R&D 250 50 5 75 (>10000) 1800-1900 360-380
Zn/Air Redox R&D 250 50 5 75 (>10000) 1440-1700 290-340
Energy Storage for ISO Fast Frequency Regulation and Renewables Integration
Flywheel Demo 5 20 0.25 85-87 1950-2200 7800-8800
(>100,000)
Li-ion Demo 0.25-25 1-100 0.25-1 87-92 1085-1550 4340-6200
(>100,000)
Advanced Demo 0.25-50 1-100 0.25-1 75-90 950-1590 2770-3800
Lead-Acid (>100,000)
Energy Storage for Utility T&D Grid Support Applications
CAES Demo 250 50 5 See note 1 1950-2150 390-430
(aboveground) (>10,000)
Advanced Demo 3.2-48 1-12 3.2-4 75-90 (4500) 2000-4600 625-1150
Lead-Acid
Sodium-Sulfur Commercial 7.2 1 7.2 75 (4500) 3200-4000 445-555
Zn/Br Flow Demo 5-50 1-10 5 60-65 (>10,000) 1670-2015 340-1350
Vanadium Demo 4-40 1-10 4 65-70 (>10,000) 3000-3310 750-830
Redox
Fe/Cr Flow R&D 4 1 4 75 (>10,000) 1200-1600 300-400
Zn/Air R&D 5.4 1 5.4 75 (4500) 1750-1900 325-350
Li-ion Demo 4-24 1-10 2-4 90-94 (4500) 1800-4100 900-1700
11-65
10581090

Energy Storage Characteristics by Application (Megawatt-scale)

Option (MWh) (MW) (hrs) (total cycles) ($/kW) ($/kW-h)
Energy Storage for Commercial and Industrial Applications
Advanced Demo- 0.1-10 0.2-1 4-10 75-90 (4500) 2800-4600 700-460
Lead-Acid Commercial
Sodium-Sulfur Commercial 7.2 1 7.2 75 (4500) 3200-4000 445-555

Zn/Br Flow Demo 0.625 0.125 5 60-63 (>10,000) 2420 485-440
2.5 0.5 5 2200
Vanadium Demo 0.6-4 0.2-1.2 3.5-3.3 65-70 (>10,000) 4380-3020 1250-910
Flow
Li-ion Demo 0.1-0.8 0.05-0.2 2-4 80-93 (4500) 3000-4400 950-1900

1. Refer to the full EPRI report for important key assumptions and explanations behind these estimates. All
systems are modular and can be configured in both smaller and larger sized not represented. Figures are
estimated ranges for the total capital installed cost estimates of “current” systems based on 2010 inputs from
vendors and system integrators. Included are the costs of power electronics if applicable, all costs for
installation, step-up transformer, and grid interconnection to utility standards. Smart-grid communication and
controls are also assumed to be included. For batteries, values are reported at rated conditions based on
reported depth of discharge. Costs include process and project contingency depending on technical maturity.
The cost in $/kW-h is calculated by dividing the total cost by the hours of storage duration.
2. For CAES and Pumped Hydro, larger and smaller systems are possible. For belowground CAES the heat rate
may range from ~3845-3860 Btu/kWh and the energy ratio is 0.68-0.78; for aboveground CAES the heat rate is
~4000 Btu/ kWh and the energy ratio is ~1.0.
3. For C&I and Residential applications lower CapEx costs may be possible if the battery system is integrated and
installed with a photovoltaic system.
In addition to the initial and life-cycle costs, key factors for applying these energy storage
technologies to wind applications are power and energy performance, usable charge and
discharge cycles, and reliability. Figure 11-42 shows the separation of the charge and discharge
equipment. In this model, the charge and discharge durations can also be designed
independently. Since wind power typically operates at a capacity factor of 30% to 40%, it is not
unusual that the charging system may be rated higher than the discharging system. This allows
capture of wind energy during a short time frame. Also, large storage capacity will allow
effective operation during periods of low wind production.
11-66
10581090
Figure 11-42
Model for Evaluation of Energy Storage with Wind Power
This model addresses the main performance considerations of separately designing power
and energy requirements. For example, as shown in the figure, system performance may benefit
from the power rating being independent of energy capacity. In addition, charging may involve
different converter equipment, and a longer period, than discharging. Available ancillary
services, such as frequency regulation or spinning reserve, may require that the discharge
converter be on line and independent of the charging converter.
In the future, energy storage will be an increasingly important component of wind plant
applications as wind penetration increases, the costs of wind intermittency are recognized, and
the costs of available energy storage technologies are reduced. Wind-storage applications are
expected to be most needed when wind power is installed in relatively weak power systems, such
as remote or island sites, and when the cost of electric energy is relatively high. Because of the
relatively large scale of wind plants and the need to meet daily load cycles, long-term and
scalable energy storage technologies are most likely for wind applications. These technologies
need to be scalable and clearly benefit from economies of scale. Examples are CAES, pumped
hydro, flow (redox batteries), and low-cost batteries such as lead-acid and advanced lead-acid
systems.
Short-term energy storage technologies for stabilizing wind fluctuations may include advanced
lead-acid, lithium-ion, and advanced supercapacitors under development. The current costs of
each of these technologies need to be evaluated by application needs. For example, advanced
lead-acid battery systems are being deployed in several wind applications on remote islands or
isolated grids. Short-term energy storage for stabilization competes directly with other solutions
including the grid’s ability to absorb fluctuating energy, improvement in wind turbine output
controls for mitigating fluctuations, the diversity and smoothing effects of many wind turbines in
the same general vicinity, and other power electronic equipment such as the DSVC (distributed
static VAR compensator).
11-67
10581090
11.5.8 Grid Planning and Operating Tools
New tools to enable grid operators to better visualize the real-time status of the grid and
anticipate contingencies are needed to moderate additions of VG resources. Several tools and
methods are being developed in this area:
• Modeling tools to forecast renewable energy output on a near-real-time frame and up to 24
hours, with confidence levels for grid operators to have advanced notice to commit adequate
capacity, plan and schedule to meet energy requirements. Benefits include minimizing
unscheduled power flows, real-time voltage and dynamic stability problems, and preventing
blackouts. Research scope should include the use of time-dependent optimal power flow
techniques to develop adaptive strategies for real-time operation.
• Methods to evaluate the capacity value of an aggregation of variable generation, energy
storage, and demand response providers, which allows for coordinated dispatch and
scheduling. This would involve energy markets to help balance the economics and risks in
power system operations. The probabilistic nature of these resources will need to be taken
into account when considering reserve provision and dispatch to meet demand.
• Planners will need to be able to ensure sufficient flexibility on the grid to deal with the
increased variability and uncertainty. This needs to recognize contributions from distributed
resources such as storage, electric vehicles and demand response. Additionally, planners
must consider more than just the traditional peak load or snapshot hours to ensure that hour-
to-hour variability can be managed.
• Visualization for grid operation needs to be extended to cover intermittent generation in a
geographical display, together with visualization of load and demand response capabilities
and the effect they have on the transmission loadings, voltages and frequencies of the power
grid. This concept, developed by EPRI, will provide greater wide-area situational awareness
to operators. Control room visualization may be a key technology for increasing intermittent
generation without significant cost or loss of grid reliability. R&D is needed to develop
automation concepts on adaptive relay protection and voltage instability load shedding to
handle the integration of substantial amounts of intermittent generation.
• Bulk System Coordination of photovoltaic for Market and Bulk System Control: Dispatch
center control of distributed renewable generation on the transmission system will be needed.
This will allow renewable generation to participate and be aggregated into energy markets as
well as allow for control to preserve system stability, power quality, and reliability at the
bulk level.
11.6 Integration Technologies and Strategies for Distribution-Connected

