Bihar Smart Energy Access

e[r] cluster
for a
smart energy access
THE ROLE OF MICROGRIDS IN PROMOTING THE INTEGRATION OF RENEWABLE ENERGY IN INDIA
2
imprint
Energynautics
Nis Martensen, Rena Kuwahata,
Dr.Thomas Ackermann
acknowledgments: Nils Goldbeck,
Anupam Kumar Sinha
Greenpeace
Manish Ram, Ramapati Kumar,
Sven Teske
Energynautics GmbH
Mhlstrae 51
63225 Langen, Germany
Telephone: +49 (0) 61 51 - 785 81 01
t.ackermann@energynautics.com
Greenpeace International
Ottho Heldring Straat 5, AZ 1060
Amsterdam, The Netherlands
Sven.teske@greenpeace.org
3
recognize the need for partnerships
among governments, enterprises,
civil society and the people to create the conditions
for a successful transition to the much desired
and required clean energy system.
Greenpeace India Society
60 Wellington Road, Richmond Town
Bangalore 560025
Tel: +91-80-41154861
Fax: +91-80-41154862 E-mail:
info@greenpeaceindia.org Visit: www.
greenpeaceindia.org
Ramapati Kumar, Manish Ram
Date
May 2012
Copyright Information
Cover Photo Harikrishna Katragadda
/Greenpeace
Solar Photovoltaic Power Plant
in Tangtse
The 100 kWp stand-alone solar
photovoltaic power plant at Tangtse,
Durbuk block, Ladakh. Located 14,500
feet AMSL in the Himalaya, the plant
supplies electricity to a clinic, a school
and 347 houses in this remote location,
for around fve hours each day.

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
4
contents
introduction and policy recommendations 6
Greenpeace Proposal to support a renewable energy]cluster 8
executive summary 10
1
access to energy the concept 18
Rural electrifcation for universal electrifcation 19
Grid extension the passive top-down approach 19
Off-grid power generation
the active bottom-up approach 19
What is the problem with the current process? 20
Integration of renewable energies 20
Smart Energy Access 21
Microgrids 21
The Concept 22
2
microgrid design case study 24
Methodology 25
Overview of Bihar 26
Bihar Microgrid Village 27
Step 1: Assessment of Renewable Energy Resources in Bihar 29
Step 2: Assessment of System Demand 31
The resulting system demand for each demand scenario
is summarised in 32
Step 3: System sizing by production optimisation with Homer 33
Unit size considerations 34
Solar PV price sensitivity 34
HOMER Simulation Results 35
Conclusions from this section 39
Step 3: Technical feasibility check 40
Load-Related Assumptions and Network Design 40
Load distribution 42
Reactive Power Demand 43
Power Flow Analysis 44
Step 4: Development of a Control Strategy
for the Switching Operation 49
The simple solution 49
Control issues 50
Cost comparison with grid extension 54
Microgrid scale-up across the state of Bihar 55
3
grid technologies and defnitions 58
Energy Access Options 59
Energy Homes Systems 59
Off-grid Distribution Systems 59
Microgrids 60
Distributed Generation Systems 60
Obstacles to Energy access Projects 60
Strengths and Weaknesses of Energy Access Options 61
4
references 64

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
solar photovoltaic power plant in tangtse
THE 100 KWP STAND-ALONE SOLAR PHOTOVOLTAiC
POWER PLANT AT TANGTSE, DuRBuK BLOCK, LADAKH.
LOCATED 14,500 FEET AMSL iN THE HiMALAyA,
THE PLANT SuPPLiES ELECTRiCiTy TO A CLiNiC, A
SCHOOL AND 347 HOuSES iN THiS REMOTE LOCATiON,
FOR AROuND FiVE HOuRS EACH DAy.
5
foreword
Energy is central to nearly every major challenge and opportunity
the world faces today. Access to energy for all is essential for
creating jobs, developing economies, increasing incomes, food
production, security or climate change. Approximately one and a half
billion people around the world, still lack access to electricity and
the majority of them live in developing nations. Countries like india
still face a severe energy crunch, with millions of rural inhabitants
still lacking access to secure and assured electricity, making this a
development priority.
The energy shortage is most acute among indias rural poor and in
states such as Bihar, where more than 80% of the population still
live in the rural areas; the majority of them still rely on conventional
and traditional fuels for their energy needs. As governments
and the energy sector seek to provide more modern and reliable
energy services to these communities, there is a growing gap in the
ever increasing demand and the shrinking supply of energy from
conventional sources, as these are fast depleting. On the other hand,
a fedgling market in cleaner, more effcient energy delivery models
is emerging. Successful (though small scale) business models such
as solar-based home electricity systems and electricity services
generated from decentralized sources such as micro hydro and
biomass gasifers are increasingly fnding a market among such
households.
To facilitate energy access and a clean energy transition, a new
approach has to be thought about. Sustainable technologies
and innovative business models have to be implemented with
governments, businesses, investors, and civil society coming together
to solve public problems. Supplying modern energy services to the
millions who now lack electricity and clean fuels is not just a social
imperative but also essential for distributive justice in a growing
economy. it can create millions of new jobs and increase productivity
in rural areas, thus bringing about sustainable prosperity.
This report shows how microgrids could be a useful concept for
accelerating current rural electrifcation efforts, support renewable
energy integration into energy systems, and create a reliable
electricity system. There could be divergence of views on the
approaches to be adopted for achieving this object. However we
should all be unanimous in recognizing the need for partnerships
among governments, enterprises, civil society and the people to
create the conditions for a successful transition to the much desired
and required clean energy system.
May 2012
G M Pillai
Founder- Director General
Pune, World institute of Sustainable Energy (WiSE)

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
Child in Ladakh
CHiLD AT THE SAMSKiT TSONGPA (WOMENS
GROuP) iN PuNPuN HAMLET, DuRBuK
BLOCK, LADAKH. THE GROuP DRAWS
ELECTRiCiTy FROM A SOLAR POWER PLANT
iN DuRBuK.
R]CLuSTER FOR A SMART ENERGy ACCESS
THE ROLE OF MiCROGRiDS iN PROMOTiNG THE iNTEGRATiON OF RENEWABLE ENERGy iN iNDiA
introduction
and policy
recommendations
from greenpeace
phunchok in the tangtse gompa iN DuRBuK BLOCK,
LADAKH. THE ROOM HAS TAKEN SEVEN yEARS
TO BuiLD AND PAiNT, AND WiLL BE COMPLETED
THiS yEAR. HiGH ON THE CLiFF, THE GOMPA
RECEiVES ELECTRiCiTy FROM THE 100 KWP SOLAR
PHOTOVOLTAiC POWER PLANT iN TANGTSE ViLLAGE
BELOW, AS DO A CLiNiC, A SCHOOL, AND 347 HOuSES.

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
7
i
n
t
r
o
d
u
c
t
i
o
n

a
n
d

p
o
l
i
c
y

r
e
c
o
m
m
e
n
d
a
t
i
o
n
s

f
r
o
m

g
r
e
e
n
p
e
a
c
e
access to energy in 2012:
the international year of sustainable energy for all
in December 2010, the united Nations General Assembly declared
that 2012 would be the international year of Sustainable Energy for
All; access to modern affordable energy services in developing
countries is essential for the achievement of the internationally
agreed development goals, including the Millennium Development
Goals
1
, and sustainable development, which would help to reduce
poverty and to improve the conditions and standard of living for the
majority of the worlds population.
The General Assemblys Resolution 65/151 called on uN Secretary-
General Ban Ki-Moon to organize and coordinate activities during
the year in order to increase awareness of the importance of
addressing energy issues, including access to and sustainability of
affordable energy and energy effciency at local, national, regional
and international levels.
in response, a new global initiative, Sustainable Energy for All,
was launched at the uN General Assembly in September 2011,
along with its own High Level Group, designed to mobilize action
from governments, the private sector and civil society globally.
The initiative has three inter-linked objectives: universal access to
modern energy services, improved rates of energy effciency, and
expanded use of renewable energy sources.
in the light of this opportunity Greenpeace initiated a renewable
energy project in the state of Bihar, a rural area of northern india
where over 82% of the population lacks access to electricity
2
. The
idea was to develop a smart grid for rural electrifcation beginning by
supplying only basic electricity needs to the area. However over time
this can be expanded, as demand and supply grows, step by step to
a full 24/7 supply with 100% renewables. unlike the current rural
electrifcation concepts, this research aimed to develop a system which
functions as a bottom-up grid expansion. To help with this project,
Greenpeace commissioned the German engineering frm Energynautics
to present the technical possibilities and options both for a village and
for an entire state based on this concept. Greenpeace publishes since
2005 global, regional and national energy scenarios under the name
Energy [R]evolution. This report continues the sucessfull series with
the frst ever energy [r]evolution concept for villages. Therefore we
call the concept presented in this report E[R] cluster.
The traditional rural electrifcation concept cannot be expanded
in a bottom-up manner to a full grid system, but generally leaves
regional systems as island systems. However, villages enjoying
such rural electrifcation programmes often lost priority for grid
connection programmes. Therefore many communities rather wanted
to wait till the expanded grid reached their village. in this report we
present a technical concept on how to use renewable energy clusters
to organize bottom up electrifcation across the entire state of Bihar.
Greenpeace hopes to inspire other developing countries with this
concept. However, in order to help implementing renewable energy
clusters, policy changes and specifc national and international
support programs are required.
policy recommendations for renewable energy
cluster
The need for an integrated approach to rural electrifcation is an
imperative for developing nations such as india, and in context for
states like Bihar to progress economically. Energy capacity increase
should align with other developments and electricity infrastructure
strategies. The elements of policy, governance and capacities
suggested in this section are based on the principle of inclusion,
states increasing investment capacity, generation of developmental
benefts and reduction of environmental damage. The proposed
framework takes a bottom-up approach to achieve universal
electrifcation; this requires extensive institutional and government
intervention. The vision for overall energy-led development for the
State of Bihar emerges as follows:
a state-wide
network of decentralised
energy plants (stand
alone and micro-grids),
developed with support
from state government
agencies in collaboration
with private entrepreneurs
for a high impact and
accelerated economic
development of the
state
The most important step in this direction would be to develop a

resource and technology plan for rural electrifcation,
Identifypopulation,energyneedsatdifferenttimesoftheyear,
investment required, local investment possibilities and fnancial
capacities of the population.
Identifynatureandamountofrawmaterialavailablefor
decentralised renewable energy systems at different times of the
year.
8
i
n
t
r
o
d
u
c
t
i
o
n

a
n
d

p
o
l
i
c
y

r
e
c
o
m
m
e
n
d
a
t
i
o
n
s

f
r
o
m

g
r
e
e
n
p
e
a
c
e
Analyseanddevelopfeasibletechnologiesandlocationsemerging
from the above- mentioned two points.
Mapthelocationsindistrictsandblocksandenlistthe
government infrastructure support, required and available.
The changes are recommended at the policy level, supported by
various implementation processes and structures.
Clusterbaseddevelopment:Clubbingofareastoformeconomic
clusters.
Allinfrastructuredevelopmenttobegearedfortechnology-related
development.
Amajorthrustonprovidingmarketlinkagesbasedonproduct
cluster formation and effective strengthening of the credit and
marketing links for security of the entrepreneur.
Trainingofrelevantgovernmentpersonnelforcommitment
and institutional support to the entrepreneurs and need for a
semi-autonomous cell or unit in the Department of industry to
supervise, coordinate and promote rural industrialisation.
Financeresourcebuildingatthegovernmentandprivatelevel
by the establishment of a council for rural industrialisation and
establishment of related coordination committees at district level.
Integratedenergyplanning:state-widenaturalresourcesand
appropriate technology.
Stateincentivisedemergenceofnewprivateenterprisemodels,
particularly energy services companies (ESCOs), developed
through innovative and alternative business opportunities in the
electricity sector by use of decentralised generation
Empowermentofrelevantgovernmentagenciestodrivethe
expansion of decentralised energy.
Showcasingthestateasanevolveddecentralisedenergymarket,
nation-wide and worldwide.
Transformthestateseconomicsandenergysupplythroughuseof
decentralised energy by creating facilitative policy structures.
Createsupportiveinstitutionalavenuesformicro-grids.
Institutionalisesupportiveregulatorystructuresandencourage
governance support through the Electricity Regulatory
Commission.
Developinnovativeenergypricingmodelsbymixingregulationand
competition.
This can be the mandate of district administrations and the Bihar
Rural Electrifcation Corporation (REC). The district level plans
will identify the areas having raw material potential for their own
energy systems outside the grid. The plan will also defne the areas
(state-wide), having economic viability for private (single, multiple,
or cooperative) ownership of electricity systems. This investigation
will provide the state with a clear district/area-wise plan to meet
the electricity needs of the state. However, infrastructure support,
fnancial assistance and policy support is required for this plan
to work effectively. The other most important step to take would
be to design a tariff mechanism to encourage micro grids and
decentralised systems. Tariff and pricing structure can have a
signifcant impact on the economic viability of distributed energy
system. Therefore, a fair tariff and pricing system is essential. The
pricing system should include:-
ThecontractualagreementbetweentheERCandtheDEoperator
should supersede the basic rate set by the ERC for the public
utility. in the early stages of any new technology, market-based
pricing can make the new system uneconomic. Hence, to encourage
DEs it is essential for the regulator to have separate tariff setting,
which
Includesanygrantsfromnationalandinternationalagencies.
Includesincentiveslabsbasedonamountofgeneration(system
effciency and system reliability).
Includesincentiveslabsbasedonsupply(numberofcustomers
served).
Canbeseasonalorperiodicdependingonthecostcycleofthe
DEs (the time frame and tariff should strike a balance between
the raw material price fuctuations and an affordable price to the
rural consumer).
Defningthegrid-DEinterconnectionstandardstoavoidanytariff
manipulations, for instance, either prohibiting access to the grid
by providing expensive power (by the utility) or by supplying to the
grid at a signifcantly high cost (by the operator).
MechanismssuchastheFTSM(FeedinTariffSupport
Mechanism) have to be explored to support the large scale
expansion of micro grids based on renewable energy technologies
which is presented in detail below.
greenpeace proposal to support a
renewable energy]cluster
This energy cluster system builds upon Greenpeaces Energy [R]
evolution scenario
3
which sets out a global energy pathway that not
only phases out dirty and dangerous fossil fuels over time to help
cut CO2 levels, but also brings energy to the 2billion people on the
planet that currently dont have access to energy. The most effective
way to ensure fnancing for the energy [r]evolution in the power
sector is via Feed-in laws.
To plan and invest in an energy infrastructure, whether for
conventional or renewable energy, requires secure policy frameworks
over decades. The key requirements are:
a. long term security for the investment
The investor needs to know the pattern of evolution of the energy
policy over the entire investment period (until the generator is paid
off). investors want a good return of investment and while there
is no universal defnition of a good return, it depends on the long
term proftability of the activity as well as on the infation rate of the
country and the short term availability of cash throughout the year
to sustain operations.
b. maximize the leverage of scarce fnancial resources
Access to privileged credit facilities, under State guarantee, are one
of the possible instruments that can be deployed by governments
9
i
n
t
r
o
d
u
c
t
i
o
n

a
n
d

p
o
l
i
c
y

r
e
c
o
m
m
e
n
d
a
t
i
o
n
s

f
r
o
m

g
r
e
e
n
p
e
a
c
e
to maximise the distribution of scarce public and international
fnancial resources, leverage on private investment and incentivize
developers to rely on technologies that guarantee long term fnancial
sustainability.
c. long-term security for market conditions
The investor needs to know if the electricity or heat from the power
plant can be sold to the market for a price which guarantees a
good return of investment (ROi). if the ROi is high, the fnancial
sector will invest; if it is low compared to other investments then
fnancial institutions will not invest. Moreover, the supply chain
of producers needs to enjoy the same level of favourable market
conditions and stability (e.g. agricultural feedstock).
d. transparent planning process
A transparent planning process is key for project developers, so
they can sell the planned project to investors or utilities. The entire
licensing process must be clear, transparent and fast.
e. access to the (micro) grid
A fair access to the grid is essential for renewable power plants. if
there is no grid connection available or if the costs to access the grid
are too high the project will not be built. in order to operate a power
plant it is essential for investors to know if the asset can reliably
deliver and sell electricity to the grid. if a specifc power plant (e.g.
a wind farm) does not have priority access to the grid, the operator
might have to switch the plant off when there is an oversupply from
other power plants or due to a bottleneck situation in the grid. This
arrangement can add high risk to the project fnancing and it may
not be fnanced or it will attract a risk-premium which will lower
the ROi.
a rural feed-in tariff for bihar
in order to help implement the E[R] clusters in Bihar, Greenpeace
suggests starting a feed-in regulation for the cluster, which will be
partly fnanced by international funds. The international program
should add a CO2 saving premium of 10 indian Rupee (iNR) per
kWh for 10 years. This premium should be used to help fnance the
required power generation as well as the required infrastructure
(grids). in the table ES the CO2 savings, rough estimation of
employment effects as well as the required total funding for the CO2
premium for the state of Bihar are shown.
e[r] cluster jobs
While the employment effect for the operation and maintenance
(O&M) for solar photovoltaics (0,4/MW), wind (0,4/MW), hydro
(0,2/MW) and bio energy (3,1/MW) are very well documented,
4

the employment effect of grid operations and maintenance are not.
Therefore Greenpeace assumed in this calculation that for each
100GWh one job will be created. This number is based on grid
operators in Europe and might be too conservative. However it is
believed that the majority of the jobs will be created by the O&M of
power generation; grid operation may be part of this work as well.
Due to the high uncertainty of employment effects from grid
operation, these numbers are only indicative.
table ES: key results for e[r] villagecluster - state of bihar (rural) - emplyoment. environment + ft
ScEnario EMPloyMEnt co2 SavinGS Fit
GENERATiON GRiD TOTAL SPECiFiC TOTAL AVERAGE ACCROSS ALL TECHNOLOGiES
JOBS JOBS JOBS t CO2 /GWh million t CO2/a iNR /kWh
ScEnario a: Solar + BioMaSS
D1 Absolute Minimum (state-wide) 1,778 10 1,788 1,100 0.8 25
D2 Low income demand (state-wide) 5,936 75 6,011 6.7 19
D3 Medium income demand (state-wide) 14,326 153 14,479 13.4 25
D4 urban households (state-wide) 16,340 447 16,787 32.0 19
ScEnario B: Solar + SMall Hydro + BioMaSS
ScEnario c: Solar + Wind + BioMaSS
Source: GREENPEACE iNTERNATiONAL
1 MiLLENiuM DEVELOPMENT GOALS FOR DETAiLS SEE :HTTP://WWW.uN.ORG/MiLLENNiuMGOALS/
2 HTTP://WWW.iNDiATRiBuNE.COM/iNDEx.PHP?OPTiON=COM_CONTENT&ViEW=ARTiCL
E&iD=8574:ONLy-164-iN-BiHAR-HAVE-ACCESS-TO-ELECTRiCiTy&CATiD=125:GENERAL-
NEWS&iTEMiD=400
3 ENERGy [R]EVOLuTiON A SuSTAiNABLE ENERGy WORLD ENERGy OuTLOOK 2012,
GREENPEACE iNTERNATiONAL, AMSTERDAM THE NETHERLANDS, JuNE 2012
4 iNSTiTuTE FOR SuSTAiNABLE FuTuRES (iSF), uNiVERSiTy OF TECHNOLOGy, SyDNEy,
AuSTRALiA: JAy RuTOViTZ, ALiSON ATHERTON

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
Tseway Motup in Ladakh
TSEWAy MOTuP AND HiS FAMiLy WATCH TELEViSiON
iN THE EVENiNG iN uDMAROO ViLLAGE, NuBRA VALLEy,
LADAKH. ELECTRiCiTy iS SuPPLiED TO THE ViLLAGE
FOR Six HOuRS EVERy EVENiNG By A 30 KVA MiCRO-
HyDRO POWER uNiT iNSTALLED ON A MOuNTAiN
STREAM ABOVE THE HiMALAyAN ViLLAGE. THE FAMiLy
ALSO HAVE A MuSiC SySTEM, AN ELECTRiC BuTTER
CHuRNER AND CFL LiGHTS iN THEiR HOME.
executive
summary
electricity management committee
(L-R) LOBZANJ TSEPHEL (56), TSERiNJ RiNGCHEN
(56) AND TASHi NAMGiAL (59), MEMBERS OF THE
ELECTRiCiTy MANAGEMENT COMMiTTEE, MEET
TO DiSCuSS THE MANAGEMENT OF THE MiCRO-
HyDRO POWER uNiT iN uDMAROO ViLLAGE, LADAKH.
EVERy CuSTOMER iN uDMAROO iS A MEMBER
OF THE COMMiTTEE, THOuGH THE SOCiAL AND
TECHNiCAL GOVERNANCE OF THE SySTEM iS THE
RESPONSiBiLiTy OF AN ELECTED BODy OF Six
ViLLAGERS.07/30/2010