With large amounts of distributed connected generation, new strategies to manage power on the
distribution level, as opposed to transmission level, will be needed. The distribution system will
need to evolve to accommodate two-way power flows. In the future, a more automated
distribution system will interact with distributed energy resources including photovoltaic
generation and battery systems. This will in turn enable better utilization of these resources and
higher penetration. As smart grid concepts are implemented on the distribution system, it will
11-68
10581090
become increasingly feasible to integrate distributed PV into grid operations. This will need large
amounts of time-synchronized data to be fed back to a control room. Better visibility of
distributed PV from the control room, coupled with PV inverter technology that can be
responsive to system needs via operator commands and price signals, would enable distributed
PV to participate in energy and capacity markets and help support system reliability. Emerging
communication and control technologies make it feasible for PV to be aggregated into “virtual
power plants.” Combining solar with demand response creates a compelling reliability product,
but this concept is in its infancy. All of these changes will add complexity to the operations of
the distribution system. The changes in the distribution grid for high-penetration PV will likely
include:
• Interactive Voltage Regulation and VAR Management: Utility voltage regulator and
capacitor controls will be interactive with each other and photovoltaic sources. The central
controller shown in Figure 11-43 will help manage the interactivity to ensure optimized
voltage and reactive power conditions.
• Protective Relaying Schemes Designed for Renewable Generation: The distribution system
and sub-transmission will include more extensive use of directional relaying,
communication-based transfer trips, pilot signal relaying, and impedance-based fault-
protection schemes (like those used in transmission). These can work more effectively with
multiple sources on the distribution system.
• Advanced Islanding Control: Extensively automated switchgear and renewable generation
with enhanced islanding detection to improve capability for detecting unintentional islands.
Also, these systems should be able to reconfigure the grid/renewable generation into
reliability-enhancing “intentional islands.”
Figure 11-43
Distributed Controller Results are Aggregated to Manage Area Power and System Voltage
Profiles
11-69
10581090
Interactive Service Restoration: Sectionalizing schemes for service restoration allow distributed
photovoltaic and other renewable generation to pick up load during the restoration process as
shown in Fig 11-44. Once separated, these must deal effectively with overloads from cold-load
pickup and the current in-rush required to recharge the system.
Rotating
PV Machine
Inverter Source Switch C
Source Switch B (Last to close)
SUBSTATION
(3rd to close)
FEEDER
Circuit Breaker Switch A PV

Rotating
(first to close) (2nd to close) Machine Inverter
Source Source
Load Load
Area 1 Load Load
Area 2
Area 3 Area 4
Figure 11-44
Illustration of Cascaded Restoration of Distributed Generators
• Improved Grounding Compatibility: Consider new devices and architectures in both

renewable generation and distribution that address grounding incompatibilities between
power system sensing, protection, and harmonic flows. Examples of such techniques are:
– Control or limit ground fault overvoltage via relaying techniques or ancillary devices
instead of effectively grounded renewable generation requirements.
– Harden the power system and loads to be less susceptible to ground fault overvoltage
(increase voltage withstand ratings).
– Change protective relaying for ground faults so a high penetration of grounding sources
does not affect the ground fault relaying.
– Change feeder grounding scheme or load serving scheme back to a grounded three-wire
system.
• Distributed Energy Storage: Energy storage of various forms will apply to correct temporary
load/generation mismatches, regulate frequency, mitigate flicker, and assist advance
islanding functions and service restoration. Table 11-12 shows the costs of distribution scale
storage.
11-70
10581090
Table 11-12
Energy Storage Characteristics by Application (Kilowatt-scale)

Option (kWh) (kW) (hrs) (total cycles) ($/kW) ($/kW-h)
Energy Storage for Distributed (DESS) Applications
Advanced Demo- 100-250 25-50 2-5 85-90 1600- 3725 400- 950
Lead-Acid Commercial (4500)
Zn/Br Flow Demo 100 50 2 60 1450-3900 725-1950
(>10000)
Li-ion Demo 25-50 25-50 1-4 80-93 2800-5600 950-3600
(5000)
Energy Storage for Residential Energy Management Applications*
Lead-Acid Demo- 10 5 2 85-90 4520-5600 2260
Commercial (1500-5000)
20 4 1400
Zn/Br Flow Demo 9-30 3-15 2-4 60-64 2000-6300 785- 1575
(>5000)
Li-ion Demo 7-40 1-10 1-7 75-92 1250-11,000 800-2250
(5000)
1. Refer to the full EPRI report for important key assumptions and explanations behind these estimates. All
systems are modular and can be configured in both smaller and larger sized not represented. Figures are
estimated ranges for the total capital installed cost estimates of “current” systems based on 2010 inputs from
vendors and system integrators. Included are the costs of power electronics if applicable, all costs for installation,
step-up transformer, and grid interconnection to utility standards. Smart-grid communication and controls are also
assumed to be included. For batteries, values are reported at rated conditions based on reported depth of
discharge. Costs include process and project contingency depending on technical maturity. The cost in $/kW-h is
calculated by dividing the total cost by the hours of storage duration.
2. For CAES and Pumped Hydro, larger and smaller systems are possible. For belowground CAES the heat rate
may range from ~3845-3860 Btu/kWh and the energy ratio is 0.68-0.78; for aboveground CAES the heat rate is
~4000 Btu/ kWh and the energy ratio is ~1.0.
3. For C&I and Residential applications lower CapEx costs may be possible if the battery system is integrated
and installed with a photovoltaic system.
4. First-of-a-kind system costs will be higher than shown. Future system costs may be lower than shown after
early demonstrations are proven and products become standardized.
These system changes and technology upgrades represent an extensive investment on the part of
electric utilities, rate payers and equipment manufacturers, and a huge change in the way the
power system is designed and operated. These changes will not come overnight and will require
many decades to implement, as well as considerable engineering planning and development to
determine the balance of features and capabilities needed against the cost and complexity of
implementation. Nonetheless, these are the approaches needed to move to high-penetration PV,
and the industry needs to begin work now on research and development so that the technologies,
tools, and approaches will be available in a timely manner.
Another configuration on the horizon is likely to coordinate multiple generators at dispersed

locations and with a variety of energy resources, including various combinations of solar, wind,
other generators and energy-storage devices. Microgrids may be applied in a broad range of sizes
and configurations. Figure 11-45 shows examples of possible microgrid “subsets” that could be
11-71
10581090
developed on a typical radial distribution system. These microgrid subsets include a single
customer, a group of customers, an entire feeder, or a complete substation with multiple feeders.
A very large substation could have up to 100 MW of capacity, eight or more feeders, and could
serve more than 10,000 customers.
Figure 11-45
Concept of Distribution Microgrids of Various Sizes and Levels Allowing Reliability Islands
and Grid Tie Operation
Challenges with microgrids are many. Regardless of their size, they must take on key control
responsibilities while operating in the islanded state; otherwise, serious damage may result. At
the same time, these distributed generators must not adversely impact reliability, voltage
regulation, or power quality on the bulk power system while the microgrid is interconnected.
Advanced inverter/controllers and energy management will need to be sufficiently sophisticated

to interface with emerging smart grid technology. As such, they must be capable of supporting
communication protocols utilized by current energy management and utility distribution level
communication systems. This shift from today’s central control system to a future intelligent
control is illustrated in Figure 11-46.
11-72
10581090
Figure 11-46
Distributed Controller Must be Integrated with Overall Distribution Control Systems to
Maximize System Value and Reduce Capacity Requirements
The master controller is the key to providing a future sophisticated microgrid operation that
maximizes efficiency, quality, and reliability. While some of the capabilities identified for an
intelligent microgrid master controller are currently being researched, other capabilities do not
yet exist. The Galvin Electricity Initiative documented the functional requirements for master
controller software in the report Master Controller Requirements Specification for Perfect Power
Systems (02162007 Rev. 3, 2007), which is available from the Galvin Electricity Initiative’s
website at www.galvinpower.org.
In 2008, the German Association of Energy and Water Industries (BDEW) introduced new grid
codes for connecting power plants to the medium-voltage grid. Generating plants connected to
the medium-voltage networks have to support grid stability and must not disconnect from the
grid during a fault, as is common practice today in many countries. With the increasing
penetration of DG sources in the medium voltage grid, it is necessary that these plants include
both dynamic and static grid-support capabilities.
Due to high penetration of DG plants in the medium-voltage grid, new medium voltage grid code
in Germany requires that these plants provide voltage stability during voltage drop in the grid,
often referred to as fault ride-through (FRT). Figure 11-47 illustrates one ride-though concept
proposed in this grid code. PV systems connected to medium voltage grid need to:
• Stay connected during a voltage drop down to 0% UC with duration up to 150 ms
• Offer stable operation above Limit 1
• Ride through the low voltages within Limits 1 and 2
11-73
10581090
• Support the grid by providing additional reactive current during a symmetrical fault
• In case of an unsymmetrical fault the reactive current must not cause voltages increase above
1.1 UC in the non faulty phases.
Figure 11-47
FRT Limiting Curves Proposed in the New German Grid Codes for Connecting PV Systems
to the Medium-Voltage Power Grid.
According to this medium-voltage grid code, a plant must be able to curtail active power output
in steps of 10% (or smaller) of the agreed rated output power. Above a system frequency of 50.2
Hz, all generating units have to reduce their power output with a gradient of 40%/Hz of the
instantaneously available power. Above 51.5 Hz and below 47.5 Hz, the plant has to disconnect
from the grid. In response to following cases, the network operator can temporarily limit the
feed-in power or disconnect the plant:
• Risk of unsafe system operation,
• Risk of bottlenecks and congestion in the network,
• Risk of unwanted islanding,
• Risk of static or dynamic grid instability,
• Risk of system instability due to frequency increase,
• Carry out repairs or construction,
• In the context of production management, feed-in management, and network security
management.
11-74
10581090
This new German grid code also requires DG plants connected to the medium-voltage grid to
provide static grid support by providing reactive power. The reactive power set point can be
either fixed or adjustable by a signal from the network operator:
• A fixed displacement factor
• A variable displacement factor depending on the active power
• A fixed reactive power value in MVar
• A variable reactive power depending on the voltage.
11.7 References
11.7.1 Endnotes
1. Wind Report 2004: E.ON Netz GmbH, Bayreuth, Germany. 2004, updated by “German
Renewable Energy Sector Shows Impressive Growth,” Greenprices, January 24, 2007.
2. Study of Electric Transmission in Conjunction with Energy Storage Technology: Lower