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
11
e
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
Microgrids can offer reliable and cost competitive electricity
services, providing a viable alternative to the conventional top-
down approach of extending grid services. The microgrid approach
is smart because it can facilitate the integration of renewable
energies, thereby contributing to national renewable energy (RE)
targets. in addition it can reduce transmission losses by having
generation close to demand. Being built from modular distributed
generation units, it can adequately adjust to demand growth. it
can operate both in island mode and grid-connected mode, making
operation fexible and can also offer grid support features.
This report demonstrates with a case study how this bottom-up
approach with microgrids would work. it focuses on development in
the state of Bihar in india.
step 1: renewable resource assessment
The frst step to this approach is to make an assessment of the
resources available in the area. in the case of Bihar, these are
biomass, hydro and solar PV power. While there are no detailed wind
measurements available, there are indications that in some areas
wind turbines could operate economically as well.
step 2: demand projections
The second step is to assess the level of electrical demand that
will need to be serviced. Once there is access to electricity services,
demand will almost always grow, accompanying economic growth.
For the case of Bihar the following demand levels were considered,
which are characterised by total energy consumption, peak demand
and daily load profles as shown in Figure 1 below.
As the proposed bottom-up electrifcation approach starts on a per-
village basis, a set of village demand profles is generated based on
these hypothetical household demand profles. The village demand
profles also contain assumptions about non-household loads such as
a school, health stations or public lighting.
The village-based electricity supply system forms the smallest
individual unit of a supply system. Therefore the matching set of
generation assets is also determined on a per-village basis.
step 3: define optimal generation mix
The third step in this approach is to design a system which can serve
the demand using the resources available in the most economic
manner. At this point it is of utmost importance that the system
design uses standard components and is kept modular so that it
can be replicated easily for expansion across the entire state. in
designing such a system, an appropriate generation mix needs to be
developed, which can meet demand 99% of the time at the lowest
production cost. This can be determined using production simulation
software such as HOMER
5
, which calculates the optimal generation
capacities based on a number of inputs about the installation and
operation costs of different types of generation technologies in india.
fgure 1: illustration how the demand (development of the household) will change over time
0
20
40
60
80
100
120
140
160
180 watts
Hour of day
0
100
200
300
400
500
600
700 watts
Hour of day
2 x CFL bulbs
1 x Ce II phone
65 kWh/year
Annual peak 30 W
TV /radio
Cooling/hea ting
500-1,000 kWh/year
Annual peak 150-500 W
Computers
Household appliances
1,200 kWh/year
Annual peak 1 kW
ABSOLUTE
MINIMUM
RURAL
HOUSEHOLD
URBAN
HOUSEHOLD
0
5
10
15
20
25
30
35
40
6 8 10 12 14 16
watts
Hour of day
2 4 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24
Source: GREENPEACE iNTERNATiONAL
12
e
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
fgure 2: process overview of supply system design by
production optimisation
PRODUCTION
OPTIMISATION
(HOMER)
OPTIMAL GENERATION
CAPACITIES THAT MEET
DEMAND 99% OF THE TIME
AT LEAST COST
AVAILABLE
RESOURCES
SYSTEM
DEMAND
GENERATION COSTS
& OPERATION
CHARACTERISTICS
Source: ENERGyNAuTiCS
step 4: network design
Once the optimal supply system design is determined, it is also
important to make sure that such a supply system can be distributed
through a physical network without breaching safe operating limits,
and that the quality of the delivered electricity is adequate for its
use. This can be done by modelling the physical system using power
system simulation software such as PowerFactory
6
. in this way
the behaviour of the electrical system under different operating
conditions can be tested, for example in steady-state power fow
calculations. Figure 3 shows a diagram of the village power system
model used in this study.
fgure 3: diagram of the powerfactory grid model
Source: ENERGyNAuTiCS
13
e
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
step 5: control system considerations
The fnal part of the system design involves the development of a
suitable strategy for switching between grid-connected and island
modes. Depending on the quality of service required by the loads
in the microgrid, the regulations stipulated in the grid code for
operation practices, and number of grid support features desired,
several different designs could be developed. For microgrids as
part of rural electrifcation efforts in developing countries however,
design simplicity and cost effciency weighs more than the benefts of
having an expensive but sophisticated control system.
Through the use of microgrids, the gap between rural electrifcation
and universal electrifcation with grid expansion can be met, while
at the same time bringing many additional benefts both for the
consumers and grid operators. By developing a system which is
modular and constructed using standard components, it makes it
easier to replicate it across wide areas with varying geographic
characteristics. The method demonstrated in this report can be
used to develop roadmap visions and general strategy directions.
it must be noted however, that detailed resource assessments, cost
evaluations, demand profle forecasts and power system simulations
are always required to ensure that a specifc microgrid design is
viable in a specifc location.
results
This report shows how microgrids could be a useful concept
for accelerating current rural electrifcation efforts, supporting
renewable energy integration into power systems, and creating
a reliable power system. The key to the wide-spread adoption of
such a strategy is to come up with system designs which are cost
competitive, and can be constructed using standard components in
a modular fashion. The process to develop such system designs is
demonstrated with a case study in Bihar, india.
in the frst instance, the microgrid can be designed as an off-
grid system with its own voltage and frequency control. it is
recommended that conventional voltage and frequency droop control
be used, rather than building a costly control scheme at this stage
of development. The frst priority for an un-electrifed village in
accessing electricity is to have access to lighting, which can extend
productive work hours. Therefore at absolute minimum, a microgrid
must be able to supply a demand based on the use of basic lighting,
being 2 compact fuorescent lamps (CFL) per household.
This can be achieved most effciently with a predominantly biomass-
powered system, such as the Husk Power Systems
7
, which are
already in application in a number of villages in Bihar today. While
the husk power system operates by charging around 18 iNR/kWh,
the system designs based on HOMER calculations indicates a system
cost of approximately 24 iNR/kWh. The discrepancy is attributed to
the fact that the optimal system calculated by HOMER is designed
to assure supply reliability 99% of the time.
Once an electricity service is available, people generally increase
their consumption. in india, this means adding appliances such
as fans, television sets and cellular phones. For a basic service
which can supply these types of appliances, a system that provides
approximately 500 kWh per year per household is required. This
can be achieved with a predominantly PV-biomass system or a PV-
biomass-hydro system (if a suitable hydro resource is available in
the vicinity of the village). Such systems can be achieved at costs of
around 19-24 iNR/kWh, or 11-13 iNR/kWh respectively.
in the E[R] Village Cluster concept, the demand grows step by
step as the economy grows and ultimately reaches the level of an
indian household in a city. Such a system would need to provide for
a consumption of around 1,200 kWh per year per household. These
systems can be achieved also at costs of 19-24 iNR/kWh, or 13-17
iNR/kWh respectively.
The more important point in achieving this system level is the way
in which the system is expanded. For example, biomass systems
based on rice husk (husk power systems) are available in unit sizes
of 32 kW and 52 kW, while hydro power is not proftable for units
sizes below 100 kW (based on the general fow conditions assumed
for the state of Bihar). On the other hand, solar PV systems are
assumed to be roof-top systems, so as not to compete for land
space with agricultural production. Therefore the unit sizes are
much smaller, in the range of 100-1,000 W. Therefore it would be
up to the system owner to decide how best to expand the system in
a piecewise fashion. However the energy clusters in the Greenpeace
concept are operated entirely with sustainable biomass and other
renewable energy sources. Three different energy mix options have
been calculated, based on different resource availabilities in the
State of Bihar.
table 1: village cluster demand overview
dEMand ScEnarioS SuPPly nEEdS
SCENARiO DEMAND PER DAy TOTAL ANNuAL DEMAND PEAK DEMAND TOTAL iNSTALLEDCAPACiTy
[kWh/day] [kWh/a] [kWpeak] [kW]
D1 Absolute Minimum village 111 40,514 22 31.5
D2 Low income village 881 321,563 99.4 106
D3 Medium income village 1,754 640,117 271 265
D4 urban households village 4,192 1,530,037 554 800
Source: ENERGy [R]EVOLuTiON CLuSTER - SMART ENERGy ACCESS, ENERGyNAuTiCS MAy 2012 AND GREENPEACE iNTERNATiONAL

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
Electric Cooker in Udmaroo Village
SONAM TSOMO MAKES DiNNER ON AN
ELECTRiC COOKER iN THE EVENiNG
iN uDMAROO ViLLAGE, NuBRA VALLEy,
LADAKH. ELECTRiCiTy iS SuPPLiED TO
THE ViLLAGE FOR Six HOuRS EVERy
EVENiNG By A MiCRO-HyDRO POWER uNiT
iNSTALLED ON A MOuNTAiN STREAM ABOVE
THE HiMALAyAN ViLLAGE.
14
E
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
Since the village of Manjharia seems to have no wind potential
which would allow to operate wind turbines economically
8
,
both scenario A and B shown in table 2 (see above) could be
implemented there. Scenario C however shows a generation mix for
villages with wind potential (6 m/s average wind speed).
When the system demand reaches a certain level, it becomes
fnancially interesting for the central grid to extend its services.
Therefore it is assumed that such microgrids will eventually be
connected to the central grid. Since microgrids retain the ability
to function as an autonomous system, when the grid fnally comes,
certain switching arrangements can be implemented to allow the
microgrid to operate both in grid-connected mode and island mode,
depending on the situation. For example, when there is a prolonged
blackout in the central grid, such as during rotating blackouts at
peak demand periods, the microgrid can disconnect from the grid
and operate in island mode, supplying loads locally. Also when the
microgrid is connected to the grid and receiving standard service
conditions, the generators connected to the microgrid can reduce
their outputs and let the cheap grid supply provide electricity to the
loads.
For a microgrid system to be able to manage the transitions to and
from grid-connected mode to island mode with minimal interruption
to grid operation, generators and loads, sophisticated control
schemes are required, which monitor, communicate and coordinate
control of all of these elements. Such a control system would be
expensive to develop, and some functions are still subjects of R&D
in laboratory conditions or in pilot project stages in regions like the
uSA, Europe and Japan.
However it would be diffcult to implement such a system in a
rural area in a developing country which has fnancial barriers as
well as less operational capacity, market fexibility, and regulatory
considerations. This is why a simplifed design concept which limits
the transition from grid-connected mode to island mode to periods
with prolonged blackouts, and following this, the transition from
island to grid-connected mode and hand over control back to the
central grid.
This strategy would involve manual switching following a system
blackout, to disconnect the microgrid from the central grid. Once
the microgrid is fully disconnected from the central grid it can be
restarted, and continue operation in island mode. Later, when the
central grid is back up and running, the microgrid would again
experience a brief blackout, while the connection to the grid is
switched back on, at which point the loads can be supplied again, by
the central grid supply
9
.
Required Investment for a E[R]village Cluster
The frst phase of this concept, corresponding to demand scenario 1,
would require an investment of 40million iNR for generation assets
and network equipment. With growing demand both the investment
in new generation and the income from selling the electricity would
grow. Compared to a supply scenario that includes Diesel, the 100%
renewable energy mix would be slightly more expensive, also due to
higher storage needs from photovoltaic power generation. unlike
the Diesel supply option, the money would remain in the community,
as the fuel is local, adding more economic benefts to the village.
in order to bridge the cost gap Greenpeace suggests supporting
E[R] village clusters with national and internal access to energy
programmes.
table 2: village cluster calculated energy mix
inStallEd caPacity ManjHaria villaGE GEnEration By tEcHnoloGy
SCENARiO PV WiND HyDRO BiO ENERGy PV WiND HyDRO BiO ENERGy
[kW] [kW] [kW] [kW] [kWh/a] [kWh/a] [kWh/a] [kWh/a]
MEDiuM PV
COST
MEDiuM PV
COST
MEDiuM PV
COST
MEDiuM PV
COST
D1 Absolute Minimum village 1.5 30 2,582 39,480
D2 Low income village 6.0 100 10,329 319,710
D3 Medium income village 25.0 240 43,037 627,320
D4 urban households village 600.0 200 1,032,885 925,648
D2 Low income village 6.0 100 40 10,329 588,285 18,355
D3 Medium income village 25.0 250 180 43,037 1,332,918 128,860
D4 urban households village 400.0 250 200 688,590 1,332,918 351,076
D2 Low income village 6.0 100 10,329 319,710
D3 Medium income village 25.0 240 43,037 627,320
D4 urban households village 400.0 100 300 688,590 240,159 867,518
15
E
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
Bottom up electrifcation, local planning with local technicians
(education & training required
10
) and top-down fnancing could
make access to energy a true success story in a win-win situation.
Up-scaling the E[R]cluster
To replicate this type of microgrid design across the entire state
of Bihar, a rough approximation based on geographical division
indicates that 13,960 villages can be supplied by a non-hydro no-
wind system and 3,140 villages with a hydro system. it is assumed
that there is potential for up to 1900 systems where wind power
may be used, and that a total number of 19,000 villages are
appropriate to cover all rural areas in the state of Bihar.
With such an expansion strategy, at minimum (corresponding to
demand scenario 2) approximately 1700MW of biomass, 314MW
of hydro and 114MW of PV power installations would be required.
At the stage when microgrids are fully integrated with the central
grid (demand scenario 4), it is expected that at least 4,000MW
of biomass, 785MW of hydro and 10,000MW of PV power
installations would be required.
Future Energy Mix in rural Bihar
Should the E[R] village clusters be implemented across all rural
areas in the state of Bihar, then all three generation options
(Scenario A, B and C) will be implemented. The generation
mix depends on the available resources and therefore will vary
signifcantly. However, we assumed that the majority of the villages
will have a mix of bio-energy, solar and small hydro - with a few
clusters including wind.
table 4: key results for e[r] village cluster - overall demand and supply - bihar state rural
dEMand StatE oF BiHar rural SuPPly StatE oF BiHar rural
SCENARiO DEMAND PER DAy TOTAL ANNuAL DEMAND PEAK DEMAND TOTAL iNSTALLED CAPACiTy
[GWh/day] [GWh/a] [MWpeak] [MW]
D1 Absolute Minimum (state-wide) 2 770 418 599
D2 Low income demand (state-wide) 17 6,110 1,889 2,140
D3 Medium income demand (state-wide) 33 12,162 5,149 5,641
D4 urban households (state-wide) 80 29,085 10,526 15,357
table 3: village cluster supply system investment overview
invEStMEnt For 3 diFFErEnt ScEnarioS
SCENARiO iNVESTMENT
POWER
GENERATiON
iNVESTMENT
NETWORK
EquiPMENT
TOTAL iNVESTMENT OPERATiON &
MAiNTAiNANCE
GENERATiON COSTS COST DELTA TO
DiESEL OPTiON
uS$ THOuSAND
iNR
uS$ THOuSAND
iNR
uS$ THOuSAND
iNR
uS$
/year
THOuSAND
iNR
uS$
/kWh/a
iNR
/kWh/a
uS$
/kWh/a
iNR
/kWh/a
D1 Absolute Minimum village 67,706 3,555 696,100 36,545 763,806 40,100 10,347 543 0.47 25 0.01 1
D2 Low income village 164,406 8,631 808,100 42,425 972,506 51,057 97,333 5,110 0.37 19 0.10 5
D3 Medium income village 439,337 23,065 808,100 42,425 1,247,437 65,490 245,636 12,896 0.47 25 0.17 9
D4 urban households village 2,205,136 115,770 1,242,200 65,216 3,447,336 180,985 278,073 14,599 0.37 19 0.09 5
D2 Low income village 385,745 20,252 850,200 44,636 1,235,945 64,887 20,060 1,053 0.22 11 0.01 1
D2 Low income village 164,406 8,631 808,100 42,425 972,506 51,057 97,333 5,110 0.37 19 0.10 5

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
ID: GP0260IHouse in Tangtse Village
CHHEMAT DORJAK (35, LEFT) AND STANZiN
DOLMA (28, RiGHT) WATCH TELEViSiON iN THEiR
HOME iN TANGTSE ViLLAGE, DuRBuK BLOCK,
LADAKH. THEiR ELECTRiCiTy iS SuPPLiED
By A SOLAR PHOTOVOLTAiC POWER PLANT iN
TANGTSE, 14,500M AMSL iN THE HiMALAyA.
THEy PAy A TWiCE-yEARLy FEE TO A LOCAL
COMMiTTEE, ELECTED TO MANAGE THE PLANT.
16
E
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
The overall installed capacity of solar photovoltaic systems would
be in the range of 12,000MW, while the rice husk and bio-energy
generators would add up to around 3,800MW. The overall regional
bio-energy potential of 16TWh/a would be suffcient to fuel the
feet of generators, signifcantly decreasing Bihars dependence on
imported fuels.
Investment requirements for the state for Bihar
The overall investment requirement for each E[R] villager cluster
would be in the range of 1 to 1.5 million uS Dollar (50 to
60millioniNR)
11
a investment volume which is quite typical for
small to medium size energy projects, but importantly without the
need to import fuel. Large parts of the money invested in E[R]
cluster would remain within the community and would contribute to
the local economy.
The case study in Bihar, india, clearly demonstrates how to realise
the active bottom-up approach of the Smart Energy Access strategy.
The transition from rural-electrifcation to universal electrifcation
is made possible by making use of the versatility of microgrids.
That is, its functionality as an off-grid system, the ability to
incorporate multiple generation sources, adapt to demand growth,
and to be integrated with the central grid, while retaining the ability
to separate and operate as an island grid if needed. By building
prototype microgrid designs, which can be created using standard
components, mixed and matched according to the type of renewable
resources available in the area, adoption and implementation can be
made effcient and effective, and suitable for wide-spread application
in developing countries.
footnotes
5 HOMER iS AN ENERGy MODELLiNG SOFTWARE FOR DESiGNiNG AND ANALySiNG HyBRiD
POWER SySTEMS. A TRiAL VERSiON OF THE SOFTWARE CAN BE DOWNLOADED FOR FREE AT
THE WEBSiTE: HTTP://WWW.HOMERENERGy.COM/
6 POWERFACTORy iS A POWER SySTEM SiMuLATiON SOFTWARE FOR DESiGNiNG AND
ANALySiNG POWER SySTEMS. iT iS A LiCENSED PRODuCT DEVELOPED By DiGSiLENT.
7 WWW.HuSKPOWERSySTEMS.COM
8 THE AVERAGE WiND SPEED iS ONLy ABOuT 4 M/S. FOR ECONOMiC OPERATiON OF WiND
TuRBiNES AT LEAST 6 M/S CAN BE CONSiDERED AS A MiNiMuM REquiREMENT.
9 THiS iS NEEDED WHERE GENERATOR MODES OF OPERATiON OR PROTECTiON SETTiNGS ARE
DiFFERENT BETWEEN iSLAND OPERATiON AND GRiD-PARALLEL OPERATiON AND CANNOT BE
SWiTCHED AuTOMATiCALLy.
10 COSTS FOR EDuCATiON PROGRAMS ARE NOT iNCLuDED iN THiS CALCuLATiON
11 MEDiuM PV COST SCENARiO, DEMAND SCENARiO 4, COST NOT iNCLuDiNG NETWORK
EquiPEMENT.
table 5: bihar state rural calculated energy mix
inStallEd caPacity BiHar StatE rural GEnEration By tEcHnoloGy
SCENARiO PV WiND HyDRO BiO ENERGy PV WiND HyDRO BiO ENERGy
[MW] [MW] [MW] [MW] [GWh/a] [GWh/a] [GWh/a] [GWh/a]
HiGH Pv coSt ScEnario
D1 Absolute Minimum (state-wide) 29 0 0 570 49 0 0 750
D2 Low income demand (state-wide) 114 0 314 1,712 196 0 1,847 5,128
D3 Medium income demand (state-wide) 475 10 785 4,372 818 27 4,185 10,349
D4 urban households (state-wide) 10,012 570 785 3,990 17,235 1,369 4,185 15,804
MEdiuM Pv coSt ScEnario
D3 Medium income demand (state-wide) 475 0 785 4,372 818 0 4,185 10,354
D4 urban households (state-wide) 10,392 190 785 3,990 17,890 456 4,185 15,673
loW Pv coSt ScEnario
D3 Medium income demand (state-wide) 2,458 0 785 4,372 4,231 0 4,185 8,724
D4 urban households (state-wide) 12,358 0 785 3,800 21,274 0 4,185 14,080
diesel truck on ladakh-manali route
A DiESEL TRuCK ON THE LADAKH - MANALi ROuTE. DiESEL iS
THE MAiN FuEL FOR TOWNS AND ViLLAGES iN THiS REMOTES
REGiON, AND MuST BE BROuGHT By ROAD: OFTEN A JOuRNEy
OF MuLTiPLE DAyS. DuE TO SNOWS AND LANDSLiDES,
DELAyS ARE COMMON.
17
E
x
e
c
u
t
i
v
e

s
u
m
m
a
r
y
1
table 6: bihar state rural supply system investment overview
invEStMEnt PoWEr GEnEration invEStMEnt nEtWork EquiPMEnt total invEStMEnt
SCENARiO MiLLiON uS$ BiLLiON iNR MiLLiON uS$ BiLLiON iNR MiLLiON uS$ BiLLiON iNR
HiGH Pv coSt ScEnario
D1 Absolute Minimum (state-wide) 1,334 70.1 13,226 694.4 14,560 764.4
D2 Low income demand (state-wide) 4,109 215.7 15,486 813.0 19,595 1,028.8
D3 Medium income demand (state-wide) 10,198 535.4 15,486 813.0 25,684 1,348.4
D4 urban households (state-wide) 60,257 3,163.5 23,602 1.239.1 83,859 4,402.6
MEdiuM Pv coSt ScEnario
D4 urban households (state-wide) 40,796 2,141.8 23,602 1,239.1 64,397 3,380.9
loW Pv coSt ScEnario
D4 urban households (state-wide) 26,860 1,410.2 23,602 1,239.1 50,462 2,649.3

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
Electric Saw in Udmaroo Village
CARPENTER TSEWANG STOBDAN OPERATES AN ELECTRiC
SAW uNiT iN uDMAROO ViLLAGE, NuBRA VALLEy,
LADAKH. HE iS A MEMBER OF A MENS SELF-HELP GROuP
THAT HAS STARTED uSiNG THE ELECTRiC MACHiNE
SiNCE A MiCRO-HyDRO POWER uNiT WAS iNSTALLED
ABOVE HiS ViLLAGE iN THE HiMALAyA. THEy CAN NOW
WORK FASTER AND TSEWANG ESTiMATES HAViNG THE
MACHiNE HAS DOuBLED THEiR iNCOME.
Woman with Solar Lantern in Bihar
A WOMAN BRiNGS A SOLAR LANTERN TO THE KOTAK
uRJA 3 KVA SOLAR STATiON. THE STATiON WAS SET
uP iN APRiL 2010 iN A HAMLET OF MuSAHARS - THE
POOREST OF THE POOR CASTE iN BiHAR - WHiCH
HAS NEVER HAD AN ELECTRiCiTy CONNECTiON.
THE SOLAR PLANT, iNSTALLED AT SiKANDARPuR,
ALSO PROViDES POTABLE WATER, COMMuNiTy TV,
A TELEPHONE BOOT AND CELLPHONE CHARGiNG
POiNTS. SiKANDARPuR ViLLAGE, DANAPuR, BiHAR.

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
1
access to energy -
the concept

1
2S

1 ACCLSS 1C LNLkG - 1nL CCNCL1
WlLh growlng demand, rlslng fossll-fuel prlces and lnLernaLlonal pressure Lo reduce
greenhouse gas emlsslons, developlng naLlons are faclng Lwo presslng lssues LhaL need
Lo be addressed ln parallel: lncreaslng energy access Lo pull people ouL of poverLy, whlle
lnLegraLlng as much renewables as posslble.
kUkAL LLLC1kIIICA1ICN ICk UNIVLkSAL LLLC1kIIICA1ICN
8ural elecLrlflcaLlon programs are consldered an lmporLanL parL of Lhe sLraLegy Lo
achleve unlversal elecLrlflcaLlon ln developlng counLrles. 1hese programs provlde access
Lo elecLrlclLy servlces elLher by exLendlng Lhe cenLral grld, or bulldlng off-grld generaLlon
capaclLy. Slnce Lhe ulLlmaLe goal ls Lo provlde rellable elecLrlclLy servlces aL affordable
cosLs, Lhe cheaper of Lhe Lwo opLlons ls usually Lhe one LhaL ls lmplemenLed.
Gr|d extens|on - the pass|ve top-down approach -
lor a vlllage LhaL ls noL yeL elecLrlfled, connecLlng Lo Lhe grld ls seen as Lhe preferred
opLlon, because lL means LhaL lL wlll be lnLegraLed dlrecLly lnLo Lhe cenLral grld and can
access cheap power supply (due Lo economles of scale). Powever, ln many developlng
counLrles Lhere lsn'L enough cenLral grld supply Lo cover all demand, and small rural
connecLlons are ofLen cuL off ln favour of malnLalnlng supply Lo clLles and lndusLrles
(resulLlng ln frequenL blackouLs). AnoLher lssue ls LhaL Lhe cenLral grld operaLors ofLen
do noL know exacLly how much power ls consumed where ln Lhe sysLem, due Lo lack of
meLerlng equlpmenL and rampanL energy LhefL. 1hls makes lL dlfflculL Lo conLrol Lhe
power quallLy such as malnLalnlng volLage and frequency wlLhln sLandard operaLlng
llmlLs. lurLhermore ln some cases, even lf exLendlng Lhe grld ls Lhe cheaper opLlon, rural
communlLles may noL be able Lo afford Lhe lnLerconnecLlon cosLs.
Cff-gr|d power generat|on - the act|ve bottom-up approach -
under Lhese clrcumsLances, Lhe opLlon of bulldlng off-grld supply sysLems such as solar
home sysLems and mlnl-grlds may be beLLer Lhan exLendlng Lhe grld. 1hls ls because lL
can provlde rellable elecLrlclLy supply aL a relaLlvely low cosL. ower can be sourced
from local renewable resources, and due Lo shorL dlsLances beLween generaLors and
consumers, Lhere are very llLLle elecLrlcal sysLem losses. lurLhermore, such sysLems can
be lnsLalled and seLup for operaLlon ln a relaLlvely shorL lengLh of Llme. 1he negaLlve
aspecLs of Lhese sysLems are LhaL Lhey are Lyplcally deslgned for a parLlcular
consumpLlon proflle, Lherefore requlre addlLlonal equlpmenL Lo caLer for lncreases ln
demand. Also, once an off-grld supply sysLem ls provlded, Lhe vlllage ls ofLen neglecLed
ln furLher elecLrlflcaLlon conslderaLlons because from Lhe cenLral sysLem's polnL of vlew
lLs work ls done on elecLrlfylng LhaL vlllage.

19
1
a
c
c
e
s
s

t
o

e
n
e
r
g
y

-

t
h
e

c
o
n
c
e
p
t
1

1
26

What |s the prob|em w|th the current process?
AlLhough ln prlnclple rural elecLrlflcaLlon programs lead Lo unlversal elecLrlflcaLlon, ln
reallLy, Lhere are obsLacles LhaL need Lo be overcome unLll Lhe deslred level of elecLrlclLy
servlce can be avallable Lo all. 1hese obsLacles are creaLed by Lhe llmlLs lmposed by Lhe
lmperfecL sysLem, such as lnadequacles of Lhe cenLral grld lnfrasLrucLure and operaLlon
capablllLy, hlgh lnlLlal cosLs of pro[ecLs, and lack of foreslghL ln sysLem deslgn LhaL ls
adopLed. 1hese acL Lo creaLe a gap beLween Lhe Lop-down approach of exLendlng Lhe
grld, and Lhe boLLom-up approach of bulldlng off-grld sysLems (llgure 4).

I|gure 4: 1op-down versus bottom-up approach to e|ectr|f|cat|on. Source: energynaut|cs
lor an un-elecLrlfled rural vlllage, galnlng access Lo elecLrlclLy servlces provlded by Lhe
cenLral grld can be Lhe cheapesL soluLlon. Powever, ofLen Lhe supply rellablllLy and
quallLy ls compromlsed because of rampanL supply shorLages and elecLrlclLy LhefL. Cn
Lhe oLher hand, home energy sysLems and off-grld dlsLrlbuLlon sysLems can provlde
more rellable servlces. 1hese opLlons are parLlcularly sulLable for vlllages LhaL are
remoLe from Lhe grld, or even lf lL ls close Lo Lhe grld cannoL afford Lhe connecLlon cosLs,
or where Lhe grld supply ls llmlLed. Cnce an off-grld soluLlon ls lmplemenLed however,
ofLen lL ls dropped from Lhe prlorlLy llsL Lo exLend grld servlces, Lherefore creaLlng a gap
LhaL prevenLs rellable unlversal elecLrlflcaLlon wlLh grld expanslon Lo be achleved.
INTEGRATION OF RENEWABLE ENERGIES
"Lnergy access |s a prerequ|s|te for deve|opment"
1hen, access Lo renewable energy ls a prerequlslLe for susLalnable developmenL. lor Lhe
maxlmum number of people Lo access renewable energy ln Lhe mosL economlc manner,
economles of scale would dlcLaLe LhaL Lhe LradlLlonal Lop-down approach of large-scale
power generaLlon feedlng lnLo a cenLral grld ls Lhe mosL approprlaLe way. Powever, for
developlng counLrles where Lhe cenLral grld reaches only a fracLlon of Lhe populaLlon,
and where lL ls dlfflculL Lo flnance large caplLal cosLs, a boLLom-up approach maybe
beLLer, slnce lL can provlde energy dlrecLly Lo unsupplled areas wlLhouL havlng such a
hlgh lnlLlal cosL.
1op-down: gr|d
extens|on
Consumers can
access besL Larls buL
servlce unrellable
8ouom-up: o-gr|d
systems
Plgher unlL cosL of
energy buL more
rellable
r]cluster for a smart energy access
tHe role of mIcrogrIDs In PromotIng tHe IntegratIon of reneWaBle energy In InDIa
20
a
c
c
e
s
s

t
o

e
n
e
r
g
y

-

t
h
e

c
o
n
c
e
p
t
1

1
27

1here are a number of boLLom-up approaches Lo renewable energy developmenL. 1hese
are: home energy sysLems, off-grld dlsLrlbuLlon sysLems, mlcrogrlds, and dlsLrlbuLed
generaLlon sysLems (llgure 3). 1he cholce of whlch energy supply opLlon Lo use depends
on Lhe resources LhaL are avallable, Lhe flnanclal capaclLy of Lhe pro[ecL execuLor,
Lechnlcal feaslblllLy and flexlblllLy, envlronmenLal lmpacLs and soclal adapLablllLy. lor
more lnformaLlon on Lhese sysLems refer Lo chapLer 3.

I|gure S: 1op-down versus bottom-up renewab|e energy deve|opment approaches. Source: energynaut|cs.
SMAk1 LNLkG ACCLSS
1o geL from rural elecLrlflcaLlon Lo a fully lnLegraLed grld whlle promoLlng renewable
energles, we suggesL mlcrogrlds for Lhe followlng advanLages/feaLures.
Microgrids
1he concepL of mlcrogrlds, ln parLlcular, wlLh tbe oblllty to swltcb betweeo lslooJ
opetotloo moJe ooJ qtlJ-coooecteJ moJe, presenLs an lnLeresLlng case. Mlcrogrlds are
dlsLrlbuLlon grlds whlch can be developed uslng a boLLom-up approach, [usL llke home
energy sysLems and off-grld sysLems, uslng local renewable energy sources Lo supply
local demand. Powever, unllke Lhese lsland sysLems, power supply ls more flexlble and
rellable because Lhere ls Lhe opLlon Lo source lL from local or cenLrallzed generaLlon,
accordlng Lo lLs avallablllLy and prlce. ln Lhls way, cheap power can be accessed from Lhe
grld durlng low demand perlods, whlle local supply can be acLlvaLed Lo power local
demand durlng blackouLs ln Lhe maln grld. lurLhermore, mlcrogrlds can conLrol Lhe
balance beLween local supply and demand, Lherefore manlpulaLe Lhe power flow aL Lhe
1op-down approach 8ouom-up approach
Pome energy
sysLem
C-grld dlsLrlbuuon
sysLem
Mlcrogrld
ulsLrlbuLed
generauon
Large-scale 8L
connecLed Lo
cenLral grld
21
1
a
c
c
e
s
s

t
o

e
n
e
r
g
y

-

t
h
e

c
o
n
c
e
p
t

1
28

polnL of common coupllng wlLh Lhe grld. 1hls enables Lhe mlcrogrld Lo supporL Lhe larger
grld ln managlng power quallLy as well as supply securlLy, by ln[ecLlng or consumlng
acLlve or reacLlve power when requlred, and aldlng ln black-sLarL slLuaLlons.
1he Concept
Smart Lnergy Access |s a strategy for un|versa| e|ectr|f|cat|on. lL lndlcaLes a paLh from
baslc rural elecLrlflcaLlon uslng locally avallable renewable resources Lo a sysLem LhaL ls
fully lnLegraLed wlLh Lhe cenLral grld. 1he meLhod suggesLed as Lhe vehlcle for Lhls
LranslLlon, ls whaL ls known as Lhe Mlcrogrld". 1he unlque feaLure of a mlcrogrld ls LhaL
lL has Lhe ablllLy Lo operaLe as an lsland sysLem, wlLhouL belng connecLed Lo Lhe cenLral
grld, buL also as an lnLegraLed sysLem, connecLed Lo Lhe cenLral grld, and operaLed as a
sLandard parL of Lhe exlsLlng grld sysLem. 1hls feaLure allows mlcrogrlds Lo be developed
as parL of a boLLom up" approach Lo grld expanslon. ln rural areas where grld-
connecLed elecLrlclLy servlces do noL exlsL, power can be generaLed locally by off-grld
supply sysLems (ln Lhe form of energy home sysLems or mlnl-grlds) relaLlvely qulckly. lf
Lhese sysLems are deslgned ln such a way LhaL Lhey can be Lurned lnLo mlcrogrlds, Lhey
can be easlly lnLegraLed lnLo Lhe cenLral grld laLer when Lhe grld ls flnally exLended
(llgure 6).

I|gure 6: Deve|opment stages of the 5mort nerqy 4ccess strategy. Source: energynaut|cs.

lurLhermore, slnce Lhe loads are supplled by local power generaLlon, Lhere wlll be less
sLress on Lhe cenLral grld Lo allocaLe supply from lLs already consLralned supply sysLem.
lf deslred Lhere ls also poLenLlal Lo source exLra supply from Lhe mlcrogrld Lo cover Lhe
shorLages ln Lhe cenLral grld, and Lhe conLrol sysLems of mlcrogrlds allow easy volLage
conLrol aL Lhe polnL of connecLlon (Lhls alds wlLh keeplng Lhe cenLral grld wlLhln healLhy
operaLlng condlLlons). lf Lhe conLrol sysLem ls sophlsLlcaLed enough, Lhere ls also
poLenLlal for Lhe mlcrogrlds Lo conLrlbuLe Lo frequency conLrol and black sLarLs, keeplng
Lhe power supply quallLy ln check and aldlng wlLh Lhe recovery of Lhe cenLral grld from
blackouLs.
If |t |s such a great too| then, why aren't m|crogr|ds used a|ready?
Crld expanslon for
unlversal
elecLrlflcaLlon uslng
boLLom-up"
approach wlLh
Mlcrogrlds
8eneflLs of
Mlcrogrlds for
cenLral grld:
uemand ls supplled
locally so Lhe sLress on
cenLral supply sysLem
ls reduced
volLage can be
conLrolled aL Lhe polnL
of common coupllng
(CC) Lo supporL Lhe
cenLral grld
uependlng on deslgn
of Lhe mlcrogrld, lL can
conLrlbuLe Lo
frequency conLrol and
black sLarL of Lhe
cenLral grld
Gr|d
extens|on
Gr|d
M|crogr|d
Is|and
operat|on
Gr|d-
connected
operat|on
No
L|ectr|c|ty
M|n|-gr|d
Lnergy home
system
22
a
c
c
e
s
s

t
o

e
n
e
r
g
y

-

t
h
e

c
o
n
c
e
p
t
1

1
28

polnL of common coupllng wlLh Lhe grld. 1hls enables Lhe mlcrogrld Lo supporL Lhe larger
grld ln managlng power quallLy as well as supply securlLy, by ln[ecLlng or consumlng
acLlve or reacLlve power when requlred, and aldlng ln black-sLarL slLuaLlons.
1he Concept
Smart Lnergy Access |s a strategy for un|versa| e|ectr|f|cat|on. lL lndlcaLes a paLh from
baslc rural elecLrlflcaLlon uslng locally avallable renewable resources Lo a sysLem LhaL ls
fully lnLegraLed wlLh Lhe cenLral grld. 1he meLhod suggesLed as Lhe vehlcle for Lhls
LranslLlon, ls whaL ls known as Lhe Mlcrogrld". 1he unlque feaLure of a mlcrogrld ls LhaL
lL has Lhe ablllLy Lo operaLe as an lsland sysLem, wlLhouL belng connecLed Lo Lhe cenLral
grld, buL also as an lnLegraLed sysLem, connecLed Lo Lhe cenLral grld, and operaLed as a
sLandard parL of Lhe exlsLlng grld sysLem. 1hls feaLure allows mlcrogrlds Lo be developed
as parL of a boLLom up" approach Lo grld expanslon. ln rural areas where grld-
connecLed elecLrlclLy servlces do noL exlsL, power can be generaLed locally by off-grld
supply sysLems (ln Lhe form of energy home sysLems or mlnl-grlds) relaLlvely qulckly. lf
Lhese sysLems are deslgned ln such a way LhaL Lhey can be Lurned lnLo mlcrogrlds, Lhey
can be easlly lnLegraLed lnLo Lhe cenLral grld laLer when Lhe grld ls flnally exLended
(llgure 6).

I|gure 6: Deve|opment stages of the 5mort nerqy 4ccess strategy. Source: energynaut|cs.

lurLhermore, slnce Lhe loads are supplled by local power generaLlon, Lhere wlll be less
sLress on Lhe cenLral grld Lo allocaLe supply from lLs already consLralned supply sysLem.
lf deslred Lhere ls also poLenLlal Lo source exLra supply from Lhe mlcrogrld Lo cover Lhe
shorLages ln Lhe cenLral grld, and Lhe conLrol sysLems of mlcrogrlds allow easy volLage
conLrol aL Lhe polnL of connecLlon (Lhls alds wlLh keeplng Lhe cenLral grld wlLhln healLhy
operaLlng condlLlons). lf Lhe conLrol sysLem ls sophlsLlcaLed enough, Lhere ls also
poLenLlal for Lhe mlcrogrlds Lo conLrlbuLe Lo frequency conLrol and black sLarLs, keeplng
Lhe power supply quallLy ln check and aldlng wlLh Lhe recovery of Lhe cenLral grld from
blackouLs.
If |t |s such a great too| then, why aren't m|crogr|ds used a|ready?
Crld expanslon for
unlversal
elecLrlflcaLlon uslng
boLLom-up"
approach wlLh
Mlcrogrlds
8eneflLs of
Mlcrogrlds for
cenLral grld:
uemand ls supplled
locally so Lhe sLress on
cenLral supply sysLem
ls reduced
volLage can be
conLrolled aL Lhe polnL
of common coupllng
(CC) Lo supporL Lhe
cenLral grld
uependlng on deslgn
of Lhe mlcrogrld, lL can
conLrlbuLe Lo
frequency conLrol and
black sLarL of Lhe
cenLral grld
Gr|d
extens|on
Gr|d
M|crogr|d
Is|and
operat|on
Gr|d-
connected
operat|on
No
L|ectr|c|ty
M|n|-gr|d
Lnergy home
system

1

1here are a few reasons for Lhls, buL Lhe maln lssues are deslgn slmpllclLy, sysLem cosL,
and Lhe lack of awareness of lLs beneflLs. lor appllcaLlon ln rural elecLrlflcaLlon, Lhe
sysLem deslgn ls somewhaL more compllcaLed Lhan a slmple off-grld dlsLrlbuLlon sysLem
because of Lhe addlLlonal capablllLy of belng able Lo swlLch beLween grld-connecLed and
lsland operaLlon modes. 1he assoclaLed sysLem cosLs are also hlgher because of Lhe
addlLlonal conLrol sLrucLure LhaL ls requlred. Powever, lf Lhese sysLems can be deslgned
uslng sLandard componenLs ln such a way LhaL Lhey are repllcable, Lhe cosLs wlll
evenLually come down as more mlcrogrld sLrucLures are lmplemenLed, due Lo
economles of scale. lurLhermore, Lhe lnlLlal lnvesLmenL need noL be for Lhe flnal full
mlcrogrld soluLlon, because mlcrogrlds can be bullL from sLandalone energy home
sysLems and off-grld dlsLrlbuLlon sysLems. 1herefore lL glves flexlblllLy ln Lhe Llmlng of
flnanclal lnvesLmenL, sLarLlng wlLh a slmpler sysLem and addlng feaLures as Llme goes by
unLll lL ls fully lnLegraLed wlLh Lhe cenLral grld sysLem.
1haL ls why ln Lhls reporL we Lry Lo glve suggesLlons on how Lo slmpllfy Lhe mlcrogrld
deslgn Lo be consLrucLed ouL of sLandard componenLs as repllcable modules whlch can
be lmplemenLed across Lhe globe for a Lruly susLalnable unlversal elecLrlflcaLlon.

23
a
c
c
e
s
s

t
o

e
n
e
r
g
y

-

t
h
e

c
o
n
c
e
p
t
1

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
2
microgrid
design case study
for bihar
Solar System on Hospital in Bihar
AN ELECTRiCiAN AT TRiPOLiA HOSPiTAL iN PATNA,
OPERATES THE CONCENTRATED SOLAR POWER (CSP)
SySTEM ON THE HOSPiTAL ROOF. THE SySTEM HAS
FOuR LARGE MiRROR DiSHES, WHiCH REFLECT
THE SuNS RAyS ONTO A HEADER PiPE CONTAiNiNG
WATER. AS THE RAyS ARE DiRECTED ONTO A SiNGLE
POiNT, THE HEAT TRANSFERRED iS VERy GREAT, AND
CREATES STEAM FROM THE WATER. TRiPOLiAS CSP
SySTEMS NOW CREATE STEAM TO STERiLiSE ALL OF
THE HOSPiTALS MEDiCAL iNSTRuMENTS, DRESSiNGS,
BEDSHEETS AND LAuNDRy, uSiNG THE FREE AND
RENEWABLE ENERGy OF THE SuN.

31

2 MICkCGkID DLSIGN CASL S1UD
1he case for mlcrogrlds ln promoLlng smarL" lnLegraLlon of renewable energy and
expandlng energy access ln developlng counLrles ls demonsLraLed wlLh a case sLudy ln
Lhe sLaLe of 8lhar, lndla.
1o glve an example of Lhe appllcaLlon, mlcrogrld deslgns sulLable for Lhe sLaLe of 8lhar ln
lndla are presenLed. Slmple sysLem deslgns are suggesLed LhaL can be adopLed ln mosL
vlllages across Lhe sLaLe, Lo achleve Lhe goal of unlversal elecLrlflcaLlon Lhrough SmarL
Lnergy Access.
METHODOLOGY
1he flrsL sLep Lo Lhls approach ls Lo make an assessmenL of Lhe resources avallable ln Lhe
area.
1he second sLep ls Lo assess Lhe level of elecLrlcal demand LhaL wlll need Lo be servlced.
1he Lhlrd and flnal sLep ln Lhls approach ls Lo deslgn a sysLem whlch can serve Lhe
demand uslng Lhe resources avallable ln Lhe mosL economlc manner. AL Lhls polnL lL ls of
uLmosL lmporLance LhaL Lhe sysLem deslgn uses sLandard componenLs and ls kepL
modular so LhaL lL can be repllcaLed easlly for expanslon across Lhe enLlre sLaLe.
1he slze of Lhe generaLlon sysLem ls deLermlned based on producLlon opLlmlsaLlon uslng
sofLware called PCML8
12
. 1hls program evaluaLes dlfferenL comblnaLlons of generaLlon
Lechnologles and deLermlnes Lhe mosL cosL efflclenL generaLlon mlx LhaL meeLs Lhe
demand.
1o ensure LhaL such a supply sysLem could acLually be dellvered Lhrough a physlcal
neLwork, power flow slmulaLlons are performed uslng sofLware called owerlacLory
13
. ln
Lhls way dlfferenL neLwork Lopologles can be evaluaLed, and equlpmenL needed Lo
malnLaln sysLem frequency and volLages can be deLermlned.
1he flnal parL of Lhe sysLem deslgn lnvolves Lhe developmenL of a sulLable sLraLegy for
swlLchlng beLween grld-connecLed and lsland modes. uependlng on Lhe quallLy of
servlce requlred by Lhe loads ln Lhe mlcrogrld, Lhe regulaLlons sLlpulaLed ln Lhe grld code
..............................
12
PCML8 ls an energy modelllng sofLware for deslgnlng and analyslng hybrld power sysLems. A
Lrlal verslon of Lhe sofLware can be downloaded for free aL Lhe webslLe:
hLLp://www.homerenergy.com/
13
owerlacLory ls a power sysLem slmulaLlon sofLware for deslgnlng and analyslng power
sysLems. lL ls a llcensed producL developed by ulgSlLLn1
25
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

32

for operaLlon pracLlces, and number of grld supporL feaLures deslred, several dlfferenL
deslgns could be developed. lor mlcrogrlds as parL of rural elecLrlflcaLlon efforLs ln
developlng counLrles however, where deslgn slmpllclLy and cosL efflclency welghs more
Lhan Lhe beneflLs of havlng a expenslve buL sophlsLlcaLed conLrol sysLem, a slmple
sLraLegy ls proposed.
1he process used Lo deslgn Lhe mlcrogrld for Lhls case sLudy ls shown ln llgure 7.