Colorado River Authority, Austin, TX, Ridge Energy Storage & Grid Services, Austin, TX,
RnR Engineering, Austin, TX, and Walter J. Reid Consulting, Austin, TX. August 2003.
3. Giovanni, Dan, Wind Energy in the Big Island: Hawaii Electric Light Company, Hilo, HI.
Presented at USDOE/DBEDT Forum. April 2002.
4. Brooks EDL, Smith J, Pease J, McGree M. “Assessing the Impact of Wind Generation on
System Operations at Xcel Energy-North and Bonneville Power Administration.” Windpower
2002. AWEA Washington, D.C.
5. Brooks, D., Lo, E., Zavadil, R., Santoso, S., Smith, J. “Characterizing the Impact of
Significant Wind Generation Facilities on Bulk Power System Operations Planning.”
UWIG, May 2003.
6. Dragoon, K., Milligan, M. “Assessing Wind Integration Costs with Dispatch Models: A Case
Study.” Windpower 2003, Austin, TX.
7. Hirst, E. “Integrating Wind Energy with the BPA Power System: Preliminary Study.”
Prepared for BPA Power Business Line, September 2002.
8. Hirst, E. “Integrating Wind Output with Bulk Power Operations and Wholesale Electricity
Markets.” Wind Energy 5(1) 2002. pp. 19–36.
9. Electrotek Concepts, Inc. “Systems Operations Impacts of Wind Generation Integration

Study.” Prepared for We Energies, June 2003.
10. United States Patent #5,083,039, Variable Speed Wind Turbine, Richardson et al. issued
January 21, 1992.
11-75
10581090
11. Asplund G. et al. “DC Transmission Based on Voltage Source Converters.” CIGRE Session
1998 paper No 14-302.
12. Grainger, W. and Jenkins, N. “Offshore Wind Farm Electrical Connection Options.” British
Wind Energy Association, 2002.
13. Torbjörn Thiringer et al. “Grid Disturbance Response of Wind Turbines Equipped With
Induction Generator and Doubly-Fed Induction Generator.” Chalmers University of
Technology, IEEE T&D Dallas, September 2003.
14. IEEE Draft Trial-Use Guide for Application of Power Electronics for Power Quality
Improvement on Distribution Systems Rated 1 kV through 38 kV, P1409, D10,
September 5, 2003.
15. Ambra Sannino, Jan Sversson, Tomas Larsson, “Power-Electronic Solutions to Power
Quality Problems.” Electric Power Systems Research 66 (2003) 71-82, Department of
Electric Power Engineering, Chalmers Univ. of Technology, Gothenburg, Sweden.
16. Short-Term Wind Generation Forecasting Using Artificial Neural Networks. EPRI, Palo
Alto, CA: October 2003. Report 1009219.
17. California Wind Energy Forecasting System Development and Testing Phase 1: Initial
Testing. California Energy Commission, Sacramento CA; EPRI, Palo Alto, CA: 2002.
1003778.
Testing. California Energy Commission, Sacramento CA, EPRI, Palo Alto, CA: 2003.
1003779.
19. Texas Wind Energy Forecasting System Development and Testing Phase 1: Initial Testing.
EPRI, Palo Alto, CA; U.S. Department of Energy, Washington D.C.: 2003. 1008032.
20. Texas Wind Energy Forecasting System Development and Testing Phase 2: 12-Month
Testing. EPRI, Palo Alto, CA; U.S. Department of Energy, Washington D.C.: 2004.
1008033.
21. Wind Energy Forecasting Applications in Texas and California. EPRI, Palo Alto, CA: 2003.
1004038.
22. California Regional Wind Energy Forecasting System Development, Volume 1: Executive
Summary. EPRI, Palo Alto, CA; California Energy Commission, Sacramento, CA: 2006.
1013262.
23. California Regional Wind Energy Forecasting System Development, Volume 2: Wind Energy
Forecasting System Development and Testing and Numerical Modeling of Wind Flow over
Complex Terrain. EPRI, Palo Alto, CA; California Energy Commission, Sacramento, CA:
2006. 1013263.
11-76
10581090
24. California Regional Wind Energy Forecasting System Development, Volume 3: Wind Tunnel
Modeling of Wind Flow over Complex Terrain. EPRI, Palo Alto, CA; California Energy
25. California Regional Wind Energy Forecasting System Development, Volume 4: California
Wind Generation Research Dataset (CARD). EPRI Palo Alto, CA, California Energy
26. John W. Zack. Wind Forecasting: “State of the Art”. CanWEA Wind and Power Systems
Seminar, May 20, 2009, Montreal, Quebec.
27. Andrew Cruden and Graham J. W. Dudgeon. Opportunities for Energy Storage Devices
Operating with Renewable Energy Systems. EESAT 2000, September 18–20, Orlando
Florida.
28. Goren Strbac, Graeme N. Bathurst, and Nick Jenkins. Integration of Renewable Generation
into the UK Market—Opportunities for Energy Storage. EESAT 2000, September 18–20,
Orlando Florida.
29. Robert E. Taylor and Joseph J. Hoagland, Using Energy Storage With Wind Energy for
Arbitrage, EESAT 2002, April 15–17, 2002 San Francisco California.
30. R. Taylor, J. Hoagland, D. Bradshaw. Energy Storage for Ancillary Services, EESAT 2002,
April 15–17, 2002 San Francisco California.
31. T. Key, H. Kamath, Electric Energy Storage: Will it Help Enable DE, Primen Perspective
on Distributed Energy, Primen, DE-PP-23, August 2003.
11.7.2 General
Wind Power Integration Assessment and Case Studies. EPRI, Palo Alto, CA: March 2004.
1004806.
EPRI-DOE Handbook of Energy Storage for Transmission and Distribution Applications).

EPRI, Palo Alto, CA; U.S. Department of Energy, Washington, D.C.: December 2003. 1001834.
Norris, B., EPRI Energy Storage Handbook: Wind Energy Storage Applications, Gridwise
Engineering Co., Danville, CA: February 2003.
Technical Assessment Guide–Storage. EPRI, Palo Alto, CA: December 2000. 1001203.
11.7.3 Renewable Generation Grid Impacts and Integration
Characterizing the Impacts of Significant Wind Generation Facilities on Bulk Power System
Operations Planning: Utility Wind Interest Group - Xcel Energy-North Case Study, Utility Wind
Interest Group, Reston. VA, EPRI, Palo Alto, CA. December 2003. 1004807.
Wind Report 2004: E.ON Netz GmbH, Bayreuth, Germany. 2004.
11-77
10581090
Eastern Wind Integration and Transmission Study, NREL, 2009
Western Wind and Solar Integration Study, NREL, 2010
11.7.4 Battery Energy Storage
Assessment of Advanced Batteries for Energy Storage Applications in Deregulated Electric

Utilities. EPRI, Palo Alto, CA: 1998. TR-111162.
Lessons Learned from the Puerto Rico Battery Energy Storage System, Sandia National
Laboratories, Albuquerque, NM: 1999. SAND99-2232.
Benjamin L. Norris, Greg J. Ball, Utility Test Results of a 2-Megawatt, 10-Second Reserve-
Power System, Sandia National Laboratories, Albuquerque, NM: 1999. SAND99-2570.
TD. Berndt, Maintenance Free Batteries, 2nd ed., Research studies Press Ltd., Taunton,
Somerset, England: 1997.
William E. Jones and David I. Feder, “Float Behavior of VRLA Cells: Theory vs. Reality,”
paper 7-4, Proceedings 17th Int. Telecom Energy Conference, The Hague, Netherlands: 1995.
Battery Energy Storage for Utility Applications: Phase I—Opportunities Analysis, Sandia
National Laboratories, Albuquerque, NM: 1994. SAND94-2605.
Chino Battery Energy Storage Plant. Vol. 1. EPRI, Palo Alto, CA: 1993. TR-101787.
Chino Battery Energy Storage Plant: First Year of Operation. EPRI, Palo Alto, CA: 1993.
TR-101786.
Development of Zinc-Bromine Batteries for Utility Applications: Commercialization Task.