I|gure 7: rocess for des|gn|ng the m|crogr|d |n the 8|har case study. Source: energynaut|cs
OVERVIEW OF BIHAR
8lhar ls a sLaLe ln Lhe norLh easL of lndla. lL conslsLs malnly of flaL plalns and ls lnLerlaced
wlLh several rlvers (llgure 8). lL has a populaLlon of over 100 mllllon
14
, of whlch almosL
83 llve ln rural areas. AgrlculLure ls Lhe maln form of economy, ownlng Lo lLs ferLlle
ground and plenLlful waLer source. 1emperaLures can vary from 3 C ln wlnLer
(uecember Lo lebruary) Lo 43 C ln summer (March Lo May)
13
, and Lhere ls a monsoon
season exLendlng from !une Lo SepLember. 1he perlod from CcLober Lo november ls
known as Lhe posL monsoon season.
..............................
14
hLLp://gov.blh.nlc.ln/roflle/defaulL.hLml
13
hLLp://www.blharonllne.ln/AbouL/roflle/Ceography/cllmaLe.hLml
SysLem deslgn
Supply sysLem based on
producuon opumsauon
uslng PCML8 soware
ulsLrlbuuon sysLem based
on slmulauon of physlcal
neLwork behavlour uslng
owerlacLory soware
ConLrol sLraLegy for
swuchlng beLween grld-
connecLed and lsland
operauon modes
uemand assessmenL
Assess Lhe level of elecLrlcal demand LhaL wlll need Lo be servlced
8esource assessmenL
Assess Lhe Lypes of generauon resources avallable ln Lhe area: e.g. blomass, solar, wlnd,
hydro, eLc
26
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

33

I|gure 8: hys|ca| map of 8|har. Source: Mod|f|ed from http:]]www.openstreetmap.org
1he elecLrlflcaLlon raLe ln 8lhar ls one of Lhe lowesL ln Lhe counLry aL 61
16
. 1here ls a
rural elecLrlflcaLlon program
17
whlch alms Lo connecL vlllages Lo Lhe grld, however, even
lf Lhere ls a connecLlon Lo Lhe grld, due Lo Lhe severe supply shorLage
18
, Lhe quallLy and
rellablllLy of servlce are ofLen compromlsed, wlLh some vlllages reporLlng only 2 days of
servlce per monLh
19
.
Bihar Microgrid Village
A Lyplcal vlllage ln 8lhar ls mosLly rural, wlLh a school, a healLh cenLre, and
admlnlsLraLlve offlce, buL normally Lhere ls very llmlLed lndusLry and commerclal
bulldlngs
20
. 1he number of people per household ls sald Lo be around Len. 1hls ls hlgher
Lhan Lhe naLlonal average of flve people per household, buL noL an aLyplcal slLuaLlon, as
lL ls common for several generaLlons Lo llve LogeLher under Lhe same roof ln rural parLs
of lndla.
..............................
16
hLLp://www.lea.org/counLry/overLy_lndla/LlecLrlflcaLlon.pdf
17
known as Lhe 8a[lv Candhl Crameen vldyuLlkaran ?o[ana (8CCv?), Lhe program has succeeded
ln elecLrlfylng 20,326 vlllages ouL of Lhe LargeL of 29,862 vlllages slnce 2003.
hLLp://gov.blh.nlc.ln/documenLs/Lconomlc-Survey-2011-Lngllsh.pdf aLe 161
18
Acoordlng Lo hLLp://gov.blh.nlc.ln/documenLs/Lconomlc-Survey-2011-Lngllsh.pdf aLe 164,
Lhere ls a peak deflclL of 31 and an energy deflclL of 46
19
hLLp://www.greenpeace.org/lndla/Clobal/lndla/reporL/Lmpowerlng-8lhar.pdf, age 11
20
hLLp://www.lcssr.org/aschlm20Champaran20flnal3813u.pdf age 2
27
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

34

lor Lhe case sLudy, a vlllage ln Lhe norLhwesL of Lhe sLaLe called Man[harla was
consldered. 1hls vlllage has a populaLlon of 6,000 llvlng ln 12 conglomeraLlons (called
clusLers ln Lhls reporL), and Lhere are approxlmaLely 600 households ln Lhe area
21
.

I|gure 9: Map of Man[har|a v|||age. Source: Greenpeace.
1he average annual lncome ln 8lhar (excludlng Lhe caplLal, aLna) ls approxlmaLely 7,873
ln8 (1,730 uSu) per caplLa [1]. 1herefore for a household of Len people Lhls amounLs Lo
[usL over 6,300 ln8 (143 uSu) per monLh. LlecLrlclLy dellvered from Lhe grld cosLs
around 3 ln8 per kllowaLL-hour [2], and Lhe average per caplLa consumpLlon ls 122 kWh
per year [3]. 1hls means LhaL approxlmaLely 300 ln8/monLh ls requlred for Lhe mosL
baslc elecLrlclLy servlce, lf Lhe grld can ln facL provlde lL. lL has been sLaLed LhaL, wlLhouL
access Lo Lhe grld, or wlLh access buL poor servlce quallLy, vlllagers are wllllng Lo pay up
Lo 13-20 ln8/kWh for elecLrlclLy servlces [2]. ln Lhe case of Man[harla, Lhe dlsLrlcL has no
..............................
21
lnformaLlon provlded by Manlsh 8am
28
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

3S

grld connecLlon. lnsLead lL ls elecLrlfled by Pusk ower SysLems
22
, a blogas power
generaLlon sysLem based on rlce husk. 1he servlce ls avallable 6 hours a day, aL a charge
of 100 ln8 per monLh per 30 W of supply. 1hls works ouL Lo be approxlmaLely 18 ln8 per
kllowaLL-hour, servlng as a cosL effecLlve alLernaLlve Lo grld-supplled elecLrlclLy.
1he closesL vlllage wlLh dlsLrlbuLlon lnfrasLrucLure ls narkahawa approxlmaLely 3 km
away. Powever, Lhls vlllage ls currenLly noL elecLrlfled. 1he closesL vlllage whlch ls
elecLrlfled ls Mangalpur across Lhe Candak rlver, wlLh a dlrecL llne dlsLance of
approxlmaLely 10 km (llgure 10).

I|gure 10: Map |ocat|on of Man[har|a. Mod|f|ed from http:]]www.geo|ys|s.com] (Map data from Goog|e
Maps)
STEP 1: ASSESSMENT OF RENEWABLE ENERGY RESOURCES IN BIHAR
1he economlcally explolLable renewable resources ln 8lhar are solar, wlnd, hydro and
blomass.
..............................
22
huskpowersystems
Man[harl
a
Candak
8lver
29
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

WlLh an average solar lrradlaLlon of 4.8 kWh/m
2
/day and average dally sunshlne of
around 8 hours, solar power ls a very aLLracLlve energy resource
23
. lf Lhe lnsLallaLlon
cosLs are compeLlLlve enough, solar phoLovolLalc Lechnology ls Llpped Lo domlnaLe Lhe
scene as lL can be lmplemenLed almosL anywhere ln Lhe sLaLe for elecLrlclLy producLlon.
lor Lhe appllcaLlon of mlcrogrlds ln rural vlllages, lL ls expecLed LhaL mosL of Lhe
developmenL wlll focus on roofLop solar v sysLems, so as noL Lo compeLe for land space
wlLh agrlculLural producLlon.
Wlnd power on Lhe oLher hand, ls noL a very aLLracLlve resource ln 8lhar. 1hls ls malnly
due Lo Lhe low average wlnd speeds across Lhe sLaLe
24
. 1he resource may be explolLable
wlLh low power wlnd generaLlon sysLems ln Lhe range of 100-1,000W Lurblnes, however
lL ls expecLed LhaL Lhe peneLraLlon raLe of such sysLems wlll noL be domlnanL due Lo Lhe
hlgher lnsLallaLlon cosLs compared Lo oLher Lechnologles.
WlLh so many rlvers and canals ln Lhe sLaLe, hydro power ls also an aLLracLlve resource.
Powever, due Lo Lhe flaL land Lopology, flow raLes and avallable head helghLs are noL
very hlgh
23
, llmlLlng Lhe appllcaLlons Lo low power hydro schemes. ln addlLlon, Lhe low-
lylng flaL plalns are prone Lo floodlng durlng monsoon season, and canals are used
exLenslvely for lrrlgaLlon. 1herefore Lhe avallable flow for operaLlon of a hydro power
planL ls hlghly seasonal, ln some cases operaLlon ls lnLerrupLed for 4 monLhs of Lhe
year
26
. Mlcro hydro Lechnology ls a well-esLabllshed and economlc sound Lechnology
however, Lherefore even wlLh Lhe llmlLaLlons on resource avallablllLy lL ls expecLed Lo
remaln an explolLable resource.
A varleLy of blomass resources are avallable ln 8lhar, such as domesLlc wasLe, naLlve
shrubs, and by-producLs from agrlculLural producLlon such as rlce and wheaL husks. 1he
laLLer opLlon ls Lhe mosL aLLracLlve as lL can co-exlsL wlLh producLlon of much needed
food producLs, and Lhe largesL wasLe producL from rural areas of 8lhar ls agrlculLural
wasLe, raLher Lhan domesLlc wasLe. lurLhermore, power generaLlon sysLems based on
rlce husk gaslflers are already appllcaLlon ln a number of vlllages
27
as economlcally vlable
alLernaLlves Lo elecLrlflcaLlon by grld exLenslon.
..............................
23
CllmaLologlcal Solar 8adlaLlon (CS8) daLa from n8LL
24
AL a hub helghL of around 30m, Lhe average wlnd speed ls approxlmaLely 4 m/s accordlng Lo
hLLp://energy.blh.nlc.ln/docs/8L-A-A-8lhar.pdf, ages 18-19
23
MosL hydropower planLs currenLly under operaLlon ln 8lhar run on a deslgn flow raLe of around
200 llLres per second and a head helghL of 4 meLres.
26
Accordlng Lo naLwar S.P. . SallenL leaLures provlded by 8lhar SLaLe PydroelecLrlc ower
CorporaLlon
27
See Pusk ower SysLems: huskpowersystems
30
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

37

1ab|e 7: L|ectr|ca| energy potent|a| from b|o-energy sources |n the state of 8|har. Source: Greenpeace
(mod|f|ed from WISL[33])
no Lnergy Source !/a
AgrlculLural 8esldues
1+2 8lce husk (30 + 100 avallablllLy) 14.3
3 8lce sLraw (30 avallablllLy) 24.0
4 Malze cobs (30 avallablllLy) 1.6
3 Sugarcane bagasse 9.3
Sub LoLal 49.3
urban WasLe
6 Munlclpal Solld WasLe (Lhermochemlcal converslon) 0.8
7 Munlclpal Solld WasLe (blochemlcal converslon) 0.3
8 Munlclpal Llquld WasLe 0.7
Sub LoLal 2.0
CLher lndusLrlal WasLes
9 ulsLlllery (spenLwash) 0.4
10 ualry (washlngs, whey) 0.0
11 Sugar (wasLe waLer, pressmud) 0.1
Sub LoLal 0.6
1oLal 31.9

STEP 2: ASSESSMENT OF SYSTEM DEMAND
1he second sLep ls Lo assess Lhe level of elecLrlcal demand LhaL wlll need Lo be servlced.
1he flrsL prlorlLy for an un-elecLrlfled vlllage ls elecLrlclLy servlce for llghLlng. 1herefore
Lhe absoluLe mlnlmum demand scenarlo assumes LhaL every household ls equlpped wlLh
Lwo ClL llghLs. 1he servlce ls assumed Lo have llmlLed avallablllLy, belng 3-6 hours of
elecLrlclLy supply servlce per day
Cnce Lhere ls access Lo elecLrlclLy, demand wlll mosL cerLalnly grow. 1he second demand
scenarlo also conslders elecLrlclLy servlces for space coollng and enLerLalnmenL, belng a
fan and a Lelevlslon, whlch ls consldered Lo be Lhe mlnlmum sLandard for a low lncome
household [4].
1he ensulng scenarlos allow for lncreased demand growLh unLll Lhe level of elecLrlclLy
consumpLlon ln a clLy ls reached. lor Lhe purpose of Lhls sLudy, lL was assumed LhaL all
households have more or less Lhe same consumpLlon paLLern, wlLh Lhe followlng
appllances consldered ln each scenarlo. 1he LoLal energy consumpLlon and peak demand
are also glven.

31
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

38

1ab|e 8: Summary of app||ances cons|dered |n the consumpt|on patterns of res|dent|a| demand for each
scenar|o. Source: energynaut|cs

Scenar|o 1
Abso|ute
m|n|mum
Scenar|o 2
Low |ncome
househo|d
Scenar|o 3
Med|um |ncome
rura| househo|d
Scenar|o 4
Urban househo|d
L|ght|ng
Ce|| phone
Ian]Coo|er
kad|o]1V
Water pump
Cther
eak demand per
househo|d
30 Wp 130 Wp 300 Wp 1 kWp
Annua| energy
demand per
househo|d
63 kWh 300 kWh 1,000 kWh 1,200 kWh

ln addlLlon Lo Lhe resldenLlal secLor, demand from Lhe followlng secLors was also
consldered for all scenarlos.
1ab|e 9: Summary of app||ances cons|dered |n the consumpt|on patterns of non-res|dent|a| demand for a||
scenar|os. Source: energynaut|cs
ub||c ||ght|ng Schoo| nea|th stat|on
Adm|n|strat|ve
bu||d|ng
L|ght|ng
Ian]Coo|er
Water pump
kefr|gerator
The resulting system demand for each demand scenario is summarised in 1able 10 below.
1ab|e 10: 1ota| annua| demand and system peak demand for each scenar|o. Source: energynaut|cs

Scenar|o 1
Abso|ute
m|n|mum
Scenar|o 2
Low |ncome
househo|d
Scenar|o 3
Med|um |ncome
rura| househo|d
Scenar|o 4
Urban househo|d
1ota| demand per
year
40 MWh 320 MWh 640 MWh 1,340 MWh
System peak
demand
20 kW 100 kW 270 kW 334 kW
32
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

39

STEP 3: SYSTEM SIZING BY PRODUCTION OPTIMISATION WITH HOMER
An opLlmal generaLlon mlx was Lhen calculaLed uslng producLlon slmulaLlon sofLware
called PCML8
28
. 1hls sofLware uses generaLlon caplLal and operaLlng cosLs as well as
operaLlon characLerlsLlcs as lnpuL Lo calculaLe dlfferenL comblnaLlons of generaLlon unlLs
and slzes. lL flnds Lhe leasL cosL comblnaLlon based on energy producLlon slmulaLlon Lo
meeL Lhe demand. 1he assumpLlons made regardlng Lhe generaLlon cosLs and operaLlon
are summarlsed ln 1able 11 below.
1ab|e 11: Generat|on cost assumpt|ons used |n the nCMLk s|mu|at|on for 8|har. Source: Var|ous.
Generat|on type
(un|t s|zes
cons|dered)
Cap|ta| cost (per kW) C&M cost Notes regard|ng operat|on
So|ar V
(0.1-1 kW)
1,260-4,204 uSu
(114,000-190,000 ln8)
[3]
3 of CALx/yr [6]
Average lrradlance: 4.87
kWh/m
2
/day
lnsLalled as solar home sysLems
rlce wlLhouL baLLery
8|omass
(10-S0 kW)
1,100-2,213 uSu
(30,000-100,000 ln8)
[7]
20 of CALx/yr [8]
Source: 8lce husk
luel cosL: 0 uSu/Lonne
29

ConsumpLlon raLe: 1.4-1.7 kg/kWh
nydro
(10-2S0 kW)
1 x 10 kW: 10,930 uSu
(493,000 ln8) [9]
2 x 20 kW: 3,340 uSu
(241,230 ln8) [10]
2 x 30 kW: 2670 uSu
(93,000 ln8) [11]
230 kW: 1,900 uSu
(83,600 ln8) [12]
3 of CALx/yr [9]
[10] [11] [12]
ueslgn flow raLe: 7,230 L/s avallable
beLween Aprll Lo november
W|nd
(6S0-1,S00 W)
4,230 uSu
(192,300 ln8) [13]
3 of CALx/yr [13]
Average wlnd speed: 6 m/s aL 20 m
hub helghL
D|ese|
30

333 uSu
(23,000 ln8) [14]
0.01 uSu/h/kW
(0.23 ln8) [13]
luel cosL: 0.97 uSu/L (30 ln8/L)
ConsumpLlon raLe: 0.3 L/kWh
8attery
(S0-200 Ah)

130-1,300 uSu
(6,800-38,730 ln8)
31

negllglble
8aLLerles are lnsLalled as parL of Lhe
solar home sysLems
Gr|d extens|on
8,000 uSu
(360,000 ln8) per km
[16]
400 uSu/yr
(18,000 ln8/yr) [16]
Lxcludes dlsLrlbuLlon grld
componenLs
..............................
28
PCML8 ls an energy modelllng sofLware for deslgnlng and analyslng hybrld power sysLems. A
Lrlal verslon of Lhe sofLware can be obLalned from Lhe webslLe: hLLp://www.homerenergy.com/
29
Crlglnally esLlmaLed Lo 33 uSu/Lonne. Powever, Lhe acLual cosL can be assumed zero because
addlLlonal revenue can be galned from selllng Lhe by-producL from gaslflcaLlon (called char),
whlch can offseL Lhe orlglnal fuel cosL.
30
ulesel ls used only for a reference slmulaLlon Lo be able Lo calculaLe a cosL dlfference, buL ls noL
lncluded ln Lhe proposed energy mlx.
31
uefaulL values are glven by PCML8. llgures were verlfled by comparlng wlLh prlce lnformaLlon
obLalned onllne.
33
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

Unit size considerations
AL Lhls polnL Lhe unlL slzes consldered need Lo Lake lnLo accounL LhaL Lhe sysLem deslgn
musL use sLandard componenLs and ls kepL modular so LhaL lL can be expanded Lo caLer
for demand growLh. lor example, blomass sysLems based on rlce husk (husk power
sysLems) are avallable ln unlL slzes of 32 kW and 32 kW [17], buL can be consldered Lo
come ln slzes varylng from 10-30 kW [7]. Small hydro power sysLems sLarLlng from 100
kW are under operaLlon, ln consLrucLlon or have been planned for uLlllslng varlous
exlsLlng canals ln 8lhar
32
, buL lL may be posslble Lo lmplemenL mlcro hydro power
schemes such as Lhose descrlbed ln Lhe Model ueLalled ro[ecL 8eporLs [9] [10] [11]
publlshed by Lhe MlnlsLry of new and 8enewable Lnergy
33
. Solar v sysLems are
assumed Lo be roof-Lop sysLems, so LhaL Lhey do noL Lo compeLe for land space wlLh
agrlculLural producLlon. 1herefore Lhe unlL slzes are much smaller, ln Lhe range of 100-
1,000 W. lurLhermore, home solar sysLems are usually equlpped wlLh baLLery sLorage ln
Lhe range of 30-200Ah because Lhe maln use for elecLrlclLy ls for llghLlng, whlch ls noL
requlred durlng Lhe day when Lhe sun ls shlnlng and solar energy ls avallable. ulesel
generaLors are avallable ln all slzes as Lhey are a more convenLlonal producL. 1herefore
lL would be up Lo Lhe sysLem owner Lo declde how besL Lo expand Lhe sysLem ln a
plecewlse fashlon.
1o demonsLraLe LhaL Lhese sysLems can be adopLed ln oLher reglons of Lhe world (noL
only llmlLed Lo 8lhar or lndla), wlnd power was also lncluded ln Lhe generaLlon mlx. Cnly
small wlnd sysLems are consldered however, wlLh unlL slzes ranglng from 630-1,300 W.
1hese Lypes of sysLems have a hub helghL of around 20 meLres, and do noL requlre
heavy clvll works Lo be lnsLalled.
1he unlL slzes LhaL should be consldered and Lhelr correspondlng prlces may dlffer
beLween Lhe Lype of resources LhaL are locally avallable and Lhe reglon of operaLlon.
1herefore, Lo use Lhls approach for developlng mlcrogrld deslgns for oLher parLs of Lhe
world, Lhe lnpuL values should be modlfled Lo reflecL Lhe local condlLlons.
Solar PV price sensitivity
1o demonsLraLe LhaL Lhere ls merlL ln acceleraLlng Lhe adopLlon of solar v sysLems and
reach lower unlL prlces Lhrough economles of scale (llgure 11), Lhe opLlmal sysLem
deslgn based on global solar v prlces were calculaLed ln addlLlon Lo Lhe currenLly
appllcable prlces ln lndla. 1hls case demonsLraLes LhaL wlLh lower unlL prlces achleved
from lncorporaLlon of hlgher volumes of solar v pro[ecLs, overall sysLem cosLs can be
..............................
32
lor deLalls, check Lhe 8lhar SLaLe PydroelecLrlc ower CorporaLlon webslLe:
hLLp://www.bshpclLd.com/
33
1he reporLs can be accessed from Lhe AlLernaLe Pydro Lnergy CenLre webpage:
hLLp://www.ahec.org.ln/publ/model/model20dprs.hLml
34
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

41

broughL down. 1herefore governmenLs should encourage pollcles LhaL acceleraLe Lhe
adopLlon of solar v power such as Lhose seen ln counLrles llke Cermany.