EPRI, Palo Alto, CA: 1991. GS-7093.
Design and Costs for a Generic 10-MW Utility Lead-Acid Battery Energy Storage Plant.
EPRI, Palo Alto, CA: 1988. AP-5845.
11.7.5 Compressed Air Energy Storage
Hybrid Plants for Power Generation in the Uninterruptible Power Substation, Phase I.
EPRI Report, to be published.
Transient Analysis of Hybrid Plants—Hybrid Plants for Power Generation in the

Uninterruptible Power Substation, Phase II, EPRI Report, to be published.
Hybrid Plants for Power Generation in the Uninterruptible Power Substation, Phase III.
EPRI Report, to be published.
Compressed Air Storage with Humidification: An Economic Evaluation. EPRI, Palo Alto,
CA: 1998. TR-111691.
11-78
10581090
Standard Compressed Air Energy Storage Plant: Design and Cost. EPRI, Palo Alto, CA: 1994.
TR-103209.
History of First U.S. Compressed Air Energy Storage (CAES) Plant. EPRI, Palo Alto, CA: 1993.
TR-101751, Vols. 1 & 2.
Proceedings: Second International Conference on Compressed Air Energy Storage. EPRI, Palo
Alto, CA: 1993. TR-101770.
Compressed Air Storage with Humidification (CASH) Coal Gasification Power Plant
Investigation. EPRI, Palo Alto, CA: 1991. GS-7453.
Recuperators for Compressed Air Energy Storage Plants. EPRI, Palo Alto, CA: 1989. GS-6571.
11.7.6 Pumped Hydro Energy Storage
“Existing Capacity and Planned Capacity Additions at U.S. Electric Utilities by Energy Source
and State—1999,” U.S. DOE, http://www.eia.doe.gov/cneaf/electricity/ipp/html1/t17p01.html.
“The Comprehensive Electricity Competition Act: A Comparison of Model Results—Appendix

B: CECA Competitive Case: Data Forecasting Products,” U.S. DOE,
http://www.eia.doe.gov/oiaf/servicerpt/ceca/apb.html.
Annual Energy Outlook 2001, Reference Case Tables Forecasting Products, U.S. DOE,
http://www.eia.doe.gov/oiaf/aeo/earlyrelease/aeotab_9.htm.
Pumped Storage Planning and Evaluation Guide. EPRI, Palo Alto, CA: January 1990. GS-6669.
Compendium of Pumped Storage Plants in the United States, American Society of Civil
Engineers: 1993. ISBN: 0-87262-991-0.
Hydroelectric Pumped Storage Technology: International Experience, American Society of Civil

Engineers: 1996. ISBN: 0-7844-0144-6.
11.7.7 Flywheel Energy Storage
A Summary of the State of the Art of Superconducting Magnetic Energy Storage Systems,
Flywheel Energy Storage Systems, and Compressed Air Energy Storage Systems, Sandia
Abacus Technology Co., An Assessment of Flywheel Energy Storage Technology for Hybrid
and Electric Vehicles, for the U.S. Department of Energy, Office of Transportation
Technologies, Electric and Hybrid Propulsion Division: July 1996.
Utility-Size Flywheel Energy Recovery and Storage System (FERS) Feasibility Study. EPRI, Palo
Alto, CA: 1989. RP 1084-20.
11-79
10581090
Draft Report on Current Status of Flywheel Energy Storage. EPRI Internal Draft Report.
EPRI, Palo Alto, CA: November 1995.
Flywheels for Electric Utility Energy Storage. EPRI, Palo Alto, CA: 1999. TR-108889.
11.7.8 SMES
W. Buckles and W. V. Hassenzahl, “Superconducting Magnetic Energy Storage,” IEEE Power

Engineering Review, Vol. 20, No. 5, pp. 16-20: May 2000.
An Assessment of High Temperature Superconductors for High Field SMES Systems.

EPRI, Palo Alto, CA: 1999. TR 110719.
A Summary of the State of the Art of Superconducting Magnetic Energy Storage Systems,
Flywheel Energy Storage Systems, and Compressed Air Energy Storage Systems, Sandia
J. T. Steiger, G. Fuchs, P. Verges, K. Fischer, L. Schultz, and A. Gladun, “Critical Current

in Silver Sheathed Bi-2223 Tapes,” IEEE Trans. Appl. Supercon. 7, 1347: 1997.
M. A. Green, R. A. Byrns, and S. J. St. Lorant, “Estimating the Cost of Superconducting

Magnets and the Refrigerators Needed to Keep Them Cold,” Advances In Cryogenic
Engineering, Vol. 37, Plenum Press, New York: February 1992.
J .E. Dagle, J. G. De Steese, M. K. Donnelly, and D. J. Trudnowski, Evaluation of

Superconducting Magnetic Energy Storage at Blythe, prepared for San Diego Gas and Electric,
Battelle, Pacific Northwest Labs, Richland, Washington: July 1995.
John H. Doudna, Analysis of Benefits from Superconducting Magnetic Energy Storage

on the ML&P System, Interim Report No. 3: April 1995. PTI Report No. R4-95.
11-80
10581090
12
GREENHOUSE GAS EMISSIONS CONTROL
12.1 Greenhouse Gases
Several greenhouse gases are known to contribute to humanity’s effect on the radiation balance
in the atmosphere and, hence, on global temperature and potential climate change. The
greenhouse gases of concern include carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O),
and certain chlorofluorocarbons (CFC) that react with and deplete the ozone layer. The estimated
lifetimes of these gases in the atmosphere before they oxidize to carbon dioxide and their
infrared absorbing strengths vary. Table 12-1 compares those characteristics relative to carbon
dioxide as a function of time.
Table 12-1
Lifetimes in the Atmosphere and Relative Infrared Absorption Strengths of the
Greenhouse Gases
Lifetime in the Infrared Absorbing Strength Per Unit Mass vs. CO2
Gas
Atmosphere 20-Year 50-Year 100-Year
Carbon Dioxide (CO2) Variable 1 1 1
Methane (CH4) 12 (±3) Years 56 21 7
Nitrous Oxide (N2O) 120 Years 280 310 170
Chlorofluorocarbons Not Given 4,900 3,800 Not Given
(CFCs)
Source: U.S. DOE EIA, Emissions of Greenhouse Gases in the U.S. 1996, October 1997.
Over a 50-year timeframe, methane absorbs 21 times as much infrared radiation as an equal mass
of carbon dioxide. Thus, we say that the infrared absorbing equivalent of a ton of methane
released by a landfill is 21 tons of carbon dioxide. Similarly, the infrared absorbing equivalents
of a ton of nitrous oxide and a ton of chlorofluorocarbons are 310 tons and 3,800 tons of carbon
dioxide, respectively.
12.2 Role of Renewable Energy Technology

Renewable generation technologies typically produce no direct emissions of fossil fuel carbon
dioxide, methane, or other greenhouse gases, and can be used to offset greenhouse gases emitted
by the fossil-fuel-fired component of the system generation mix. Consequently, renewable
energy can be considered to be a greenhouse gas emissions-reduction technology as well as a
renewable energy power technology. In addition, should greenhouse gas emission-reduction
12-1
10581090
Greenhouse Gas Emissions Control
mandates be enacted in the future, renewables would become an important component of a

greenhouse gas emissions-reduction strategy and would likely play a key role in carbon dioxide
emissions trading.
The carbon dioxide emissions-reduction potential of a renewable energy power plant is a
function of the fuel mix of the existing generation system. The effective carbon dioxide
emissions-reduction cost is a function the CO2 emission rate and the difference between the
average generation costs of the base system and those of the renewable energy power plant.
12.3 Fossil Carbon Intensity of Fuels
The carbon intensity of generation technologies is measured by the carbon emissions released by
burning a fuel to generate a megawatt-hour of electricity. It depends on the carbon and heat
contents of the fuel as well as the net thermal efficiency of the power generation technology.
Table 12-2 compares fuel heat and carbon content with fossil carbon intensity for coal, oil and
natural gas [1] [2]. Dry wood is included in the table to highlight the zero carbon intensity of
sustainably grown biomass fuel. The data are presented in units of mass of carbon released per
unit of fuel in both mass and higher heating value (HHV) heat content units. Sub-bituminous
coal has the highest carbon intensity per unit energy content (27.77 kg carbon/GJ) and natural
gas has the lowest fossil carbon intensity (13.8 kg carbon/GJ).
Table 12-2
Fossil Carbon Intensity of Coal, Oil, Natural Gas and Wood Fuels
Heat Content—HHV Carbon Content Fossil Carbon Intensity

Fuel Type
(Btu/lb) (MJ/kg) (lb-C/lb) (kg-C/kg) (lb-C/MBtu) (kg-C/GJ)
Coal: Bituminous 13,700 31.798 0.782 0.782 57.08 24.59
Coal: Sub-bituminous 9,000 20.889 0.580 0.580 64.44 27.77
Oil 20,000 46.420 0.92 0.92 46.0 19.8
Natural Gas 23,400 54.311 0.75 0.75 32.1 13.8
Wood (dry) 8,000 18.568 0.45 0.45 0* 0*
*Note: Fossil carbon intensity is the measure relevant to greenhouse gas. By this measure, wood from renewably-grown trees
has zero carbon intensity. If the carbon content of the fuel were put into the same formula used for fossil fuels, the carbon
intensity of wood is 54.2 lb-C/MBtu (23.4 kg-C/GJ).
Table 12-3 compares the impacts of fuel carbon content and net heat rate to the fossil carbon
intensity of selected coal- and natural-gas-fired generation technologies. The data are presented
in units of mass of CO2 and carbon (C) released per megawatt-hour of electricity generated.
Existing coal-fired power plants operating at 34.1% thermal efficiency (HHV) exhibit the highest
generation carbon intensity per megawatt-hour generated (0.26 metric ton C/MWh) while
advanced, natural-gas fuel cell plants operating at 63.7% net thermal efficiency exhibit the
lowest generation carbon intensity (0.08 metric ton C/MWh) (Note that one metric ton = one
tonne).
12-2
10581090
Table 12-3
Carbon Intensity of Generation Technologies
Fuel—Technology Fossil Carbon Emission

Carbon Content Net Heat Rate
(Net HHV Efficiency) CO2 C
(lb/MBtu) (Btu/kWh)
English Units (ton/MWh) (ton/MWh)
Coal
Typical existing (0.341) 56.7 10,000 1.04 0.28
Pulverized, 95% scrubbed 56.7 9,087 0.94 0.26
(0.376)
Advanced, IGCC (0.467) 56.7 7,308 0.76 0.21
Natural Gas
Existing Steam Plant 32.1 10,300 0.61 0.17
(0.331)
Advanced, CC (0.538) 32.1 6,350 0.37 0.10
Advanced, CT (0.427) 32.1 8,000 0.47 0.13
Advanced, fuel cell (0.637) 32.1 5,361 0.32 0.09
Carbon Content Net Heat CO2 C
SI Units
(kg/GJ) Rate (kJ/kWh) (tonne/MWh) (tonne/MWh)
Coal
Typical existing (0.341) 24.43 10,550 0.95 0.26
Pulverized, 95% scrubbed 24.43 9,587 0.86 0.23
(0.376)
Advanced, IGCC (0.467) 24.43 7,710 0.69 0.19
Natural Gas
Existing Steam Plant 13.83 10,867 0.55 0.15
(0.331)
Advanced, CC (0.538) 13.83 6,699 0.34 0.09
Advanced, CT (0.427) 13.83 8,440 0.43 0.12
Advanced, fuel cell (0.637) 13.83 5,656 0.29 0.08
12.4 CO2 Emissions Offsets
12.4.1 Factors Affecting CO2 Emissions Offset
Calculation of the CO2 emissions offset resulting from adding wind, solar, and other renewable
generation to a utility system at first appears to be straightforward and uncomplicated. However,
several indirect factors complicate the calculation and typically result in actual net CO2 emissions
offsets that are lower than those estimated using the straightforward levelized method presented
below. These factors include:
• Renewable generation penetration into the generation mix
• Renewable generation duty cycle characteristics
12-3
10581090
• Hydro, pumped storage, and peaking capacity in the existing generation mix
• Impacts on the dispatch and annual capacity factor of each component of the generating mix
• Possible increased cycling of baseload coal and gas units
• Increased spinning reserves for load following and regulation.
The impacts of these factors vary with the renewable generation type and grid characteristics and
tend to be greatest for wind, which is a variable generation resource and not dispatchable, and
lowest for biomass, geothermal, and hydro, which are typically dispatchable. The impacts for
solar photovoltaic and thermal generation are intermediate between those of wind and the other
renewables, as discussed further below.
12.4.1.1 Wind
In most cases, the daily operating cycle of wind generation does not follow the daily load
demand cycle. For example, peak wind generation may occur at night during the winter and in
the afternoon during the summer. In addition, wind generation can ramp up or down quickly as
weather fronts move by, creating the need to rapidly increase or decrease other generation
connected to the grid in order to follow load. When the penetration of wind generation in the mix
is a few percent, the impacts are usually minimal and the electricity grid can easily absorb the
wind generation. However, if wind penetration rises above 10% to 15%, especially during low
load periods at night, the impacts become more significant. In the latter case, addition of wind
generation usually requires an increase in spinning reserves for load following and regulation,
increased cycling of base load coal units, and possible curtailment of wind generation at night.
Each of these factors decreases the net CO2 emissions offset resulting from adding wind to the
generating mix. In fact, the net emissions offset can even be negative, e.g. the CO2 emissions can
actually increase. For more information, consult recent references that address wind integration
impacts (UWIG-EPRI [3], White [4], and E.ON Netz GMBH [5]).
12.4.1.2 Solar PV and Thermal
Solar PV and thermal generation are also variable resources, but their operating cycles are
predictable and more synchronized with the daily load demand cycle. In addition, solar thermal
generation is usually configured to include backup gas-firing heating or thermal energy storage
to allow the plant to operate during off-peak hours. Thus, with backup gas firing, their actual net
CO2 emissions offsets would be somewhat lower than the levelized emissions offsets presented
below.
12.4.1.3 Biomass, Geothermal, and Hydro
Biomass, geothermal, and hydro units are typically dispatchable and their CO2 emissions offsets
should be close to the levelized offsets presented below.
12-4
10581090
12.4.1.4 Planting Trees and Other Fast-Growing Crops
In addition to the CO2 emissions offsets of renewable generation technologies, plantations of

fast-growing trees and other biomass crops such as eucalyptus, hybrid poplar, and switch grass
act as carbon sinks that absorb CO2 from the atmosphere via photosynthesis.
For example, as discussed in Reference 8, the annual growth of a hybrid poplar plantation in the
Pacific Northwest absorbs 5 to 8 tons of carbon per acre or 18.3 to 29.3 ton CO2/acre per year
(11.2 to 18.0 metric tons C/hectare per year, where one hectare = 10,000 square meters).
As shown in Table 12-3, a coal-fired power plant generates up to 1.04 tons of CO2/MWh (0.95
metric ton carbon/MWh). Therefore, an acre of hybrid poplar would absorb the CO2 produced by
the generation of 17.6 to 28.2 MWh from coal.
A 500-MW coal-fired power plant operating at 85% capacity factor generates 3,723,000
MWh/year and 3,872,000 tons CO2/year. In order to absorb the entire CO2 volume generated by
the plant each year, the required hybrid poplar plantation would cover 132,000 to 211,000 acres,
or 206 to 330 square miles.
12.4.2 Levelized CO2 Emissions Offsets
This section presents charts and other information to estimate the CO2 emissions offsets resulting
from adding renewable generation to an electricity system. The estimates assume that 1) each
MWh of electricity generated by renewable resources replaces 1 MWh of electricity that would
otherwise be generated by the existing generation mix; and 2) the relative contribution of each
component of the mix to generation and CO2 emissions does not change.
Figure 12-1 plots the fossil CO2 emission offset (metric ton CO2/MWh) for coal, oil, and natural
gas generation as a function of the percentage of total kilowatt-hours of generation contributed
by the respective fossil fuel sources. The chart shows separate curves for 100% fossil fuel
generation by coal, oil, and natural gas. These curves provide representative data for illustration
only and can vary greatly depending on the specific generation technology involved (see Table
12-3). The levelized CO2 emission offset for a specific blend of coal, oil, and natural gas
generation is estimated by either interpolating a value between the three curves or by calculating
a weighted average of the values determined by the coal, oil, and gas curves on the chart.
Figure 12-2 shows the levelized CO2 emissions control cost ($/metric ton CO2) as a function of
the cost premium of renewable power relative to the existing system generation cost and the CO2
emissions offset. Figure 12-3 is similar except that it shows the CO2 emission control cost as a
function of the fossil fuel mix and assumes that 50% of the generation is contributed by coal, oil,
and natural-gas-fired technologies.
12-5
10581090
CO2 Emissions Offset vs.% Fossil Fuel Generation