I|gure 11: V modu|e pr|ce exper|ence curve. Source: Greenpeace
34

AnoLher facL Lo conslder ls LhaL prlces for v modules have dropped conslderably ln
recenL years, and are llkely Lo drop furLher ln Lhe fuLure. 1o conslder Lhe wlde range of
prlces posslble ln a sLudy LhaL alms Lo pro[ecL lLs resulLs lnLo Lhe fuLure as far as aL leasL
one decade, Lhree prlces for solar modules ln uS-uollar per kllowaLL have been used:
- Plgh v cosL scenarlo: 190,000 ln8/kllowaLL (4200 uSu/kllowaLL) - Lhls reflecLs
prlces used for slmllar sLudles as recenL as one year ago.
- Medlum v cosL scenarlo: 114,000 ln8/kllowaLL (2320 uSu/kllowaLL) - afLer a
heavy prlce drop over Lhe lasL monLhs, Lhls represenLs a more reallsLlc prlce
assumpLlon as of Loday.
- Low v cosL scenarlo: 37,000 ln8/kllowaLL (1260 uSu/kllowaLL) - Lhls ls a
hypoLheLlcal prlce LhaL exLrapolaLes Lhe recenL prlce developmenL only a few
years lnLo Lhe fuLure.
HOMER Simulation Results
Bihar microgrid with and without hydropower
1he followlng Lables show Lhe opLlmal generaLlon capaclLles for a mlcrogrld ln 8lhar wlLh
(1able 13) and wlLhouL (1able 12) access Lo sulLable hydro power resources.
..............................
34

hLLp://www.greenpeace.org/lnLernaLlonal/Clobal/lnLernaLlonal/publlcaLlons/cllmaLe/2011/llnal
20SolarCeneraLlon20vl20full20reporL20lr.pdf, age 30
35
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

42

1ab|e 12: Cpt|ma| generat|on capac|t|es for 8|har m|crogr|d w|thout hydropower. Source: energynaut|cs.
Demand scenar|o (system peak demand)
So|ar V
(kW)
8|omass
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 4000 0.48
Scenar|o 2 (100 kW) 6 100 4000 0.38
Scenar|o 3 (270 kW) 23 240 16000 0.48
Scenar|o 4 (SS4 kW) 600 200 140000 0.30
1ab|e 13: Cpt|ma| generat|on capac|t|es for 8|har m|crogr|d w|th hydropower. Source: energynaut|cs.
Demand scenar|o (system peak demand) So|ar V
(kW)
8|omass
(kW)
nydro
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 0 4000 0.48
Scenar|o 2 (100 kW) 6 40 100 4000 0.22
Scenar|o 3 (270 kW) 23 180 230 10000 0.26
Scenar|o 4 (SS4 kW) 400 200 230 40000 0.34
1he consumpLlon ln scenarlos 1 Lo 3 ls llmlLed Lo cerLaln hours of Lhe day, even Lhough
mosL of Lhe generaLlon capaclLy ls avallable for producLlon Lhe whole day. 1he more
energy ls produced by Lhe unlL, Lhe lower Lhe cosL per kllowaLL-hour wlll become. 1hls
sysLem ls deslgned Lo provlde a LoLal capaclLy of only 30 waLLs per household, whlch ls
noL enough Lo power many appllances aparL from ClL llghLlng and cell phone charglng.
Slnce Lhe goal of smarL energy access ls Lo move beyond Lhls absoluLe baslc scenarlo, lL
would noL make sense Lo base Lhe mlcrogrld deslgn on Lhls sysLem deslgn.
lgnorlng demand scenarlo 1, whlch ls Lhe absoluLe mlnlmum and unreallsLlc for a
mlcrogrld, Lhe cosL of energy (CoL) resulLlng from Lhe opLlmal sysLem deslgn ranges
from 0.38 uSu Lo 0.30 uSu per kllowaLL-hour (19-23 ln8/kWh) for sysLems wlLhouL
hydro, and beLween 0.22 uSu and 0.34 uSu per kllowaLL-hour (11-17 ln8/kWh) for
sysLems wlLh hydropower. 1hese flgures are ln range wlLh Lhe servlce provlded by Pusk
ower SysLem (18 ln8/kWh, correspondlng Lo 0.36 uSu/kWh), buL ls much hlgher
compared Lo Lhe elecLrlclLy prlce offered by Lhe grld aL 3 ln8/kWh (0.07 uSu/kWh).
1herefore ln areas where Lhere ls no access Lo Lhe grld, Lhese deslgns offer a cosL-
compeLlLlve power supply opLlon. Lven ln areas where Lhere ls access Lo Lhe grld, Lhe
rellablllLy and quallLy of servlce provlded by Lhe grld ls llkely Lo be much worse Lhan
whaL can be provlded by local generaLlon. WlLh Lhe above sysLem deslgns, all mlcrogrld
loads can be served excepL for some lnsLances ln demand scenarlos 3 and 4, whlch
amounL Lo less Lhan 1 of Lhe LoLal annual consumpLlon. 1hls ls because Lhe sysLem ls
deslgned wlLh a 99 supply rellablllLy sLandard.

36
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

43

Microgrids with wind power
1he followlng Lable shows Lhe opLlmal generaLlon capaclLles for a mlcrogrld ln a reglon
wlLh reasonable wlnd power resources.
1ab|e 14: Cpt|ma| generat|on capac|t|es for m|crogr|ds w|th w|nd power. Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
W|nd
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 0 4000 0.48
Scenar|o 2 (100 kW) 6 100 100 4000 0.38
Scenar|o 3 (270 kW) 23 240 230 12000 0.48
Scenar|o 4 (SS4 kW) 200 300 230 40000 0.46
lgnorlng demand scenarlo 1, whlch ls Lhe absoluLe mlnlmum and unreallsLlc for a
mlcrogrld, Lhe cosL of elecLrlclLy resulLlng from Lhe opLlmal sysLem deslgn for a mlcrogrld
ln 8lhar wlLh access Lo hydro power ranges from 0.38 uSu Lo 0.48 uSu per kllowaLL-hour
(19-24 ln8/kWh). AlLhough Lhls ls hlgher compared Lo Lhe elecLrlclLy prlce offered by Lhe
grld aL 0.07 uSu/kWh (3 ln8/kWh), lL ls sLlll somewhaL ln range wlLh Lhe servlce
provlded by Pusk ower SysLem (18 ln8/kWh, correspondlng Lo 0.36 uSu/kWh).
Effect of PV price reduction
1o conslder Lhe effecL of a 40 prlce reducLlon wlLhln a only a few years as currenLly
seen on Lhe world-wlde module markeLs, PCML8 slmulaLlons were performed wlLh
lower solar v prlces for mlcrogrlds ln 8lhar wlLh and wlLhouL hydropower, and wlLh and
wlLhouL wlnd power. 1he followlng Lables show Lhe opLlmal generaLlon capaclLles for
Lhe correspondlng comblnaLlons of resources.
1ab|e 1S: Cpt|ma| generat|on capac|t|es for 8|har m|crogr|d w|thout hydropower and reduced so|ar V
pr|ce. Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 4000 0.47
Scenar|o 2 (100 kW) 6 100 4000 0.37
Scenar|o 3 (270 kW) 23 240 16000 0.47
Scenar|o 4 (SS4 kW) 600 200 160000 0.37
1ab|e 16: Cpt|ma| generat|on capac|t|es for 8|har m|crogr|d w|th hydropower and reduced so|ar V pr|ce.
Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
nydro
(kW)
D|ese|
(kW)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 0 4000 0.47
Scenar|o 2 (100 kW) 6 40 100 4000 0.22
Scenar|o 3 (270 kW) 23 180 230 10000 0.24
Scenar|o 4 (SS4 kW) 400 200 230 60000 0.26
37
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

44

1ab|e 17: Cpt|ma| generat|on capac|t|es for m|crogr|ds w|th w|nd power and reduced so|ar V pr|ce.
Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
W|nd
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 0 4000 0.46
Scenar|o 2 (100 kW) 6 100 0 4000 0.27
Scenar|o 3 (270 kW) 23 240 0 16000 0.30
Scenar|o 4 (SS4 kW) 400 300 100 80000 0.28
As can be seen, Lhe resulLs for Lhe mlcrogrlds wlLh and wlLhouL hydropower hardly
change aL all ln comparlson Lo Lhe base assumpLlons. uue Lo Lhe relaLlvely small amounL
of v, Lhe v prlce reducLlon does noL have a very blg lmpacL on Lhe cosL of energy.
Cne dlfference can be seen ln Lhe scenarlo wlLh wlnd, where Lhe v prlce reducLlon now
favours more v wlLh sLorage over Lhe wlnd power conLrlbuLlon.
Effect of accelerated solar PV price reduction
lf governmenL pollcles were deslgned Lo encourage Lhe adopLlon of solar v power, lL ls
expecLed LhaL Lhe lnsLallaLlon cosL for solar v sysLems wlll fall by up Lo 70 due Lo
economles of scale. 1o demonsLraLe Lhls effecL, PCML8 slmulaLlons were performed
wlLh lower solar v prlces for mlcrogrlds ln 8lhar wlLh and wlLhouL hydropower, and
wlLh and wlLhouL wlnd power. 1he followlng Lables show Lhe opLlmal generaLlon
capaclLles for Lhe correspondlng comblnaLlons of resources.
1ab|e 18: Cpt|ma| generat|on capac|t|es for 8|har m|crogr|d w|thout hydropower and further reduced
so|ar V pr|ce. Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 4000 0.46
Scenar|o 2 (100 kW) 6 100 8000 0.36
Scenar|o 3 (270 kW) 130 180 16000 0.43
Scenar|o 4 (SS4 kW) 700 200 160000 0.27
1ab|e 19: Cpt|ma| generat|on capac|t|es for 8|har m|crogr|d w|th hydropower and further reduced so|ar
V pr|ce. Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
nydro
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 0 4000 0.46
Scenar|o 2 (100 kW) 6 100 80 4000 0.21
Scenar|o 3 (270 kW) 23 240 230 10000 0.24
Scenar|o 4 (SS4 kW) 400 200 230 80000 0.19

38
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

4S

1ab|e 20: Cpt|ma| generat|on capac|t|es for m|crogr|ds w|th w|nd power and further reduced so|ar V
pr|ce. Source: energynaut|cs.
So|ar V
(kW)
8|omass
(kW)
W|nd
(kW)
Storage
Capac|ty
(Ah)
CoL
(USD]kWh)
Scenar|o 1 (20 kW) 1.3 30 0 4000 0.46
Scenar|o 2 (100 kW) 6 100 0 4000 0.36
Scenar|o 3 (270 kW) 130 240 0 16000 0.43
Scenar|o 4 (SS4 kW) 700 200 0 160000 0.27
lL can be seen Lhe furLher reduced v prlce now ellmlnaLes Lhe use of wlnd power even
ln reglons wlLh good wlnd poLenLlal. ln comparlson Lo Lhe prevlous scenarlos wlLh hlgher
solar prlces, Lhe energy prlces only drop slgnlflcanLly ln load scenarlo 4. 1he reason ls
LhaL Lhe assumed load proflle ln Lhls scenarlo has a hlgher correlaLlon wlLh solar
lrradlaLlon, whlch leads Lo a naLurally hlgher share of v ln Lhe energy mlx. WlLh lower
prlces for v modules Lhls dlrecLly lnfluences Lhe cosL of energy.
Conclusions from this section
1he PCML8 sLudy allowed us Lo flnd Lhe opLlmal lnsLallaLlon capaclLles of a comblnaLlon
of generaLlon resources for each demand scenarlo. 1he flrsL scenarlo represenLs Lhe
absoluLe mlnlmum level of elecLrlclLy servlce, whlle Lhe second ls Lhe baslc level of
servlce expecLed Lo be provlded ln a rural vlllage ln lndla. 1he Lhlrd and fourLh scenarlos
reflecL fuLure slLuaLlons when Lhe demand level grows as Lhe economlc sLandard
lmproves.
lf a hydro resource ls avallable, Lhls causes Lhe sysLem cosL Lo drop. 1hls would be
sulLable for areas where canals already exlsL, such as vlllages ad[acenL Lo Lhe Sole,
Candak, or kosl rlver canals. As a maLLer of facL, several hydro power planLs already
exlsL, and mlcro hydro schemes are planned for canals sLemmlng from Lhe Sole ln Lhe
souLhwesL of Lhe sLaLe already. AlLhough lL ls an economlc and easlly explolLable
resource dependlng on Lhe locaLlon, due Lo Lhe sLlgma wlLh hydro schemes Lhere are
noL many plans along Lhe kosl and Candak rlvers Lo bulld hydro power planLs. Also, even
lf a vlllage ls close Lo a rlver, a bypass canal needs Lo be creaLed whlch can be cosLly.
1herefore lL ls recommended LhaL Lhls resource be explolLed ln areas where Lhe canals
already exlsL. AnoLher lssue wlLh hydro ls LhaL Lhe maln use of Lhe canals ls for lrrlgaLlon.
1herefore Lhe avallable resource and flow raLe varles slgnlflcanLly dependlng on Lhe Llme
of Lhe year. ln some cases Lhe hydro power planL ls Laken ouL of operaLlon for 4 monLhs
because of lrrlgaLlon
33
. 1herefore lL musL be palred wlLh anoLher form of power
generaLlon such as blomass or dlesel. Wlnd resources ln Lhe sLaLe of 8lhar are noL
opLlmal for economlc operaLlon of wlnd power planLs. Lven lf Lhe wlnd speed were
..............................
33
Accordlng Lo naLwar S.P. . SallenL leaLures provlded by 8lhar SLaLe PydroelecLrlc ower CorporaLlon
39
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

46

hlgher, Lhe lnlLlal lnvesLmenL requlred ln comparlson Lo blomass, hydro and v ls such
LhaL lL does noL feaLure hlghly.
Slmllarly, alLhough belng an abundanL source, solar power also, Lake a hlgh share ln Lhe
opLlmal generaLlon mlx due Lo Lhe relaLlvely hlgh caplLal cosLs. 1hls ls Lhe case even lf
Lhe phoLovolLalc prlce ls assumed Lo fall by 40 compared Lo Lhe currenL prlces ln lndla,
brlnglng lL ln llne wlLh global prlces.
STEP 3: TECHNICAL FEASIBILITY CHECK
1o ensure LhaL Lhe generaLed elecLrlclLy can be dellvered safely Lo Lhe cusLomers, power
sysLem modelllng and slmulaLlon sofLware called owerlacLory
36
was used Lo model Lhe
operaLlon of Lhe dlsLrlbuLlon grld. 1he neLwork layouL for Lhe vlllage of Man[harla ln
8lhar was modelled, and Lhe behavlour of Lhe elecLrlcal sysLem under dlfferenL
operaLlng condlLlons was observed. lor Lhls case sLudy, Lhe opLlmal Lopology of Lhe
neLwork, requlred cable slzes, volLage conLrol and locaLlon of generaLlon asseLs were
deLermlned based on sLeady-sLaLe power flow sLudles.
1he deslgn of Lhe dlsLrlbuLlon neLwork alms Lo fulfll Lhe followlng ob[ecLlves:
1he llnes and Lransformer musL be capable of safely supplylng any peak load slLuaLlon
from any of Lhe load scenarlos:
a. Llnes and Lransformers musL noL reach Lhelr Lhermal operaLlng llmlLs aL peak
load.
b. 1he volLage aL any consumer shall noL exceed Lhe commonly used Lolerances
around Lhe nomlnal volLage of 10.
1he neLwork of Lhe analysed case sLudy musL be self-conLalned ln LhaL lL does noL
necessarlly requlre connecLlon Lo furLher ouLslde neLworks for economlc lnsLallaLlon and
operaLlon.
Allowlng volLages ouLslde Lhe 10 Lolerances musL be avolded because elecLrlc
equlpmenL may noL operaLe correcLly when Lhe volLage ls Loo hlgh or Loo low, whlch can
resulL ln equlpmenL damage or even danger Lo llfe.
Load-Related Assumptions and Network Design
1he case of Lhe Man[harla sysLem can bulld upon exlsLlng sLrucLures, as Lhere ls already
a rlce-husk based blomass power generaLor supplylng consumers ln Lhe vlclnlLy. A skeLch
of Lhe exlsLlng sysLem ls shown ln Ieh|er! Verwe|sque||e konnte n|cht gefunden
werden.. 1he general sLrucLure of Lhls sysLem ls kepL, however llnes and conflguraLlon
musL be upgraded Lo be able Lo supply even Lhe peak load from demand scenarlo 4.
..............................
36
owerlacLory ls a power sysLem slmulaLlon sofLware for deslgnlng and analyslng power sysLems. lL ls a llcensed producL
developed by ulgSlLLn1

46

hlgher, Lhe lnlLlal lnvesLmenL requlred ln comparlson Lo blomass, hydro and v ls such
LhaL lL does noL feaLure hlghly.
Slmllarly, alLhough belng an abundanL source, solar power also, Lake a hlgh share ln Lhe
opLlmal generaLlon mlx due Lo Lhe relaLlvely hlgh caplLal cosLs. 1hls ls Lhe case even lf
Lhe phoLovolLalc prlce ls assumed Lo fall by 40 compared Lo Lhe currenL prlces ln lndla,
brlnglng lL ln llne wlLh global prlces.
STEP 3: TECHNICAL FEASIBILITY CHECK
1o ensure LhaL Lhe generaLed elecLrlclLy can be dellvered safely Lo Lhe cusLomers, power
sysLem modelllng and slmulaLlon sofLware called owerlacLory
36
was used Lo model Lhe
operaLlon of Lhe dlsLrlbuLlon grld. 1he neLwork layouL for Lhe vlllage of Man[harla ln
8lhar was modelled, and Lhe behavlour of Lhe elecLrlcal sysLem under dlfferenL
operaLlng condlLlons was observed. lor Lhls case sLudy, Lhe opLlmal Lopology of Lhe
neLwork, requlred cable slzes, volLage conLrol and locaLlon of generaLlon asseLs were
deLermlned based on sLeady-sLaLe power flow sLudles.
1he deslgn of Lhe dlsLrlbuLlon neLwork alms Lo fulfll Lhe followlng ob[ecLlves:
1he llnes and Lransformer musL be capable of safely supplylng any peak load slLuaLlon
from any of Lhe load scenarlos:
a. Llnes and Lransformers musL noL reach Lhelr Lhermal operaLlng llmlLs aL peak
load.
b. 1he volLage aL any consumer shall noL exceed Lhe commonly used Lolerances
around Lhe nomlnal volLage of 10.
1he neLwork of Lhe analysed case sLudy musL be self-conLalned ln LhaL lL does noL
necessarlly requlre connecLlon Lo furLher ouLslde neLworks for economlc lnsLallaLlon and
operaLlon.
Allowlng volLages ouLslde Lhe 10 Lolerances musL be avolded because elecLrlc
equlpmenL may noL operaLe correcLly when Lhe volLage ls Loo hlgh or Loo low, whlch can
resulL ln equlpmenL damage or even danger Lo llfe.
Load-Related Assumptions and Network Design
1he case of Lhe Man[harla sysLem can bulld upon exlsLlng sLrucLures, as Lhere ls already
a rlce-husk based blomass power generaLor supplylng consumers ln Lhe vlclnlLy. A skeLch
of Lhe exlsLlng sysLem ls shown ln Ieh|er! Verwe|sque||e konnte n|cht gefunden
werden.. 1he general sLrucLure of Lhls sysLem ls kepL, however llnes and conflguraLlon
musL be upgraded Lo be able Lo supply even Lhe peak load from demand scenarlo 4.
..............................
36
owerlacLory ls a power sysLem slmulaLlon sofLware for deslgnlng and analyslng power sysLems. lL ls a llcensed producL
developed by ulgSlLLn1
table 21.
40
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

47

Radial Design Ease of operation
1he neLwork ls lald ouL as a radlal dlsLrlbuLlon sLrucLure. 1hls ls Lhe naLural cholce for Lhe
glven case sLudy slnce Lhe exlsLlng generaLlon asseLs are locaLed ln Lhe geographlcal
cenLre of Lhe vlllage clusLer. 8esldes, radlal sLrucLures are Lhe easlesL Lo operaLe.
Location of generation assets
lor pracLlcal reasons lL ls assumed LhaL addlLlonal generaLlon asseLs as deLermlned ln
Lhe PCML8 sLudy are lnsLalled close Lo Lhe exlsLlng blomass power generaLor and can
Lhus be connecLed Lo Lhe same cenLral node. A connecLlon Lo a hypoLheLlcal
Lransmlsslon sysLem wlLh hlgher volLage Lhrough a Lransformer ls also expecLed Lo be
locaLed ln Lhe same area. 1he beneflL of Lhls cholce ls LhaL power flows wlll Lhen be
slmllar ln off-grld and grld connecLed operaLlon, and Lhe llnes and faulL proLecLlon
equlpmenL can be plcked and parameLerlzed ldenLlcally.
Line lengths and initial topology
1he vlllage clusLers were asslgned Lhree-leLLer abbrevlaLlons Lo obLaln Lhe names for Lhe
nodes ln Lhe neLwork model. 1he dlsLances beLween Lhe vlllages were measured ln
Coogle LarLh and used as Lhe lengLhs of Lhe llnes connecLlng Lhe vlllages.
1ab|e 21: Inter-c|uster ||ne topo|ogy and ||ne |engths
Name Node 1 Node 2 Length |n km
8A2-8nA 8AZ 8PA 0.430
8A2-SI1 8AZ Sl1 0.364
C-8A2 LAn1 8AZ 0.730
C-kA8 LAn1 8A8 0.248
C-kAM1 LAn1 8AM1 0.100
CnA-GC8 CPA CC8 0.472
DnU-CnA uPu CPA 0.239
MAN1-MAN2 MAn1 MAn2 0.830
MAN1-I MAn1 l 0.269
kA8-DnU 8A8 uPu 0.373
kAM1-MAN1 8AM1 MAn1 0.300
kAM1-kAM2 8AM1 8AM2 0.120
SI1-8Ak Sl1 8A8 0.348
SI1-SIS Sl1 SlS 0.313

WlLhln each clusLer Lhere are addlLlonal llnes connecLlng Lhe acLual consumers
(lnsLlLuLlons, households, llghLlng) Lo Lhe maln polnL of connecLlon of Lhe clusLer. uue Lo
Lhe small slze of Lhe clusLers, Lhese llnes are consldered Lo be shorL ln comparlson Lo Lhe
lnLer-clusLer llnes and are neglecLed ln Lhe calculaLlons.
Choosing cable cross-section
An lnlLlal esLlmaLlon of Lhe requlred cable/llne cross-secLlon was made based on a few
slmple assumpLlons:
41
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

lor reasons of slmpllclLy, Lhe same cable cross-secLlon would be selecLed for all lnLer-
clusLer llnes. l.e., lndlvldual llnes are noL adapLed accordlng Lo Lhe load expecLed for
Lhem.
lor each load scenarlo, Lhe load would be spllL equally beLween Lhe Lhree feeders
connecLlng Lhe clusLers Lo Lhe cenLral node.
1he maxlmum currenL ln each feeder ls deLermlned from Lhe nomlnal volLage, Lhe
maxlmum load on Lhe feeder, and Lhe number of conducLors lnvolved (number of
phases).
1he mlnlmum cross-secLlon needed Lo wlLhsLand Lhls currenL wlLhouL overheaLlng was
deLermlned from Lables ln [18].
1aklng Lhe longesL (llne) dlsLance beLween a clusLer node and Lhe cenLral node and
assumlng a power facLory of 0.9 (lnducLlve), Lhe volLage drop was analysed. 1hls lmplles
an unreallsLlc worsL-case conslderaLlon ln LhaL Lhe enLlre load does noL normally have Lo
be LransmlLLed over Lhe full llne lengLh.
1he calculaLlons based on Lhese conslderaLlons led Lo Lhe followlng concluslons:
1he Lhermal llmlLs of Lhe llnes are noL a problem, compared Lo Lhe necesslLy of llmlLlng
Lhe volLage drop across Lhe llnes.
Cnly Lhe blggesL cross-secLlons wlLhln Lhe range of sLandard dlsLrlbuLlon cable Lypes may
be sufflclenL ln order Lo llmlL Lhe volLage drop.
SLandard cross-secLlons of dlsLrlbuLlon cables are ln Lhe range of 70-300 mm per
conducLor. lor Lhe slmulaLlon model, a cross-secLlon of 240 mm per conducLor ln a 4-
conducLor cable (Lhree phases plus neuLral) was chosen. ower flow calculaLlons would
Lhen show wheLher Lhe volLage varlaLlon aL Lhe loads can be kepL wlLhln a few per cenL,
see furLher below.
LxacL cable Lypes chosen are n?8? 2x240sm 0.6/1kv for scenarlo 1 and n?8? 4x240sm
0.6/1kv for scenarlos 2-4. ConducLor maLerlal ls copper
37
.
Load distribution
lor Lhe PCML8 sLudy Lhe LoLal load was composed of separaLe proflles generaLed for
AdmlnlsLraLlon, School, PealLh SLaLlon, ubllc LlghLlng, and households. lor Lhe power
flow sLudy Lhese proflles musL be dlsaggregaLed and asslgned Lo locaLlons ln Lhe grld.
Administration, School, Health Station
1he admlnlsLraLlon ls usually locaLed ln Lhe blggesL clusLer ln Lhe vlclnlLy. ln Lhls case Lhe
Man[harla vlllage, hence Lhe admlnlsLraLlve load proflle ls asslgned Lo a load elemenL aL
Lhe node named MAn. Slmllar argumenLs can be made for Lhe locaLlon of Lhe School
and Lhe healLh sLaLlon, Lherefore Lhey are also assumed Lo be locaLed aL Lhe MAn node.
..............................
37
lor a pracLlcal reallsaLlon lL ls llkely LhaL alumlnlum cables wlll be chosen lnsLead, because Lhey are cheaper. Powever,
slnce alumlnlum has sllghLly lower conducLlvlLy Lhan copper, even hlgher cable cross-secLlons wlll be needed.
42
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

49

Public Lighting
ubllc llghLlng ls noL only requlred ln Lhe blggesL vlllage, buL essenLlally ln all clusLers
allke. 1he LoLal publlc llghLlng demand as assumed ln Lhe PCML8 sLudy ls Lherefore
dlsLrlbuLed Lo equal parLs among Lhe 12 clusLers.
Households
1he PCML8 sLudy ls based on Lhe number of 600 households. 1hese 600 are dlsLrlbuLed
across Lhe 12 clusLers for Lhe owerlacLory sLudy as follows:
C|uster Number C|uster Number of nouseho|ds
1 8AM 41
2 l 48
3 MAn 102
4 8AZ 94
S 8PA 32
6 Sl1 42
7 SlS 40
8 8A8 32
9 8A8 29
10 CC8 66
11 uPu 36
12 CPA 18
1hls dlsLrlbuLlon key ls an esLlmaLe based on Lhe area" of each clusLer, obLalned from
Lhe polygon area as drawn ln Coogle LarLh.
Reactive Power Demand
Whlle Lhe reacLlve power demand does noL have conslderable lmpacL on Lhe acLlve
power demand, lL has a blg lmpacL on Lhe volLage drop across Lhe dlsLrlbuLlon llnes.
1ab|e 22: Assumed Consumer ower Iactors
Consumer 1ype ower Iactor (SL1) ower Iactor (SL2, SL3, SL4)
Adm|n|strat|on 0.90 capaclLlve 0.90 lnducLlve
nea|th Stat|on 0.93 lnducLlve 0.93 lnducLlve
Schoo| 0.93 lnducLlve 0.93 lnducLlve
ub||c L|ght|ng 0.80 capaclLlve 0.93 lnducLlve
nouseho|ds 0.80 capaclLlve 0.93 lnducLlve

lnducLlve loads lead Lo a hlgher volLage drop across dlsLrlbuLlon llnes and Lherefore
requlre hlgher cross-secLlons ln order Lo llmlL Lhe volLage drop for Lhe same acLlve
power consumpLlon. 1he consumpLlon ln Lhe mlnlmum load scenarlo (SL1) ls domlnaLed
by ClL llghLlng, whlch behaves more capaclLlve Lhan oLher Lypes of equlpmenL. 1he
oLher load scenarlos also lnclude oLher Lypes of consumer devlces and are Lherefore
assumed Lo have a more lnducLlve characLer.
43
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

Power Flow Analysis
1he same load proflles as used as Lhe lnpuL daLa for PCML8 are used for Lhe
consumpLlon paLLerns, as descrlbed above. All daLa ls provlded as hourly daLa for one
year. ower llow analysls ln each scenarlo ls run for all hours of Lhe year Lo obLaln Lhe
mlnlmum and maxlmum volLages as well as losses and requlred energy lmporL/exporL
(due Lo PCML8 allowlng a cerLaln amounL of non-dellvered energy and Lhe need Lo
compensaLe Lhe grld losses).
1he maln crlLerlon for a valld conflguraLlon ls LhaL Lhe volLage range noL exceeds much
more Lhan 3 beLween Lhe lowesL and hlghesL volLage aL any low-volLage node. Whlle ln
pracLlce a volLage devlaLlon of +-10 around Lhe nomlnal volLage ls allowed, we have Lo
Lake lnLo accounL LhaL Lhe volLage aL Lhe cenLral node may also flucLuaLe due Lo volLage
regulaLlon lssues aL Lhe generaLlon asseLs or due Lo volLage varlaLlons ln Lhe (medlum
volLage) Lransmlsslon grld. Cnly allowlng a delLa of 3 wlLhln Lhe low-volLage sysLem
should provlde Lhe requlred safeLy margln.
Initial one-phase per feeder configuration
Study Case S81-80-{8ase)
1terated over 8760 hours

Lnergy 1mport = 9101.96 kWh
Lnergy Lxport = 1776.68 kWh
6rd losses = 1072.68 kWh

Max Lxport: 1.0979 kW at hour: 7S24
Max 1mport: 2.27S0 kW at hour: 44
Max load: 24.S686 kW at hour: 44
Mn load: 0.0000 kW at hour: 6
Max 6eneraton: 21.S000 kW at hour: 44
Mn 6eneraton: -0.0000 kW at hour: 4
Max loss: 0.6822 kW at hour: 44
Mn loss: 0.0000 kW at hour: 2986

Max q 1mport: 0.0SS9 kvar at hour: 78S7
Mn q 1mport: -17.62S1 kvar at hour: 2229
Max q load: 0.1SSS kvar at hour: 21
Mn q load: -17.84S4 kvar at hour: 2229

Max vo1tage lv: 1.0S00 p.u. at hour: 6 at node: 8^8
Mn vo1tage lv: 1.0422 p.u. at hour: 44 at node: 8^8
Max vo1tage Mv: 1.0228 p.u. at hour: 78S7 at node: Mv_8uS
Mn vo1tage Mv: 1.0228 p.u. at hour: 2229 at node: Mv_8uS
Extension to three-phase
1o cope wlLh Lhe demand ln scenarlos 2 and 3, lL ls necessary Lo upgrade Lhe llnes from
slngle phase Lo Lhree-phase. 1hls lncreases Lhe number of conducLors from 2 Lo 4, buL
reduces Lhe volLage drop slgnlflcanLly. 1o make Lhls work ln pracLlce, lL wlll requlre LhaL
consumers wlll be connecLed ln slmllar paLLerns Lo each of Lhe Lhree phases.
Load Scenarlo 2:
Study Case S82-80

Power Flow Analysis




Load Scenarlo 2:
Study Case S82-80
44
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2


Lnergy 1mport = 10S92.27 kWh
Lnergy Lxport = 127S.8S kWh

Max Lxport: 8.S612 kW at hour: 7711
Max load: 79.2888 kW at hour: 2228
Max 6eneraton: 76.9178 kW at hour: 2228
Mn 6eneraton: 0.0000 kW at hour: 0

Max q 1mport: 26.S812 kvar at hour: 2228
Mn q 1mport: -0.2621 kvar at hour: 8072
Max q load: 26.0928 kvar at hour: 2228
Mn q load: 0.0268 kvar at hour: 221

Max vo1tage Mv: 1.02S2 p.u. at hour: 2228 at node: Mv_8uS
Load Scenarlo 3:
Study Case S82-80


Max Lxport: 22.0S42 kW at hour: 21
Max 1mport: 26.S020 kW at hour: 721S
Max load: 222.2662 kW at hour: 2226
Max 6eneraton: 222.2624 kW at hour: 2226
Mn 6eneraton: 0.0000 kW at hour: 0

Max q 1mport: 79.7724 kvar at hour: 2226
Mn q 1mport: -0.2621 kvar at hour: 7
Max q load: 72.47S8 kvar at hour: 2226

Max vo1tage Mv: 1.0282 p.u. at hour: 2226 at node: Mv_8uS

Power Flow Analysis




Load Scenarlo 2:
Study Case S82-80
45
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

As can be seen from Lhe volLage range aL low volLage (Lv) nodes, Lhe volLage range
already exceeds 3. 1hls lndlcaLes LhaL for hlgher consumpLlon lL wlll be necessary Lo
upgrade Lhe grld agaln.
Extensions to accommodate peak load in the maximum load scenario
keeplng Lhe volLage range below 3 even ln load scenarlo 4 can be achleved by Laklng
Lhe followlng measures:
1wo addlLlonal feeders, connecLlng clusLers 8A8 and CC8 dlrecLly Lo Lhe cenLral power
planL vla new llnes: C-8A8 (0.82 km), and C-CC8 (0.97 km).
lnsLalllng parallel cables ln a few secLlons: C-8AM1-MAn1-MAn2 and 8AZ-Sl1 would
need a second cable, on secLlon C-8AZ Lhe number of parallel cables musL be lncreased
Lo Lhree.
1he llnes Sl1-8A8 and CPA-CC8 are opened Lo reLaln a radlal sLrucLure.
CLher alLernaLlves would be Lo lnsLall addlLlonal connecLlon polnLs Lo Lhe (as of yeL
hypoLheLlcal) medlum-volLage sysLem, whlch would make operaLlng Lhe sysLem ln lsland
slgnlflcanLly more complex.
upgradlng Lhe sysLem as descrlbed above leads Lo Lhe followlng resulLs:
Study Case S84-80-nductve-Lxt6rd

Lnergy 1mport = 208SS.44 kWh
Lnergy Lxport = 22S.44 kWh
6rd losses = 16992.S9 kWh

Max Lxport: 8.8SS1 kW at hour: 2266
Max 1mport: 2S1.9S87 kW at hour: 2492
Max load: SS7.S181 kW at hour: 2492
Max 6eneraton: 22S.0008 kW at hour: 2492
Mn 6eneraton: S2.4S68 kW at hour: S622
Max loss: 19.S040 kW at hour: 2492

Max q 1mport: 202.2997 kvar at hour: 2492
Mn q 1mport: 12.78S7 kvar at hour: 2266
Max q load: 182.2429 kvar at hour: 2492

Max vo1tage lv: 1.0492 p.u. at hour: 2266 at node: l^u1
Mn vo1tage lv: 0.9826 p.u. at hour: 2492 at node: S1S
Max vo1tage Mv: 1.0S00 p.u. at hour: 0 at node: Mv_8uS
Mn vo1tage Mv: 1.0S00 p.u. at hour: 0 at node: Mv_8uS
46
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2
S3
I|gure 12: Network w|th upgrades to accommodate |oad scenar|o 4.
I|gure 13: Upgraded gr|d mode|

CC8
C PA
uPu
8A8
8AM
MAn
l
8AZ
8PA
S l1
S lS
8A8
0 0.3 1.0 km
CenLral S LaLlon
S lngle Cable Llne
uouble Cable Llne (upgrade)
1rlple Cable Llne (upgrade)
new S lngle Cable Llne
Gr een peace
International
PowerFac tory 14.1.1
Bi har Vi ll age Case Study
Provi ding Elect ri c ity Acc ess from Publ ic Uti li ti es
and/or from D is tributed Energy Res ourc es
i nSmal l and Mi cro Gri ds
Project: GP/MG
Graphi c :Grid
Date: 3/28/2012
Annex: S R1
PLANT
DH U
RAB
SIS
BHA BAZ
PIP
MAN2
BAR
MV_BU S
RAM1 CH A MAN1
RAM2
GOB
SIT
C-GOB
C
-B
A
R
S IS _L S IS _H
B HA _H B AZ_ H
B AZ-B HA
S IT _H
C-B AZ
B AZ_ L
P IP _H
MAN_H
S chool
Heal th S ta ti on
A dmini st ra ti on
C-RA M1
RAM_H RAM_L
R
A
M
1
-R
A
M
2
RAB _H RAB _L
RAB -DHU
DHU_H
CHA _H
GOB_H
MAN_L
P hot ovol ta ics
G
~
B iomas s
G
~
Dies el
B AR_H
S IT -B AR
E xte rn al Gr i d
T
R
F
B HA _L
GOB_L
DHU_L
P IP _L
S IT _L
RAM1-MAN1
M
A
N
1
-M
A
N
2
C
H
A
-G
O
B
D
H
U
-
C
H
A
C
-R
A
B
M
A
N
1
-P
IP
S IT -S IS
B
A
Z
-S
IT
D
Ig
S
IL
E
N
T
47
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

Lach load scenarlo comes wlLh a cerLaln peak load, whlch ls Lhe mosL lmporLanL facLor ln
dlmenslonlng Lhe dlsLrlbuLlon grld. Whlle load scenarlo 1 can sLlll be supplled uslng
slngle-phase feeders, all oLher scenarlos already need Lhree-phase feeders wlLh sLrong
cables. ln Lhe case of Lhe maxlmum assumed load, lL ls even necessary Lo upgrade
lndlvldual secLlons Lo mulLlple parallel cables, and new cables have Lo be puL ln.
8ased on Lhese slmulaLlons, a sysLem wlLh Lhe followlng key feaLures was suggesLed.
8adlal dlsLrlbuLlon sLrucLure: 1hls ls Lhe naLural cholce for a sysLem wlLh large blomass
and dlesel generaLlon asseLs locaLed aL Lhe geographlcal cenLre of Lhe vlllage. 8esldes,
radlal sLrucLures are Lhe easlesL Lo operaLe.
SLandard cross-secLlons of dlsLrlbuLlon cables are ln Lhe range of 70-300 mm per
conducLor. lor Lhe slmulaLlon model, a cross-secLlon of 240 mm per conducLor ln a 4-
conducLor cable (Lhree phases plus neuLral) was chosen.
LocaLlon of generaLlon asseLs: lor pracLlcal reasons lL ls assumed LhaL addlLlonal
generaLlon asseLs as deLermlned as approprlaLe by PCML8 wlll be lnsLalled near Lhe
cenLral blomass power generaLor and can Lhus be connecLed Lo Lhe same cenLral node.
A connecLlon Lo a hypoLheLlcal Lransmlsslon sysLem of hlgher volLage ls also expecLed Lo
be locaLed ln Lhe same area. 1he beneflL of Lhls cholce ls LhaL power flows wlll Lhen be
slmllar ln off-grld and grld connecLed operaLlon, and Lhe llnes and faulL proLecLlon
equlpmenL can be plcked and parameLerlzed ldenLlcally.
volLage aL cenLral node can be conLrolled locally when Lhe mlcrogrld ls ln lsland mode.
Powever Lhere wlll sLlll be varlaLlons, whlch depend on Lhe ouLpuL characLerlsLlcs of Lhe
generaLors connecLed Lo LhaL node. lL ls vlLal LhaL Lhe volLage varlaLlons sLay wlLhln
llmlLs, parLlcularly aL Lhe end of Lhe dlsLrlbuLlon feeders. AlLhough Lhe acLual volLage
drop across Lhe feeders depends on Lhe loads, a Lolerance of 3 has been assumed
beLween Lhe cenLral node and end of Lhe feeder. 1he volLage drop observed ln Lhe
slmulaLlons was addressed by relnforclng Lhe dlsLrlbuLlon neLwork. ln grld connecLed
mode volLage conLrol meLhod wlll depend on Lhe properLles of Lhe dlsLrlbuLlon
Lransformer whlch forms Lhe connecLlon wlLh Lhe grld.

48
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

STEP 4: DEVELOPMENT OF A CONTROL STRATEGY FOR THE SWITCHING OPERATION
1he crlLlcal feaLure of whaL makes a mlcrogrld unlque, whlch ls Lhe ablllLy Lo swlLch from
grld-connecLed mode Lo lsland mode and vlsa-versa, ls dlscussed ln Lhls secLlon. 1he alm
ls Lo suggesL sysLem deslgns LhaL achleve Lhls goal aL moderaLe lnlLlal lnvesLmenL cosLs.
1here may be several reasons for a mlcrogrld Lo swlLch from grld-connecLed Lo lsland
operaLlon mode. lor example lf Lhere ls a blackouL on Lhe cenLral grld lnduced by a faulL,
lf Lhe reglon ls sub[ecL Lo a roLaLlng blackouL due Lo lnsufflclenL supply, or lf Lhe quallLy
of supply ls undeslrable for Lhe appllcaLlon ln Lhe mlcrogrld.
ln connecLlng and dlsconnecLlng a mlcrogrld Lo and from Lhe grld, however, Lhere are a
number of slgnlflcanL Lechnlcal challenges. 1hese are relaLed Lo malnLalnlng sLable
operaLlon durlng and afLer Lhe swlLchlng evenL and requlre a sophlsLlcaLed conLrol
sLraLegy lnvolvlng monlLorlng, communlcaLlon and coordlnaLlon beLween generaLors
and sysLem operaLors.
uependlng on Lhe quallLy of servlce requlred by Lhe loads ln Lhe mlcrogrld, Lhe
regulaLlons sLlpulaLed ln Lhe grld code for operaLlon pracLlces, and number of grld
supporL feaLures deslred, several dlfferenL deslgns could be developed. lor more deLalls
on Lhese deslgns, check references [19], [20], and [21].
The simple solution
lor appllcaLlon ln developlng counLrles as a Lool Lo lmprove energy access however,
raLher Lhan deslgnlng such a compllcaLed and expenslve sysLem, lL may be sufflclenL Lo
resLrlcL Lhe swlLchlng from grld-connecLed mode Lo lsland mode ln Lhe case of supply
from Lhe cenLral grld belng unavallable for a prolonged perlod of Llme (e.g. due Lo
roLaLlng power cuLs), and Lhe LranslLlon from lsland mode Lo grld-connecLed mode when
Lhe grld supply ls back.
Such LranslLlons would be handled by manual swlLchlng, upon conflrmaLlon LhaL Lhere
wlll be a prolonged blackouL, and LhaL Lhe cenLral grld ls ready Lo recelve Lhe mlcrogrld
connecLlon back agaln. lf Lhere were a conLroller, Lhls LranslLlon would be alded wlLh Lhe
boosLlng of local generaLlon Lo maLch local load before Lhe mlcrogrld ls dlsconnecLed,
and preparlng Lhe volLage, frequency, and phase of Lhe lsland sysLem Lo be ln
synchronlzaLlon wlLh Lhe grld when re-connecLlng Lo Lhe grld. Powever, wlLhouL a
conLroller, Lhe manual swlLchlng would be accompanled by a brlef blackouL of Lhe
mlcrogrld, followed by a black sLarL operaLlon. Synchronous generaLors such as dlesel
genseLs and blomass sysLems fed by blomass gaslflers (llke Lhe Pusk ower SysLem), as
well as some modern solar v lnverLers and wlnd power converLers can supporL sysLem
blacksLarLs, Lherefore make Lhls soluLlon posslble by chooslng Lhe rlghL Lype of
componenLs. 1hls sysLem ls noL exacLly Lhe mosL eleganL soluLlon. WlLhouL a mlcrogrld
conLroller whlch can coordlnaLe Lhe ouLpuLs of local generaLlon and load, such a
swlLchlng evenL would consLlLuLe a brlef blackouL of Lhe mlcrogrld sysLem each Llme Lhe
mlcrogrld ls swlLched Lo and from grld connecLed mode Lo lsland mode. Powever, lL
removes Lhe need for a compllcaLed and expenslve conLrol scheme, Lherefore could be
49
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

S6

an economlcally accesslble soluLlon whlch sLlll provldes a more rellable supply compared
Lo a sysLem whlch relles solely on Lhe grld supply, and a cheaper supply compared Lo a
sysLem whlch does noL have Lhe opLlon Lo access grld supply.
The advanced solution
As Lhe economlc capaclLy develops, and power sysLem operaLlon capablllLles as well as
regulaLlons accommodaLe for lL, more sophlsLlcaLed conLrol sLraLegles may be
lmplemenLed. 1hls would aLLempL Lo make Lhe LranslLlons beLween grld-connecLed and
lsland modes smooLher, by lmplemenLlng a conLrol scheme LhaL can coordlnaLe Lhe
ouLpuLs of Lhe dlsLrlbuLed generaLors wlLh Lhe local demand.
Such a full-fledged sysLem would requlre a number of dlfferenL componenLs, cosLlng
approxlmaLely 3 mllllon ln8 (100,000 uSu) for one mlcrogrld (see Ieh|er! Verwe|sque||e
konnte n|cht gefunden werden.) [22]. lL should be noLed LhaL Lhls prlce esLlmaLlon ls
only lndlcaLlve, real prlces ln acLual pro[ecL lmplemenLaLlons may dlffer slgnlflcanLly.
Some componenLs such as clrculL breakers, load conLrollers, and mlcro generaLors
conLrollers are already requlred ln Lhe slmple soluLlon approach.
1ab|e 23: Investment Costs to Lstab||sh the M|crogr|d. Source: [22], pr|ces converted from LUk to INk
Asset 1ype Lst|mated Cost |n Ind|an kupees
LV c|rcu|t breaker 133200
rotect|on of MV gr|d]G Interface 666100
I|ywhee| based storage system 832600
Load Contro||ers 879200
M|cro generators Contro||ers 319300
M|cro Gr|d Centra| Contro||er 999100
Commun|cat|on System 1143700

Control issues
lL ls essenLlal LhaL a mlcrogrld ln lsland operaLlon mode can malnLaln volLage and
frequency sLablllLy ln response Lo demand flucLuaLlons as well as weaLher lnduced
varlaLlons ln renewable energy based power generaLlon. lurLhermore power quallLy
should be kepL wlLhln cerLaln sLandards, and Lhere should be sLraLegles Lo ensure sLable
operaLlon durlng neLwork dlsLurbances.
When Lhe mlcrogrld ls swlLched back Lo grld-connecLed mode, lL musL be made sure LhaL
Lhe volLage, frequency, and phase of Lhe mlcrogrld ls synchronlsed wlLh LhaL of Lhe
cenLral grld aL Lhe polnL of common coupllng. 1hls requlres a mlcrogrld conLroller aL Lhe
lnLerface wlLh Lhe grld. WlLh such a conLroller lL would also be posslble Lo conLrol Lhe
amounL of power ln[ecLed/consumed by Lhe mlcrogrld aL Lhe polnL of common coupllng,
table 23) [22].