1.0
0.9
0.8
Metric Ton CO2/MWh
0.7
100% Coal
0.6
0.5
100% Fuel Oil
0.4
100% Natural Gas
0.3
0.2
0.1
0.0
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Fossil Fuel Generation, % of Total MWh
Figure 12-1
Levelized CO2 Emissions Offset vs. % Fossil Fuel Generation
CO2 Emissions Control Cost vs. Cost Premium of

300
CO2 Emissions Offset, Metric Ton/MWh
250
0.2
200
$/Metric Ton CO2
0.3
150
0.4
100
0.6
50
1.0
0
-50
-10 0 10 20 30 40 50 60
Cost Premium vs. System Generation Cost, $/MWh
Figure 12-2
Levelized CO2 Emissions Control Cost vs. Cost Premium and CO2 Emissions Offset
12-6
10581090
CO2 Emissions Control Cost vs. Cost Premium

50% Fossil Fuel Generation
250
200
$/Metric Ton CO2
150
100% Natural Gas
100% Fuel Oil
100
100% Coal
50
-50
-100
-10 0 10 20 30 40 50 60
Cost Premium vs. System Generation Cost, $/MWh
Figure 12-3
Levelized CO2 Emissions Control Cost vs. Cost Premium and Fossil Fuel Mix at 50%
Generation from Fossil Fuels
To illustrate the application of the charts, if a fossil generation mix is 70% coal and 30% natural
gas, and fossil fuels supply 90% of the kilowatt-hours generated, Figure 12-1 indicates the CO2
emissions factors for coal and natural gas (at 90%) are about 0.85 and 0.46 metric ton CO2/MWh,
respectively. Thus the weighted CO2 emissions offset factor is (0.85 x 70%) + (0.46 x 30%) =
0.74 metric ton CO2/MWh. If the cost premium is $10/MWh, then the CO2 Emissions Control
Cost is $10/MWh / 0.74 metric ton/MWh = $13.60/metric ton CO2.
12.5 Factors Affecting Generation Cost of Renewable Technologies
Renewable energy generation costs are sensitive to various factors. The following section
discusses the factors that affect the generation cost and, in turn, the levelized carbon dioxide
reduction cost of using a given renewable energy technology. All levelized costs are in constant
end-of-year or third-quarter 2010 U.S. dollars (zero inflation rate). The Federal Production Tax
Credit and Investment Tax Credit are not considered in this analysis.
12.5.1 Wind
Wind generation cost is most sensitive to the annual average wind speed at the site. As wind
speed increases, wind energy generation increases, and thus the unit cost of energy ($/MWh)
decreases. Table 12-4 presents wind plant performance and cost data vs. location. The source of
the data is the 2009 EPRI report, Engineering and Economic Evaluation of Utility-Scale Wind
Power Plants [6].
12-7
10581090
Table 12-4
Wind Plant Performance and Cost vs. Location
(4th Quarter 2010 $, Source: EPRI [6])
Base New York

Location Case California Texas Michigan New York Offshore
Rated Capacity (MW) 100 100 200 150 50 202

Reference Date
4Q 2010 4Q 2010 4Q 2010 4Q 2010 4Q 2010 4Q 2010
of Costs
Total Capital
$2,159 $2,663 $2,101 $2,345 $3,277 $4,240
Requirement $/kW)
Operation &
Maintenance Cost $31.8 $43.0 $31.6 $37.5 $40.0 $131.3
($/kW-yr)
Annual Capacity
35% 33% 31% 28% 29% 45%
Factor
Levelized Cost of
$58.1 $77.9 $64.6 $81.0 $101.7 $110.2
Electricity ($/MWh)
The levelized wind energy cost ranges between about $58/MWh at the 100-MW base case plant
and about $110/MWh at the 202-MW New York offshore wind plant.
Figure 12-4 presents the resulting levelized CO2 emissions control cost vs. system generation
cost at the 100-MW California wind project for five CO2 emissions offset factors between 0.2
and 1.0 metric ton CO2/MWh. The analysis assumes the CO2 emissions offset factor is 0.755
metric ton CO2/MWh, which is consistent with the advanced IGCC coal plant example in Table
12-3 and a representative mix of coal and cleaner natural gas combustion. The levelized CO2
emissions control cost decreases as both the CO2 emissions offset factor and system generation
cost increase, and it becomes negative for system generation costs higher than the $78/MWh
wind generation cost. Negative values of the levelized CO2 emissions control cost correspond to
a net credit for reducing CO2 emissions.
Figure 12-5 presents the levelized CO2 emissions control cost vs. system generation costs for
each of the six wind plant cases described in Table 12-4. The CO2 control cost increases as the
wind generation cost increases and is highest for the 202-MW New York offshore wind plant
and lowest for the 100-MW base case plant.
12-8
10581090
CO2 Emissions Control Cost - Wind Power

100-MW California Project, 76 $/MWh
4th Quarter 2010 $
400
0.2 Emissions Offset (Metric Ton CO2/MWh)
300
$/Metric Ton CO2
0.3
200 0.4
0.6
100
1.0
0
-100
-200
-300
0 20 40 60 80 100 120 140 160
System Generation Cost, $/MWh
Figure 12-4
Levelized CO2 Emissions Control Cost at 100-MW California Wind Plant vs. CO2 Emissions
Offset and System Generation Cost (4th Quarter 2010 $)
CO2 Emissions Control Cost - Wind Power

Emissions Offset: 0.755 Metric Ton CO2/MWh
4th Quarter 2010 $
200
150
NY Offshore 202 MW
100
$/Metric Ton CO2
NY Onshore 50 MW
50
MI Onshore 150 MW
0
-50
CA Onshore 100 MW
-100
TX Onshore 200 MW
-150
Base Case Onshore 100 MW
-200
-250
0 25 50 75 100 125 150 175 200
Figure 12-5
Levelized CO2 Emissions Control Cost vs. Location and System Generation Cost (4th
Quarter 2010 $)
12-9
10581090
12.5.2 Biomass
Biomass generation cost is most sensitive to the delivered biomass fuel cost, the installed cost of
the biomass power system, and the net thermal efficiency of the power cycle. The installed cost
is expected to remain relatively stable over time, but the delivered biomass fuel cost may decline
as new crop planting, cultivation, harvesting, and handling technologies improve. The net
thermal efficiency of the power cycle has the potential to improve as well.
Table 12-5 summarizes the total capital requirement, fixed and variable O&M, plant
performance, and levelized cost of electricity for a 60-MW, 100% biomass repowered , a 250-
MW biomass-cofired plant (at 10% heat input from biomass), and a 50-MW biomass-fired
bubbling fluidized bed boiler power plant.
Figure 12-6 shows the levelized CO2 emissions control cost for the three cases as a function of
the system generation cost. The source of the data is the 2010 EPRI report, Engineering and
Economic Evaluation of Biomass Power Plants [7]. Assuming the system generation cost is
$50/MWh, the levelized CO2 cost ranges between minus $25/metric ton CO2 for the 10%
biomass-cofired plant and plus $101/metric ton CO2 for the 100% biomass-fired bubbling
fluidized bed plant. The cofiring cost is based on the 25-MW contribution of the biomass fuel to
the 250-MW output of the coal-fired plant and incremental cost of the biomass fuel vs. coal. A
negative CO2 control cost corresponds to a net credit for reducing CO2 emissions.
Table 12-5
Levelized Cost of Electricity for 50-MW Biomass-Fired Stoker and Fluidized Bed Boiler
Power Plants ($/MWh, 3rd Quarter 2010 $, Source: EPRI [7])
100%
10% Biomass 100% Biomass
Biomass
Cofiring Bubbling FBC
Repowering
Total Capital Requirement 1,970 1,690 5,590

Fixed OM 132 21.6 113
Variable O&M 3.5 4.9 5.8

Btu/kWh 12,600 9,760 13,964
GJ/kWh 13,284 10,290 14,723
Fuel $/GJ 3.37 0.24 3.37
Levelized COE $/MWh
Capital 22 20 56
Fixed O&M 18 3 15
Variable O&M 3 5 6
Fuel 45 2 50
TOTAL 88 31 126
12-10
10581090
CO2 Emissions Control Cost - Biomass Power

Biomass Fuel Cost for Repowering and FBC: 3.74 $/GJ
Incremental Fuel Cost for Cofiring: + 0.26 $/GJ
CO2 Emissions Offset: 0.755 metric ton CO2/MWh
200
100% Biomass Bubbling FBC (50 MW)
150
$/Metric Ton CO2
100% Biomass Repowering (60 MW)