S6


Control issues

S6


Control issues

S6


Control issues

S6


Control issues
50
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

Lherefore glvlng Lhe posslblllLy for lL Lo be coordlnaLed wlLh Lhe operaLlon of Lhe cenLral
grld Lo offer volLage and frequency conLrol servlces lf needed.
Frequency and voltage control in island mode
When Lhe mlcrogrld ls operaLlng ln lsland mode, Lhe common meLhod used Lo conLrol
volLage and frequency ls Lhe droop conLrol meLhod. A slmple approach ls presenLed ln
paper [23]. 1hls approach ls Lo lmplemenL convenLlonal f/ and v/C droop conLrol
mechanlsms Lo Lhe baLLery lnverLer. ln Lhls way, expenslve sysLems LhaL requlre
communlcaLlon and exLra cabllng for a conLrol node can be avolded. lurLhermore, Lhe
ouLpuL frequency can be esLlmaLed by measurlng Lhe acLlve power ouLpuL of Lhe
lnverLer and Laklng Lhe frequency droop lnLo accounL. 1hls approach presenLs several
advanLages as low cosL, slmple expanslon of Lhe sysLem, lncreased redundancy and
slmpllfled supervlsory conLrol [24].
Frequency and voltage control in grid-connected mode
When Lhe mlcrogrld ls connecLed Lo Lhe cenLral grld, Lhe easlesL soluLlon ls Lo have Lhe
volLage aL Lhe connecLlon polnL conLrolled by Lhe cenLral grld and noL do any acLlve
volLage conLrol ln Lhe mlcrogrld lLself. 1hls allows uslng slmple Lransformer Lypes
wlLhouL on-load Lap changlng capablllLy, buL requlres an effecLlve volLage regulaLlon
reglme ln Lhe cenLral grld where volLages are always kepL wlLhln a sufflclenLly small
band. WlLhln Lhe mlcrogrld all feeders wlll experlence a volLage drop Lhe furLher away lL
geLs from Lhe Lransformer node and Lhe larger Lhe load ls connecLed Lo Lhe feeder
branch. lL was found durlng Lhe power facLory slmulaLlon parL of Lhls sLudy LhaL lL ls
llkely LhaL Lhe volLages aL Lhe end of Lhe feeders wlll be Loo low lf Lhe feeders cover a
dlsLance of longer Lhan 1 km. 1o mlLlgaLe Lhls problem, Lhe llnes would have Lo be
relnforced or Lhe volLage would need Lo be supporLed, whlch could be done by ln[ecLlon
of acLlve or reacLlve power aL cerLaln polnLs along Lhe feeder. AnoLher opLlon ls Lo uLlllze
volLage conLrolllng Lransformers elLher aL Lhe connecLlon Lo Lhe cenLral grld or even
wlLhln Lhe same nomlnal volLage level ln long feeders.
Slnce Lhe mlcrogrld conLalns generaLors, lL ls posslble Lo lnclLe supporL from generaLors
Lo malnLaln an adequaLe volLage level aL Lhe connecLlon polnL, or Lo provlde supporL Lo
Lhe cenLral grld. Powever, such an arrangemenL would have Lo be assessed and
approved by Lhe cenLral grld operaLors before puL lnLo appllcaLlon Lo make sure LhaL lL
does noL lnLerfere wlLh conLrol schemes lmplemenLed ln Lhe cenLral grld.
When Lhe mlcrogrld ls connecLed Lo Lhe grld, Lhe frequency wlll be dlcLaLed by grld.
Harmonic effects
Parmonlc effecLs are caused by cerLaln Lypes of elecLronlc loads, such as, ClLs,
compuLers, laser prlnLers, sLereos, radlos, 1vs, fax machlnes and many oLher elecLronlc
devlces. lL lncreases Lhe losses ln Lhe sysLem, produces nolse and vlbraLlons, and
decreases Lhe llfe-Llme of moLors and Lransformers. As Lhe use of elecLronlc devlces ever
lncreaslng and are also expecLed Lo be ln Lhe mlcrogrld, lL ls an lssue LhaL needs Lo be
consldered.
51
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

S8

Accordlng Lo Lhe lLLL-3 19 recommended llmlLs [26], a rule of Lhumb assessmenL can be
applled, LhaL lf Lhe sysLem capaclLy ls greaLer Lhan Lwlce Lhe demand, Lhere ls mlnlmal
volLage dlsLorLlon and no lnLerference wlLh oLher elecLrlcal equlpmenL caused by
harmonlcs. Applylng Lhls rule of Lhumb, Lhe harmonlc effecLs for dlfferenL load scenarlos
for a Lyplcal vlllage ln 8lhar (l.e. Man[harla vlllage from Lhe case sLudy), we can flnd ouL
wheLher or noL a slgnlflcanL value of harmonlc loss ls presenL for each scenarlo (see
1able 24).
1ab|e 24: Demands (L|ectron|c Dev|ces) and m|t|gat|on methods for d|fferent scenar|os
Demand
scenar|o
Demand
(correspond|ng to
e|ectron|c dev|ces)
(kW)
System s|ze
(kW)
narmon|c |osses
s|gn|f|cant?
M|t|gat|on Method
Scenar|o1 22 22 ?es Load connecLed, PSS
38
*
Scenar|o2 27.6 100 no noL 8equlred
Scenar|o3 82.8 270 no noL 8equlred
Scenar|o4 n.A. 334 no 1ransformer Secondary
ConnecLed, PSS

1able 24 shows LhaL ln scenarlo 2, scenarlo 3 and scenarlo 4 Lhe sysLem capaclLy (sysLem
slze) ls greaLer Lhan Lwlce Lhe demand correspondlng Lo elecLronlc devlces, so Lhe
harmonlcs or Lhe volLage dlsLorLlon produced ls mlnlmal and can be neglecLed. no
mlLlgaLlon meLhods are requlred buL can sLlll be lmplemenLed. Whereas, ln case of
scenarlo 1, Lhe sysLem capaclLy (sysLem slze) ls equal Lo Lhe demand (for elecLronlc
devlces), Lherefore, hlgh values of harmonlcs are presenL. 1hls Parmonlc loss can be
mlLlgaLed by Lhe use of load connecLed Parmonlc Suppresslon SysLems (PSS).
Load connecLed PSS ls an effecLlve buL Ledlous harmonlc mlLlgaLlon meLhod. Lvery load
Lermlnal musL be flLLed wlLh an PSS, whlch ln Lhe case of Lhls mlcrogrld means each ClL
unlL (of whlch Lhere are over 1,000). uue Lo Lhls reason, lL ls noL a pracLlcal soluLlon, and
Lhere ls no oLher sulLable meLhod Lo mlLlgaLe Lhe harmonlc problem. Powever, Lhe
sysLem correspondlng Lo demand scenarlo 1 ls noL a reallsLlc sysLem for Lhe smarL
energy access sLraLegy, Lherefore no furLher soluLlon ls requlred.
Power flow at point of connection (Grid-connected mode only)
Mlcrogrlds should noL ln[ecL acLlve power lnLo hlgher volLage neLworks Lo Lhe exLenL
LhaL Lhe reverse flows cause a problem (l.e. dlsrupLlng C8). When swlLchlng Lo
lsland/grld mode conLroller has Lo balance local supply wlLh load so LhaL power flow aL
..............................
38
PSS: Parmonlc Suppresslon SysLem
52
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

CC ls zero. lor Lhls, meLerlng and conLroller ls requlred. lf Loo many communlcaLlon
and conLrol equlpmenL need Lo be lnsLalled, slmple would be Lo momenLarlly blackouL
Lhe sysLem. 1hls way durlng power cuLs/roLaLlng blackouLs whlch lasL a whlle, can be
reduced Lo a momenLary blackouL. Cf course lL ls beLLer Lo avold all blackouLs buL lL may
provlde a saLlsfacLory soluLlon aL less cosL for appllcaLlons ln rural vlllages.
1he general concluslons are LhaL such a sysLem deslgn ls feaslble, buL Lo achleve smooLh
LranslLlon beLween grld-connecLed and lsland operaLlon modes sophlsLlcaLed conLrol
equlpmenL ls requlred whlch may noL be easlly accesslble flnanclally. 1herefore an
alLernaLlve sysLem ls proposed, where each Llme a sLaLe LranslLlon ls requlred Lhe
mlcrogrld power supply ls black-sLarLed. Such as sysLem may be operaLed manually, and
llmlLed Lo Llmes when a prolonged supply shorLage ls experlenced elLher ln Lhe local
generaLlon pool or ln Lhe cenLral grld. 1hls may seem a clumsy approach aL flrsL,
however, ln areas where Lhe cenLral grld ls ofLen plagued by blackouLs and cuL off supply
Lo vlllages ln favour of loads ln Lhe clLles, Lhls dellvers far superlor energy supply ln Lerms
of rellablllLy and quallLy.
As Lhe local economy grows, Lhanks Lo Lhe access Lo elecLrlclLy, dependlng on Lhe needs
of Lhe cusLomers ln Lhe mlcrogrld, Lhe swlLchlng sysLem may be auLomaLed and made
more sophlsLlcaLed (so LhaL Lhere ls no need Lo blackouL Lhe sysLem for each LranslLlon),
buL Lhls would depend on how prepared are Lhe cusLomers Lo flnance Lhe cosLs of such a
sysLem. lor example, Lhls slLuaLlon may become Lrue lf lndusLrlal cusLomers LhaL requlre
superlor supply quallLy rlse up ln Lhe vlllage, or Lhe economlc power ln Lhe vlllage
reaches a cerLaln level LhaL lndlvldual households expecL a beLLer servlce quallLy.
Some of Lhe advanced feaLures a mlcrogrld can offer lf a full conLroller ls lmplemenLed,
ls Lhe ablllLy Lo offer frequency and volLage conLrol servlces Lo Lhe cenLral grld, as well as
conLrlbuLe Lo black sLarLs lf Lhe cenLral grld experlences a black ouL. 1hese are aspecLs of
mlcrogrlds LhaL are belng lnvesLlgaLed as ongolng research ln reglons llke Lhe uSA
(CL81S), Lurope (MC8L mlcrogrlds), and !apan (Pachlnohe).
39

ldeally, slmulaLlons of such operaLlon could also be carrled ouL uslng power sysLem
slmulaLlon sofLware such as owerlacLory. Powever, Lhls requlres deLalled modelllng of
Lhe dynamlc characLerlsLlcs of each componenL ln Lhe sysLem, Lherefore has noL been
carrled ouL ln Lhls case sLudy.

..............................
39
Some advanced feaLures have been recenLly demonsLraLed ln a uLlllLy-scale dlsLrlbuLlon grld ln
uenmark, ln Lhe Cell ConLroller lloL ro[ecL run by LnerglneL.dk. lor more lnformaLlon see
hLLp://www.energlneL.dk/Ln/lC8SknlnC/nyheder/Slder/uen-usLyrllge-vlnd-kan-sLyres.aspx
53
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

60

COST COMPARISON WITH GRID EXTENSION
1o deLermlne wheLher sysLem cosLs can be a supporL Lo [usLlfy Lhe boLLom-up approach
of bulldlng a mlcrogrld over exLendlng Lhe grld, lL ls useful Lo compare Lhem wlLh grld
exLenslon cosLs. 1hls ls done comparlng Lhe opLlmal sysLems cosLs calculaLed by PCML8
wlLh Lhe cosL of exLendlng Lhe grld, based on Lhe approxlmaLlon glven by CusL eL al
40
.
lL can be seen from llgure 14 LhaL all mlcrogrlds have a hlgher sysLem cosL compared Lo
Lhe raLes LhaL would be pald by vlllage resldenLs Lo access grld-connecLed servlces.
Powever, Lhe furLher away from Lhe cenLral grld Lhe vlllage, Lhe cosLs of exLendlng Lhe
grld lncreases. 1herefore aL some polnL Lhe mlcrogrld sysLem cosL would be cheaper
Lhan accesslng grld servlces by exLendlng Lhe grld. 1hls ls known as Lhe bteok-eveo qtlJ
Jlstooce.
1he cosLs of elecLrlclLy produced by Lhe mlcrogrlds are ploLLed agalnsL Lhe prlce of grld-
supplled elecLrlclLy lf Lhe grld ls exLended ln llgure 14. lL can be seen LhaL Lhe grld-
supplled elecLrlclLy ls Lhe cheapesL soluLlon when Lhe dlsLance Lo Lhe grld ls less Lhan
3 km. Powever, dependlng on wheLher small hydropower ls avallable ln Lhe mlcrogrld
and also dependlng on Lhe prlces for solar v, Lhe break-even grld dlsLance varles
beLween 3 km and 20 km. 1he currenLly lmplemenLed Pusk ower SysLem has a break-
even grld dlsLance of around 13 km.

I|gure 14: Cost compar|son of m|crogr|d systems (w|th d|ese|) w|th gr|d extens|on. Mod|f|ed from [2]
..............................
40
bttp.//www.eptq.qtoop.com.oc.ok/wp-cooteot/oplooJs/2008/11/eptq07J0.pJf
0
3
10
13
20
23
30
33
0 3 10 13 20 23
C
o
s
t

(
I
N
k
]
k
W
h
)
Connect|v|ty (km from ex|st|ng gr|d)
Costs of opt|ma| systems (Urban Demand)
Crld Cff-grld 8emoLe
Crld exLenslon: 1 ln8/kW/km
Crld elecLrlflcaLlon cosL: 3 ln8/kWh
LffecLlve Larlff: 0-3 ln8/kWh
Pusk ower SysLem: 18 ln8/kWh
Plgh v cosL, wlLh Pydro: 17.9 ln8/kWh
Medlum v cosL, no Pydro, no Wlnd
Low v cosL, no Pydro, no Wlnd
Medlum v cosL, wlLh Pydro
Plgh v cosL, no Pydro, no Wlnd
Low v cosL, wlLh Pydro
54
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

61

Microgrid scale-up across the state of Bihar
CurrenLly ln 8lhar, Lhere are over 19,000 un-elecLrlfled vlllages across Lhe sLaLe (Lhls ls
more Lhan half of Lhe LoLal number of vlllages ln 8lhar)
41
. 8ased on a rough geographlc
assessmenL of Lhe sLaLe, lL can be esLlmaLed LhaL approxlmaLely 3,140 vlllages lle ln Lhe
reglon ad[acenL Lo rlvers and wlLh access Lo bypass canals (See llgure 13).

I|gure 1S: Map of 8|har w|th reg|ons ad[acent to r|vers and w|th access to bypass cana|s marked |n red
shad|ng.
lf Lhese are powered by mlcrogrlds wlLh hydro (8lhar mlcrogrld wlLh hydro opLlmal
sysLem), and Lhe remalnlng un-elecLrlfled vlllages ln Lhe sLaLe of 8lhar are powered by
mlcrogrlds wlLhouL hydro (Man[harla opLlmal sysLem), aL mlnlmum (correspondlng Lo
demand scenarlo u2) approxlmaLely 1700 MW of blomass, 314 MW of hydro and 114
MW of v power lnsLallaLlons would be requlred, aL a cosL of approxlmaLely 20 bllllon
uS-uollar (lncludlng generaLlon asseLs and neLwork equlpmenL cosL). AL Lhe sLage when
mlcrogrlds are fully lnLegraLed wlLh Lhe cenLral grld (correspondlng Lo demand scenarlo
u4), lL ls expecLed LhaL aL leasL 4,000 MW of blomass, 783 MW of hydro and 10,000 MW
of v power lnsLallaLlons would be requlred, aL a cosL of around 64 bllllon uS-uollar
(wlLh medlum v prlce). 1hls ls summarlsed ln
..............................
41
hLLp://www.lndlaenergyporLal.org/engsLaLs/sLaL16.hLm
330 vlllages
1,320 vlllages
710 vlllages
780 vlllages

1able 23 and 1able 26 below.

55
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2

63

1ab|e 2S: Summary of state-w|de expans|on of opt|ma| m|crogr|d capac|t|es (MW)
key 8esulLs for L[8] vlllageclusLer - energy mlx opLlons - 8lhar SLaLe 8ural
Insta||ed capac|ty 8|har State kura| Generat|on by 1echno|ogy
Scenar|o V W|nd nydro
8|o
energy
V W|nd nydro
8|o
energy
[MW] [MW] [MW] [MW] [CWh/a] [CWh/a] [CWh/a] [CWh/a]
n|gh V Cost Scenar|o
u1 AbsoluLe Mlnlmum (sLaLe-wlde) 29 0 0 370 49 0 0 730
u2 Low lncome demand (sLaLe-wlde) 114 0 314 1 712 196 0 1 847 3 128
u3 Medlum lncome demand (sLaLe-wlde) 473 10 783 4 372 818 27 4 183 10 349
u4 urban households (sLaLe-wlde) 10 012 370 783 3 990 17 233 1 369 4 183 13 804
Med|um V Cost Scenar|o
u3 Medlum lncome demand (sLaLe-wlde) 473 0 783 4 372 818 0 4 183 10 334
u4 urban households (sLaLe-wlde) 10 392 190 783 3 990 17 890 436 4 183 13 673
Low V Cost Scenar|o
u3 Medlum lncome demand (sLaLe-wlde) 2 438 0 783 4 372 4 231 0 4 183 8 724
u4 urban households (sLaLe-wlde) 12 338 0 783 3 800 21 274 0 4 183 14 080
Source: Lnergy [8]evoluLlon ClusLer - SmarL Lnergy Access, energynauLlcs May 2012 and Creenpeace lnLernaLlonal

1ab|e 26: Investment overv|ew for state-w|de expans|on |n 8|har
key 8esulLs lnvesLmenL Cvervlew - 8ural 8lhar
Generat|on and Network Lqu|pment kura| 8|har
Scenar|o
Investment ower
Generat|on
Investment Network
Lqu|pment
1ota| Investment
mllllon uS$ bllllon ln8 mllllon uS$ bllllon ln8 mllllon uS$ bllllon ln8
n|gh V Cost Scenar|o
u1 AbsoluLe Mlnlmum (sLaLe-wlde) 1 334 70.1 13 226 694.4 14 360 764.4
u2 Low lncome demand (sLaLe-wlde) 4 109 213.7 13 486 813.0 19 393 1 028.8
u3 Medlum lncome demand (sLaLe-wlde) 10 198 333.4 13 486 813.0 23 684 1 348.4
u4 urban households (sLaLe-wlde) 60 237 3 163.3 23 602 1 239.1 83 839 4 402.6
Med|um V Cost Scenar|o
Low V Cost Scenar|o
Source: Lnergy [8]evoluLlon ClusLer - SmarL Lnergy Access, energynauLlcs May 2012 and Creenpeace lnLernaLlonal

56
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2
57
m
i
c
r
o
g
r
i
d

d
e
s
i
g
n

c
a
s
e

s
t
u
d
y

f
o
r

b
i
h
a
r
2
Solar Photovoltaic Power Plant in Tangtse
THE 100 KWP STAND-ALONE SOLAR PHOTOVOLTAiC POWER PLANT AT TANGTSE, DuRBuK
BLOCK, LADAKH. LOCATED 14,500 FEET AMSL iN THE HiMALAyA, THE PLANT SuPPLiES
ELECTRiCiTy TO A CLiNiC, A SCHOOL AND 347 HOuSES iN THiS REMOTE LOCATiON, FOR
AROuND FiVE HOuRS EACH DAy.

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
MACHiNES iN SAAP WORKSHOP iN BiHARELECTRiCiAN
KOSMOS BHAKLA MEASuRES THE CHARGiNG LEVEL
OF A BATTERy BANK iN SOLAR ALTERNATiVES AND
ASSOCiATED PROGRAMMES (SAAP), A JESuiT-RuN
RESEARCH CENTRE PROViDiNG SOLAR SOLuTiON TO
ENERGy NEEDS. THE BATTERiES ARE CHARGED WiTH
ELECTRiCiTy GENERATED By SOLAR PHOTOVOLTAiC
PANELS ON THE ROOF. ST. MARyS COMPOuND,
PHuLWARiA SHARiF, PATNA, BiHAR.