100
50
-50
10% Biomass Cofiring (250/25 MW)
-100
-150
0 20 40 60 80 100 120 140
Figure 12-6
Levelized CO2 Emissions Control Cost vs. System Generation Cost for 100% Biomass-
Repowered and Bubbling Fluidized Bed Plants and 10% Biomass Co-fired Coal Plant
(Incremental Cost vs. Coal for Cofired Plant, 3rd Quarter 2010 $,Source: EPRI [7])
12.5.3 Solar Photovoltaics
Photovoltaic (PV) generation cost is most sensitive to the annual insolation rate at the site and
the installed cost of the PV system. Installed cost is expected to continue to decline over time as
new technology is developed and commercialized. Table 12-6 presents the 20-year levelized cost
of electricity for 10-MW central station solar photovoltaic power plants at each combination of
four locations and three PV technologies. The locations include Las Vegas, Nevada; Alamosa,
Colorado; Jacksonville, Florida; and Columbus, Ohio. The PV technologies include fixed flat-
plate and single-axis tracking modules using crystalline silicon cells and two-axis tracking
concentrating PV using multijunction cells. The estimated levelized cost of electricity ranges
between $247/MWh for the fixed flat-plate CdTe system at Las Vegas and $421/MWh for the
single-axis tracking c-Si system at Columbus. The source of the data is the 2009 EPRI report,
Engineering and Economic Evaluation of Central-Station Photovoltaic Power Plants. [7]
Figure 12-7 shows the expected sensitivities of the levelized CO2 emissions control costs for the
six technologies to the system generation cost at the Las Vegas site (top) and the Jacksonville
site (bottom). The CO2 control cost decreases with system generation cost. The analysis assumes
the CO2 emissions offset factor is 0.755 tonne CO2/MWh.
12-11
10581090
Table 12-6
Performance and Cost Estimates for 10-MW Solar Photovoltaic Power Plants (4th Quarter
2010 $, Source: EPRI [8])
Fixed Flat- Fixed Flat- Fixed Flat-

Plate a-Si Plate CdTe Plate c-Si
Rated Capacity 10 MW 10 MW 10 MW
Total Capital Requirement ($/kW) $3,834 $3,721 $4,027

$53 $53 $48
($/kW-yr)
First-Year Capacity Factor

Las Vegas, Nevada 22.6% 22.2% 22.4%
Alamosa, Colorado 19.1% 20.7% 21.1%
Jacksonville, Florida 17.0% 16.9% 16.9%
Columbus, Ohio 14.6% 14.6% 14.4%
Levelized Cost of Electricity ($/MWh)

Las Vegas, Nevada $249 $247 $258
Alamosa, Colorado $294 $264 $275
Jacksonville, Florida $330 $324 $342
Columbus, Ohio $384 $380 $402
1-Axis Tilted 1-Axis 2-Axis

Tracking c-Si Tracking c-Si Tracking CPV
Rated Capacity 10 MW 10 MW 10 MW
Total Capital Requirement ($/kW) $4,557 $5,066 $4,937

$62 $62 $67
($/kW-yr)
First-Year Capacity Factor

Las Vegas, Nevada 25.0% 26.9% 27.7%
Alamosa, Colorado 25.3% 26.8% 27.9%
Jacksonville, Florida 18.4%
Columbus, Ohio 15.8%
Levelized Cost of Electricity ($/MWh)

Las Vegas, Nevada $267 $258 $260
Alamosa, Colorado $263 $275 $258
Jacksonville, Florida $362
Columbus, Ohio $421
12-12
10581090
CO2 Emissions Control Cost - Solar PV

10-MW PV Plant, CO2 Emissions Offset: 0.755 metric ton
December 2010 $
400
Las Vegas, NV
350
$/Metric Ton CO2
300
1-Axis cSi
250
200
150
Fixed CdTe
100
50
0
0 25 50 75 100 125 150 175 200
CO2 Emissions Control Cost - Solar PV

10-MW PV Plant, CO2 Emissions Offset: 0.755 metric ton/MWh
December 2010 $
600
Columbus, OH
500
$/Metric Ton CO2
1-Axis cSi
400
Fixed CdTe
300
200
100
0
0 25 50 75 100 125 150 175 200
Figure 12-7
Levelized CO2 Emissions Control Cost vs. System Generation Cost for 50-MW Solar
Photovoltaic Plants Las Vegas, Nevada (Top) and Columbus, Ohio (Bottom) (December
2010 $, Source: EPRI [8])
12-13
10581090
12.5.4 Geothermal
Geothermal generation cost is most sensitive to the geothermal energy production and the
installed cost of the geothermal wells and power system.
Table 12-8 presents the 20-year levelized cost of electricity for 50-MW flash-steam and 50-MW
binary-cycle geothermal plants. The data are from the 2010 EPRI report, Engineering and
Economic Evaluation of Geothermal Power Plants [9].The levelized COE is $66/MWh and
$70/MWh for the flash-steam and binary-cycle geothermal plants.
Figure 12-8 shows the resulting CO2 emission control costs vs. the base system generation cost
for the 50-MW flash-steam and 20-MW binary-cycle geothermal plants. The analysis assumes
the CO2 emissions offset factor is 0.717 tonne CO2/MWh, including the 95% adjustment for the
CO2 emissions released during the production of geothermal power.
Table 12-7
Levelized Cost of Electricity for 50-MW Flash-Steam and Binary-Cycle Geothermal Power
Plants (December 2010 $, Source: EPRI [9])
Flash Binary
Geothermal Technology
Steam Cycle
Rated Capacity, MW 50 50
Capital Cost, $/kW $4,989 $5,192

Fixed O&M, $/kW-yr $76.4 $80.6
Variable O&M, $/kW-yr $9.60 $10.50
Capacity Factor 96% 96%
Levelized COE, $/MWh
Capital @ 8.3 %/yr $47.7 $52.4

O&M $9.1 $9.6
Fuel Cost $/GJ $9.6 $7.6

TOTAL $66.4 $69.6
Figure 12-8 shows that, as the base system generation cost increases, the CO2 emissions control
cost decreases and becomes negative for generation cost greater than about $66/MWh for flash-
steam and $70/MWh for binary-cycle geothermal plants.
12-14
10581090
CO2 Emissions Control Cost - Geothermal Power

Basis: MW Binary Cycle and Flash-Steam Plants
December 2010 $
CO2 Emissions Offset: 0.755 metric ton CO2/MW x 95%
120
100
80
$/Metric Ton CO2
60
50-MW Binary-Cycle
40
20
50-MW Flash-Steam
0
-20
-40
-60
0 20 40 60 80 100
Base System Cost, $/MW
Figure 12-8
Levelized CO2 Emissions Control Cost for 50-MW Flash-Steam and Binary-Cycle Power
Plants vs. Base System Generation Cost (December 2010 $, Source: EPRI [9])
12.5.5 Solar Thermal
Solar thermal generation cost is highly sensitive to the annual insolation rate at the site and the
installed cost of the solar thermal system. The installed cost is expected to continue to decline
over time as new technology is developed and commercialized.
Table 12-8 presents the levelized cost of electricity for four solar-thermal power technologies.
They include 250-MW parabolic trough plants with and without 6-hour thermal energy storage,
100-MW tower with 10-hour storage, and 100-MW dish-Stirling plants. The data assume 30-
year project life and constant December 2010 dollars, and do not include the federal Production
Tax Credit or Investment Tax Credit. The source of the data is the 2010 EPRI report, Solar
Thermal Technology Status and Performance and Cost Estimates–2010 [10].
12-15
10581090
Table 12-8
Constant-Dollar Levelized Cost of Electricity for Solar Thermal Technologies (December
2010 $)
Central
Parabolic Parabolic
Receiver:
Trough: Trough:
10-hr Dish Stirling
No Thermal 6-hr Thermal
Thermal
Storage Storage
Storage
Rated Capacity, MW 250 250 100 100
Total Capital Requirement, $/kW 4,070 6,160 6,510 4,540
Fixed O&M $/kW-yr 64 68 66 62
Variable O&M, $/MW-hr 0 0 0 0
Net Heat Rate, GJ/kWh 0 0 0 0
Fuel, $/GJ 0 0 0 0
Capacity Factor 25.4% 40.7% 49.1% 23.7%
Levelized COE $/MWh
Capital @ 6.68%/yr 210 199 174 250
O&M 45 30 24 47
Fuel 0 0 0 0
TOTAL 255 229 198 297
4,000 x 25
Plant Size (no. of units x unit size) 2 x 125 MW 2 x 125 MW 100 MW
kW (100 MW)
Capacity Factor 25.4% 40.7% 49.1% 23.7%
LCOE with ITC, $/MWh 116 100 86 133
FOM Portion of LCOE 35 23 19 36
LCOE without ITC, $/MWh 196 175 152 228
FOM Portion of LCOE 35 23 19 36
Figure 12-9 presents the resulting levelized CO2 emissions control costs for the four solar
thermal power plant technologies. The analysis assumes the CO2 emissions offset factor is 0.755
metric ton CO2/MWh. The CO2 emissions control cost decreases with increasing system
generation cost. The solar thermal power tower with 10-hour thermal energy storage has the
lowest CO2 emissions control cost and the dish-Stirling plant has the highest cost.
12-16
10581090
CO2 Emissions Control Cost - Solar Thermal Technologies