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
3
grid technology +
defnitions

6S

3 GkID 1LCnNCLCGILS AND DLIINI1ICNS
ENERGY ACCESS OPTIONS
1o lncrease energy access, Lhe cenLral grld can be exLended Lo rural areas where
elecLrlclLy servlces dld noL reach before. ln Lhls way Lhe cenLral grld can lncrease lLs
cusLomer base and sell more elecLrlclLy. Powever, ln cases where Lhere ls lnsufflclenL
supply capaclLy, or Lhe cosL of exLendlng Lhe grld ouLwelghs Lhe beneflLs galned by
selllng addlLlonal energy, Lhe case for grld exLenslon ls noL as favourable any longer.
lurLhermore, Lo lnLegraLe renewable energy supply ln Lhe cenLral grld, lnvesLmenLs ln
large-scale 8L pro[ecLs are needed. 1hls ls why alLernaLlve opLlons are offered as energy
access opLlons. 1hese are lnLroduced below.
Energy Homes Systems
Lnergy home sysLems are for baslc elecLrlflcaLlon of a
slngle household or bulldlng. lL conslsLs of a small
generaLor, ofLen solar v or wlnd powered, wlLh a
baLLery and a charge conLroller. 1hls Lype of sysLem ls
deslgned Lo be connecLed dlrecLly Lo uC loads of 13-130
Wp, such as lamps, radlos, Lelevlslons and refrlgeraLors.
lor AC loads, an lnverLer would be necessary. 1he
charge conLroller connecLs Lhe generaLor Lo Lhe baLLery
and conLrols Lhe sLaLe of charge. CfLen Lhe baLLery wlll
have 3 Lo 3 days sLorage capaclLy.
Off-grid Distribution Systems
Cff-grld dlsLrlbuLlon sysLems are essenLlally a small
verslon of Lhe cenLral grld. lL connecLs several loads wlLh
several generaLlon sources and a cenLral conLroller
manages Lhe supply-demand balance. 1hese sysLems are
ofLen used for powerlng remoLe communlLles where Lhe
cosL of exLendlng Lhe cenLral grld ls consldered Loo hlgh.
1here are many sysLems ln appllcaLlon around Lhe world
and lL ls well sulLed Lo explolLaLlon of local renewable
energy resources. uue Lo Lhe seasonal flucLuaLlons and
weaLher dependency of many of Lhese sources
however, lL ls besL Lo use a comblnaLlon of energy
sources ln order Lo have a conLlnuous supply. lor
example varlable sources such as solar, wlnd and run-of-
rlver hydro should be coupled wlLh baLLerles, dlesel
genseLs or blomass-based generaLors LhaL can adapL
qulckly Lo flucLuaLlng condlLlons. 1hese are called hybrld
llgure 16: ConflguraLlon of a Solar
Lnergy Pome SysLem. Source: [33]
llgure 18: ConflguraLlon of a Pybrld
SysLem wlLh uC-AC connecLlon. Source:
[33]
llgure 17: ConflguraLlon of a Pybrld
SysLem wlLh AC connecLlon. Source: [33]

sysLems and can be deslgned as uC sysLems, mlxed uC/AC sysLems or pure AC sysLems.
Microgrids
Mlcrogrlds are dlsLrlbuLlon sysLems conslsLlng of loads and generaLors balanced wlLh a
conLroller, as wlLh Lhe off-grld dlsLrlbuLlon sysLem, buL connecLed Lo Lhe grld aL a slngle
polnL of connecLlon (called Lhe polnL of common coupllng). 1he unlque feaLure of
mlcrogrlds ln Lhls sense ls Lhe ablllLy Lo swlLch beLween grld-connecLed and lsland
operaLlon modes. 1hls allows for flexlble operaLlon Lo access grld supply aL Llmes when
local demand ls hlgh, exceedlng Lhe local generaLlon capaclLy, or when Lhe grld power
prlces are low, whlle uslng local generaLlon when Lhere ls a supply shorLage ln Lhe
cenLral grld, or when Lhe grld power prlces are hlgh. 1hls lncreases Lhe rellablllLy of
supply Lo Lhe mlcrogrld cusLomers whlle slmulLaneously supporLlng Lhe cenLral grld by
creaLlng a generaLlon/demand block LhaL can conLrolled Lo acL as a load or a generaLor
and malnLaln Lhe volLage aL Lhe polnL of common coupllng. AddlLlonally, lL may be
posslble for Lhe mlcrogrld Lo conLrlbuLe Lo black sLarLlng Lhe cenLral grld.
Distributed Generation Systems
ulsLrlbuLed generaLlon sysLems are small-scale generaLors connecLed Lo Lhe dlsLrlbuLlon
neLwork of Lhe cenLral grld. LlecLrlcally lL acLs Lo reduce Lhe dlsLrlbuLlon load LhaL Lhe
cenLral grld sees aL Lhe polnLs of common coupllng. Slnce reacLlve power can also be
ln[ecLed, volLage supporL can also be offered Lo a cerLaln exLenL. unllke Lhe mlcrogrlds
however, dlsLrlbuLed generaLlon are noL made Lo balance local demand by a cenLral
conLroller, and cannoL run dlsconnecLed from Lhe cenLral grld.
OBSTACLES TO ENERGY ACCESS PROJECTS
AlLhough Lhere are programs Lo lmprove energy access by exLendlng Lhe grld, ln many
cases Lhe reallLy ls far behlnd LargeLs. WhaL are Lhe maln causes for delayed dellvery of
energy servlces? ln Lhe reporL on Afrlca's power lnfrasLrucLure by Lhe World 8ank [27],
Luropean-funded mlcrogrld program ln Afrlca [28] and a reporL on rural elecLrlflcaLlon ln
lndla [21], Lhe followlng lssues are reporLed as belng barrlers Lo rural elecLrlflcaLlon.

Microgrids

59
g
r
i
d

t
e
c
h
n
o
l
o
g
y

+

d
e
f
i
n
i
t
i
o
n
s
3

Microgrids

60
g
r
i
d

t
e
c
h
n
o
l
o
g
y

+

d
e
f
i
n
i
t
i
o
n
s
3

67

I|nanc|a|
ro[ecL cosL Loo hlgh
CaplLal cosL Loo hlgh Lack of flnanclal capaclLy of:
- grld owner
- power producer
- servlce user
1axes are Loo hlgh
C&M cosL Loo hlgh luel prlce Loo hlgh
1ransmlsslon and dlsLrlbuLlon losses Loo hlgh:
- sysLem deslgn
- energy LhefL
8evenue Loo low vlllage slze Loo small
uLlllLy fee affordable Lo consumer ls Loo low
Cperat|ona|
lnsufflclenL sysLem capaclLy
Lo manage pro[ecL
lnLegraLlon
Supply capaclLy
shorLage
Lack of lnfrasLrucLure
Lnergy LhefL
lrequenL dlsrupLlon
ln supply of grld
power
1he prevalence of backup generaLors
lncreaslng use of leased emergency power
Adm|n|strat|ve
ro[ecL lmplemenLaLlon Loo
complex
lnadequaLe
provlslons ln
regulaLlon
Lack of provlslon for uC, 8L, flexlble generaLlon
Lack of provlslon Lo manage mlsmaLches beLween servlce LhaL
can be provlded by grld and Lhe servlce needed by cusLomer
lnadequaLe
lnformaLlon
no 8L poLenLlal sLudles have been done or daLa avallable
Lack of evaluaLlon procedures Lo ldenLlfy besL pracLlces
Lack of organlsaLlonal schemes and well deflned procedures for
Lhe managemenL and malnLenance of elecLrlflcaLlon pro[ecLs
uncoordlnaLed
pollcles
Slow and complex admlnlsLraLlve processes
Soc|a|
ro[ecL noL supporLed by
local populaLlon
Lack of access Lo
educaLlon
Lack of educaLlon Lo develop undersLandlng and accepLance of
new energy sysLems and Lo culLlvaLe respecL Lo malnLaln Lhe
sysLem, lack of quallfled local Lechnlclans, eLc
LnvlronmenLal CpposlLlon Lo cerLaln Lechnologles due Lo aesLheLlcs obsLrucLlon,
compeLlng land space wlLh agrlculLural producLlon, eLc

STRENGTHS AND WEAKNESSES OF ENERGY ACCESS OPTIONS
8ased on Lhe flndlngs ln Lhe above Lable, for Lhe energy access opLlon Lo be wldely
adopLed ln developlng counLrles, lL would need Lhe followlng feaLures:
8easonable and affordable caplLal cosL
8easonable and affordable C&M cosL
8easonable and affordable power prlces
Mlnlmal lmpacL on exlsLlng regulaLlon and operaLlon pracLlces
Can be operaLed and malnLalned by local communlLles
Can be lnLegraLed wlLh cenLral grld ln Lhe fuLure
uoes noL demand addlLlonal capaclLy from cenLral grld when connecLed

Microgrids

61
g
r
i
d

t
e
c
h
n
o
l
o
g
y

+

d
e
f
i
n
i
t
i
o
n
s
3

68

1he sLrengLhs and weaknesses ln addresslng Lhe obsLacles ofLen seen ln developlng
counLrles of Lhe appllcaLlon of each of Lhese opLlons ln developlng counLrles are glven ln
1able 27. 1hey are compared wlLh grld exLenslon (Lop-down) and dlsLrlbuLed generaLlon.
1ab|e 27: Strengths and weaknesses of d|fferent energy access opt|ons. Source: energynaut|cs
Lnergy Access
A|ternat|ves
ros Cons
Standa|one nLS
(|s|and)
LowesL caplLal cosL due Lo small unlL slze
no Lransmlsslon or dlsLrlbuLlon llnes
need Lo be bullL
lasL and slmple consLrucLlon
no change ln regulaLlon requlred
8ellable supply can be susLalned when
grld ls down
Supply accesslble Lo only one user
(no beneflL from economy of scale)
M|n|-gr|d system
(|s|and)
no Lransmlsslon llnes need Lo be bullL
Supply can be shared beLween mulLlple
users
grld ls down
ulsLrlbuLlon llnes need Lo be bullL
8equlres locallzed balanclng conLrol
LffecLlve paymenL sLrucLure needs
Lo be esLabllshed
Mlnl-grld operaLor needs Lo be
esLabllshed
M|crogr|d system
(|s|and gr|d)
lnvesLmenL ln Lransmlsslon exLenslon can
be delayed unLll lL can be economlcally
[usLlfled
users
grld ls down
Supply from grld can be accessed when
energy ls cheaper and/or ln case of local
generaLlon ouLage
Lxcess local supply can be fed Lo grld
Mlcrogrld can supporL grld volLage and
black sLarL operaLlon
ulsLrlbuLlon llnes need Lo be bullL
8equlres locallzed balanclng conLrol
8equlres an effecLlve lndependenL
paymenL sLrucLure unLll grld access
ls avallable
An operaLor of Lhe lsland grld needs
Lo be esLabllshed unLll grld access ls
avallable
8equlres communlcaLlon and
conLrol equlpmenL Lo swlLch
beLween operaLlon modes
8equlres revlew of proLecLlon
equlpmenL and regulaLlon
Gr|d extens|on
w|th DG
(gr|d)
Access Lo economles of scale
Access Lo cenLrallzed balanclng conLrol
no new paymenL sLrucLure needs Lo be
developed for consumers
1ransmlsslon and dlsLrlbuLlon llnes
need Lo be bullL
8ellablllLy of supply ls dependenL on
cenLrallzed sysLem (small vlllages are
unllkely Lo become hlgh prlorlLy
areas for malnLalnlng dellvery of
quallLy and rellablllLy of supply)
Small load on a long feeder ls prone
Lo volLage problems
62
g
r
i
d

t
e
c
h
n
o
l
o
g
y

+

d
e
f
i
n
i
t
i
o
n
s
3

69

D|str|buted
generat|on
(gr|d)
users
Access Lo cenLrallzed balanclng conLrol
uC can supporL grld volLage and black
sLarL operaLlon
no new paymenL sLrucLure needs Lo be
developed for consumers
PlghesL caplLal cosL because
Lransmlsslon and dlsLrlbuLlon llnes
need Lo be bullL as well as
generaLlon unlLs
8ellablllLy of supply ls dependenL on
cenLrallzed sysLem (small vlllages are
unllkely Lo become hlgh prlorlLy
areas for malnLalnlng dellvery of
quallLy and rellablllLy of supply)
8equlres revlew of regulaLlon
ower purchase agreemenLs need Lo
be esLabllshed

63
g
r
i
d

t
e
c
h
n
o
l
o
g
y

+

d
e
f
i
n
i
t
i
o
n
s
3
Child with Solar Lantern in Bihar
KHuSBOO KuMARi (LEFT) (8) AND REKHA KuMARi (12) STuDy iN A ROOM LiT By A PORTABLE
SOLAR LANTERN. THE LANTERN PROViDES LiGHT FOR ABOuT FOuR HOuRS DAiLy AND REquiRES
CHARGiNG EACH DAy AT THE KOTAK uRJA 3 KVA SOLAR STATiON. THE SOLAR STATiON WAS SET uP
iN APRiL 2010 iN A HAMLET OF MuSAHARS - THE POOREST OF THE POOR CASTE iN BiHAR - WHiCH
HAS NEVER HAD AN ELECTRiCiTy CONNECTiON. THE SOLAR PLANT, iNSTALLED AT SiKANDARPuR,
ALSO PROViDES POTABLE WATER, COMMuNiTy TV, A TELEPHONE BOOT AND CELLPHONE
CHARGiNG POiNTS. SiKANDARPuR ViLLAGE, DANAPuR, BiHAR.

H
a
r
i
k
r
i
s
h
n
a

K
a
t
r
a
g
a
d
d
a

/

G
r
e
e
n
p
e
a
c
e
4
references

hLLp://www.energynaut|cs.com 71

4 kLILkLNCLS

[1] CovernmenL of 8lhar llnance ueparLmenL, Lconomlc Survey 2010-11,"
CovernmenL of 8lhar llnance ueparLmenL, aLna, 2011.
[2] !. CusL, A. Slngh and k. neuhoff, 8ural LlecLrlflcaLlon ln lndla: Lconomlc and
lnsLlLuLlonal AspecLs of 8enewables," unlverslLy of Cambrlge , Cambrldge, 2007.
[3] lannlng Commlsslon ower & Lnergy ulvlslon, CovernmenL of lndla, Annual
8eporL 2011-12 on Lhe worklgn of SLaLe ower uLlllLles and LlecLrlclLy
ueparLmenLs," CovernmenL of lndla, uelhl, 2011.
[4] lnLernaLlonal Lnergy Agency, World Lnergy CuLlook 2011: Lnergy for All -
llnanclng Access for Lhe oor," CLCu/lLA, arls, 2011.
[3] CovernmenL of lndla: MlnlsLry of new and 8enewable Lnergy, 8evlsed CaplLal
Subsldy and 8enchmark cosL of Lhe Sv sysLem," 1 Aprll 2011. [Cnllne]. Avallable:
hLLp://mnre.gov.ln/flle-
manager/userllles/8evlsed_CaplLal_Subsldy_and_8enchmark_cosLofLhe_Sv_sys
Lem.pdf. [Accessed 28 lebruary 2012].
[6] M. 8. nounl, S. C. Mulllck and 1. C. kandpal, hoLovolLalc pro[ecLs for
decenLrallzed power supply ln lndla: A flnanclal evaluaLlon," oetqy lollcy, vol. 34,
pp. 3727-3738, 2006.
[7] M. 8. nounl, S. C. Mulllck and 1. C. kandpal, rovldlng elecLrlclLy access Lo
remoLe areas ln lndla: An approach Lowards ldenLlfylng poLenLlal areas for
decenLrallzed elecLrlclLy supply," keoewoble & 5ostolooble oetqy kevlews, vol.
12, pp. 1187-1220, 2008.
[8] M. 8. nounl, S. C. Mulllck and 1. C. kandpal, 8lomass gaslfler pro[ecLs for
decenLrallzed power supply ln lndla: A flnanclal evaluaLlon," oetqy lollcy, vol. 33,
pp. 1373-1383, 2007.
[9] AlLernaLe Pydro Lnergy CenLre, Model ueLalled ro[ecL 8eporL: Mlcro Pydro
LlecLrlc 8ased LlecLrlcal ower lanL (1 x 10 kW)," MlnlsLry of new and
8enewable Lnergy, CovernmeLn of lndla, new uelhl, 2007.
LlecLrlc 8ased ower lanL (2 x 20 kW)," MlnlsLry of new and 8enewable Lnergy,
CovernmenL of lndla, new uelhl, 2007.
LlecLrlc based LlecLrlcal ower lanL (2 x 30 kW)," MlnlsLry of new and 8enewable
Lnergy, CovernmenL of lndla, new uelhl, 2007.
65
r
e
f
e
r
e
n
c
e
s
4


[12] CovernmenL of lndla, 8oad Map for uevelopmenL of ower SecLor ln 8lhar: A
8eporL of Lhe Speclal 1ask lorce on 8lhar," CovernmenL of lndla, new uelhl, 2007.
[13] Wlndcare lndla, 5moll wloJ 1otbloe, ColmbaLore, 1amll nadu: Wlndcare lndla.
[14] 8. 8aner[ee, Comparlson of opLlons for dlsLrlbuLed generaLlon ln lndla," oetqy
lollcy, vol. 34, pp. 101-111, 2006.
[13] 8. 8aner[ee, Comparlson of opLlons for dlsLrlbuLed generaLlon ln lndla," oetqy
lollcy, vol. 34, pp. 101-111, 2006.
[16] Alllance for 8ural LlecLrlflcaLlon, Shlnlng a LlghL for rogress - 8esL pracLlses of
Lhe Alllance for 8ural LlecLrlflcaLlon: whaL renewable energy can achleve ln
developlng counLrles," Alllance for 8ural LlecLrlflcaLlon, 8russels.
[17] C. 8oyle, Lmpowerlng 8lhar: Case SLudles for 8rldglng Lhe Lnergy ueflclL and
urlvlgn Change," Creenpeace lndla, 8engaluru, 2010.
[18] A88 AC, SchalLanlagen, 8. Auflage Prsg., uusseldorf: Cornelsen verlag Schwann-
ClrardeL, 1988.
[19] n. PaLzlargyrlou, P. Asano, 8. lravanl and C. Marnay, Mlcrogrlds: An Cvervlew of
Cngolng 8esearch, uevelopmenL, and uemonsLraLlon ro[ecLs," l lowet &
oetqy Moqozloe, vol. 3, no. 4, pp. 78-94, !uly/AugusL 2007.
[20] 1. L. uel Carplo Puayllas, u. S. 8amos and 8. L. vasquez-Arnez, Mlcrogrld
SysLems: CurrenL SLaLus and Challenges," ln 1toosmlssloo ooJ ulsttlbotloo
coofeteoce ooJ xposltloo. lotlo Ametlco, Sao aolo, 2010.
[21] l. MlLra, 1. uegner and M. 8raun, ulsLrlbuLed CeneraLlon and Mlcrogrlds for
Small lsland LlecLrlflcaLlon ln ueveloplng CounLrles - A 8evlew," 5olot oetqy
5oclety of loJlo, vol. 18, no. 1, pp. 6-20, !anuary-!une 2008.
[22] . Molses CosLa and M. A. MaLos, Lconomlc Analysls of Mlcrogrlds lncludlng
8ellablllLy AspecLs," ln 9tb lotetootloool coofeteoce oo ltoboblllstlc MetboJs
ApplleJ to lowet 5ystems, SLockholm, Sweden, 2006.
[23] l. u. kanellos, A. l. 1souchnlkas and n. u. PaLzlargyrlou, Mlcro-Crld SlmulaLlon
durlng Crld-ConnecLed and lslanded Modes of CperaLlon," ln lotetootloool
coofeteoce oo lowet 5ystems 1toosleots, MonLreal, 2003.
[24] A. Lngler, AppllcablllLy of uroops ln Low volLage Crlds," lotetootloool Iootool of
ulsttlboteJ oetqy 5ootces, vol. 1, no. 1, pp. 3-13, 2004.
[23] M. C. !. C. a. A. C. !.L. Pernndez1, Parmonlcs ln ower SysLems," ln lotetootloool
coofeteoce oo keoewoble oetqles ooJ lowet, valencla (Spaln), 13Lh Lo 17Lh
Aprll, 2009.
[26] C. Poward and . L. Murphy, wbot ls Meoot by Meetloq tbe kepoltemeots of
l-519-1992?, PlLachl Amerlca.
66
r
e
f
e
r
e
n
c
e
s
4


[27] A. Lberhard, C. 8osnes, M. ShkaraLan and P. vennemo, Afrlca's ower
lnfrasLrucLure: lnvesLmenL, lnLegraLlon, Lfflclency," 1he World 8ank, WashlngLon
uC, 2011.
[28] !. . Alzola, P. Camblong and 1. nlang, Mlcrogrlds: romoLlon of mlcrogrlds and
renewable energy sources for elecLrlflcaLlon ln developlng counlrles," 8C8C1lkL8,
Zamudlo, Spaln, 2008.
[29] SCuA8L u, ower SysLem Parmonlcs Causes and LffecLs of varlable lrequency
urlves 8elaLlve Lo Lhe lLLL 319-1992 SLandard," SCuA8L u roducL uaLa 8ulleLln,
8alelgh, nC, u.S.A., AugusL, 1994.
[30] M. C. !. C. a. A. C. !.L. Pernndez1, Parmonlcs ln ower SysLems," ln lotetootloool
coofeteoce oo keoewoble oetqles ooJ lowet, valencla (Spaln), 13Lh Lo 17Lh
Aprll, 2009.
[31] u. !. C. !onaLhan k. lel, LCCnCMlC Anu LLLC18lCAL 8LnLll1S Cl PA8MCnlC
8LuuC1lCn ML1PCuS ln CCMML8ClAL lAClLl1lLS," 07/04.
[32] Creenpeace lnLernaLlonal, L8LC, 8enewables 24/7 lnfrasLrucLure needed Lo Save
Lhe CllmaLe," 2009.
[33] World lnsLlLuLe of SusLalnable Lnergy, 8enewable Lnergy oLenLlal AssessmenL
and 8enewable Lnergy AcLlon lan for 8lhar," ueparLmenL for lnLernaLlonal
uevelopmenL (ullu), CovL of uk, une, 2011.
LlecLrlc 8ased LlecLrlcal ower lanL (1 x 10 kW)," MlnlsLry of new and
8enewable Lnergy, CovernmenL of lndla, new uelhl, 2007.
[33] . SLrau, M. Landau, M. vandenbergh, C. 8opp, C. 8rlsa, M. veLLer, C. Clanla and
M. Wollny, Cff-grld power supply and global elecLrlflcaLlon," lv A 1oplcs
2009, pp. 94-101, 2009.

67
r
e
f
e
r
e
n
c
e
s
4
e [ r ] c l u s t e r
f o r a
s m a r t e n e r g y a c c e s s
T H E R O L E O F M I C R O G R I D S I N P R O M O T I N G T H E I N T E G R A T I O N O F R E N E W A B L E E N E R G Y I N I N D I A
Greenpeace is a global organisation that uses non-violent
direct action to tackle the most crucial threats to our
planets biodiversity and environment. Greenpeace is a non-
proft organisation, present in 40 countries across Europe,
the Americas, Africa, Asia and the Pacifc. it speaks for 2,8
million supporters worldwide, and inspires many millions
more to take action every day. To maintain its independence,
Greenpeace does not accept donations from governments
or corporations but relies on contributions from individual
supporters and foundation grants.
Greenpeace has been campaigning against environmental
degradation since 1971 when a small boat of volunteers and
journalists sailed into Amchitka, an area west of Alaska,
where the uS Government was conducting underground
nucleartests.Thistraditionofbearingwitnessinanon-
violent manner continues today, and ships are an important
part of all its campaign work.
Greenpeace india Society
60 Wellington Road, Richmond Town
Bangalore 560025
Tel: +91-80-41154861
Fax: +91-80-41154862
E-mail: info@greenpeaceindia.org
Visit: www.greenpeaceindia.org
front & back cover images SOLAR PHOTOVOLTAiC POWER PLANT iN TANGTSE HARiKRiSHNA KATRAGADDA / GREENPEACE.

Bihar Smart Energy Access

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

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

Bihar Smart Energy Access

Hochgeladen von

Copyright:

Verfügbare Formate

e[r] cluster

The most important step in this direction would be to develop a

I|gure 12: Network w|th upgrades to accommodate |oad scenar|o 4.

I|gure 13: Upgraded gr|d mode|

Das könnte Ihnen auch gefallen