Net CO2 Emissions Offset: 0.755 metric ton CO2/MWh
Basis: December 2010 $
400
300
Dish Stirling
$/Metric Ton CO2
Parabolic Trough
200
100
Tower
10-hr Storage
0
Parabolic Trough
6-hr Storage
-100
0 25 50 75 100 125 150 175 200
Figure 12-9
Levelized CO2 Emissions Control Cost for Four Solar Thermal Technologies (December
2010 $, Source: EPRI [10])
12.6 References
1. Greenhouse Gas Reduction with Renewables. EPRI, Palo Alto, CA: 2000. TR-113785.
2. Renewable Energy Technology Characterizations. EPRI: 1997. TR-109496.
3. Characterizing the Impacts of Significant Wind Generation Facilities on Bulk Power System
Operations Planning: Utility Wind Interest Group - Xcel Energy-North Case Study.
EPRI, Palo Alto, CA: 2003. 1004807.
4. White, David J., “Danish Wind: Too Good to be True?”, The Utilities Journal, pp 37-39,
July 2004.
5. Wind Report 2004, E.ON Netz Gmbh, Bayreuth, Germany 2003.
6. Engineering and Economic Evaluation of Utility-Scale Wind Power Plants. EPRI, Palo Alto,
CA: 2009. 1017599.
7. Engineering and Economic Evaluation of Biomass Power Plants, 100% Biomass

Repowering, Biomass Co-firing, and Bubbling Fluidized Bed Biomass Combustion. EPRI,
Palo Alto, CA: 2010. 1019762.
12-17
10581090
8. Engineering and Economic Evaluation of Central-Station Solar Photovoltaic Plants. EPRI,

Palo Alto, CA 2009 1017600.
9. Engineering and Economic Evaluation of Geothermal Power Plants. EPRI, Palo Alto, CA:
2010 1019761.
10. Solar Thermal Technology Status and Performance and Cost Estimates–2010. EPRI, Palo
Alto, CA: 2010. 1022504.
11. Stanton, B., Eaton, J., Johnson, J., Rice, D., Schutte, B., and Moser, B., “Hybrid Poplar in the
Pacific Northwest, the Effects of Market-Driven Management,” Journal of Forestry, June
2002.
12-18
10581090
10581090
Export Control Restrictions The Electric Power Research Institute Inc., (EPRI, www.epri.com)
Access to and use of EPRI Intellectual Property is granted with the spe- conducts research and development relating to the generation, delivery
cific understanding and requirement that responsibility for ensuring full and use of electricity for the benefit of the public. An independent,
compliance with all applicable U.S. and foreign export laws and regu- nonprofit organization, EPRI brings together its scientists and engineers
lations is being undertaken by you and your company. This includes as well as experts from academia and industry to help address challenges
an obligation to ensure that any individual receiving access hereunder in electricity, including reliability, efficiency, health, safety and the
who is not a U.S. citizen or permanent U.S. resident is permitted access environment. EPRI also provides technology, policy and economic
under applicable U.S. and foreign export laws and regulations. In the analyses to drive long-range research and development planning, and
event you are uncertain whether you or your company may lawfully supports research in emerging technologies. EPRI’s members represent
obtain access to this EPRI Intellectual Property, you acknowledge that it more than 90 percent of the electricity generated and delivered in the
is your obligation to consult with your company’s legal counsel to deter- United States, and international participation extends to 40 countries.
mine whether this access is lawful. Although EPRI may make available EPRI’s principal offices and laboratories are located in Palo Alto, Calif.;
on a case-by-case basis an informal assessment of the applicable U.S. Charlotte, N.C.; Knoxville, Tenn.; and Lenox, Mass.
export classification for specific EPRI Intellectual Property, you and your
Together...Shaping the Future of Electricity
company acknowledge that this assessment is solely for informational
purposes and not for reliance purposes. You and your company ac-
knowledge that it is still the obligation of you and your company to make
your own assessment of the applicable U.S. export classification and
ensure compliance accordingly. You and your company understand and
acknowledge your obligations to make a prompt report to EPRI and the
appropriate authorities regarding any access to or use of EPRI Intellec-
tual Property hereunder that may be in violation of applicable U.S. or
foreign export laws or regulations.
Program:
© 2010 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power
Research Institute, EPRI, and TOGETHER...SHAPING THE FUTURE OF ELECTRICITY are
registered service marks of the Electric Power Research Institute, Inc.
1019760
Electric Power Research Institute

3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94303-0813 USA
800.313.3774 • 650.855.2121 • askepri@epri.com • www.epri.com
10581090

Epri Wind Systems

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Epri Wind Systems

Hochgeladen von

Copyright:

Verfügbare Formate

2010 TECHNICAL REPORT

Renewable Energy Technology Guide: 2010

Final Report, December 2010

EPRI Project Manager

Other EPRI Staff Contributing to this Report

ELECTRIC POWER RESEARCH INSTITUTE

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER

THE FOLLOWING ORGANIZATION PREPARED THIS REPORT:

Electric Power Research Institute (EPRI)

Results and Findings

Challenges and Objectives

Applications, Value, and Use

2 ECONOMIC METHODOLOGY AND ASSUMPTIONS ..........................................................2-1

4 BIOMASS ENERGY ...............................................................................................................4-1

5 SOLAR PHOTOVOLTAICS ...................................................................................................5-1

6 GEOTHERMAL ENERGY ......................................................................................................6-1

8 OCEAN TIDAL ENERGY .......................................................................................................8-1

9 OCEAN WAVE ENERGY .......................................................................................................9-1

10 RIVER IN-STREAM ENERGY ............................................................................................10-1

11 RENEWABLE ENERGY GRID INTEGRATION TECHNOLOGIES ...................................11-1

12 GREENHOUSE GAS EMISSIONS CONTROL ..................................................................12-1

Energy Source 2004 2005 2006 2007 2008 2009

94.60 Quadrillion Btu/year 7.74 Quadrillion Btu/yr

3,814 Billion kWh/yr 379 Billion kWh/yr

Natural Gas Nuclear Power

Wind 36,698 Est. 180,000

Parabolic Trough 421 950

Dish/Engine <1 2.5

Central Receiver <1 45

Compact Linear Fresnel Reflector 5.0 9.4

Low-Impact and Emerging Hydro:

Ocean Wave Energy < 0.1 <1

1.2 EPRI’s Role

1.4 Definitions and Units

The RETG consists of 12 chapters. Following Chapter 1: Introduction, they are:

2.1 Regulated Utility Power Projects

2.1.1 Fixed Charge Rates

2.1.1.1 Annual Fixed Charge Rates

2.1.3 Other Capital Expenses—Capital Additions

2.1.4 Calculating Costs per Kilowatt-Hour

2.1.4.1 Levelized Costs per Kilowatt-Hour

Comparing costs on a levelized or a present-value-of-revenue-requirements basis is appropriate

Levelized busbar costs are calculated by the following steps:

2.2 Non-Utility Generator Power Projects

2.2.1 Types of Non-Utility Generators

An understanding of the economic evaluation methods for non-utility projects requires an

2.2.1.1 PURPA Cogenerators and Small Power Producers

2.2.1.2 Exempt Wholesale Generators

Other major provisions of the EPAct92 include:

2.2.1.3 Merchant Power Plants

2.2.2 Development of an Economic Pro Forma for a Merchant Plant

2.2.2.1 Differences Between Regulated Utility and Non-Utility Power Projects

2.2.2.1.2 Income Taxes and Depreciation

2.2.2.1.3 Capital Expenditures

2.2.2.1.4 Debt Financing

2.2.2.1.5 Economic Methodology

2.2.3 Conceptualizing the Analysis

2.2.3.1 Total Capital Requirement

2.2.3.1.1 Construction Cost

Total Project $1,000 0 485,280 485,280 100.0%

Owner's Cost $1,000 0 0 2,045 2,045 10 0.4%

Working capital $1,000 0 0 6,260 6,260 31 1.3%