Sie sind auf Seite 1von 204

Green Energy and Technology

Vicente Diaz-Casas Editors

Wind Farms
Green Energy and Technology
More information about this series at
Laura Castro-Santos Vicente Diaz-Casas


Floating Offshore Wind


Laura Castro-Santos Vicente Diaz-Casas
Departamento de Enxeara Naval e Departamento de Enxeara Naval e
Ocenica Ocenica
Universidade da Corua Universidade da Corua
Ferrol, A Corua Ferrol, A Corua
Spain Spain

ISSN 1865-3529 ISSN 1865-3537 (electronic)

Green Energy and Technology
ISBN 978-3-319-27970-1 ISBN 978-3-319-27972-5 (eBook)
DOI 10.1007/978-3-319-27972-5

Library of Congress Control Number: 2016930818

Springer International Publishing Switzerland 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microlms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specic statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by SpringerNature

The registered company is Springer International Publishing AG Switzerland

The average-term energy strategies show the ocean as one of the main sources of
renewable energy. Therefore, reaching new solutions to exploit the ocean energy
has become one of the main objectives of research and development in the eld of
renewable energies. Nowadays, offshore wind farms have come true and shallow
water regions with high wind resources have already been assigned. Thus, the next
step in this eld is to go into deep water regions. It is in these regions where floating
offshore wind platforms should be taken into consideration.
The main objective of this book is to provide an overview about floating offshore
wind farms, whose study is still being developed. In this sense, ten chapters have
been developed. They talk about present and future perspectives of floating offshore
wind farms (Present and Future of Floating Offshore Wind), their life cycle (Life-
Cycle Cost of a Floating Offshore Wind Farm) and economic aspects of their fea-
sibility (Economic Feasibility of Floating Offshore Wind Farms), the types of
floating offshore wind platforms (Floating Offshore Wind Platforms), the CFD
applied to this type of energy (CFD Applied to Floating Offshore Wind Energy),
their mooring and anchoring systems (Mooring and Anchoring), their offshore wind
resource assessment (A Review of Resource Assessment Methods in the Offshore
Wind Energy Sector and A Spatiotemporal Methodology for Deep Offshore
Resource Assessment), the tools taken into account for the ocean energy maritime
spatial planning (Tools for Ocean Energy Maritime Spatial Planning) and, nally,
their operation and maintenance (Operation and Maintenance of Floating Offshore
Wind Turbines).
All this information can be useful for professionals, enterprises and students who
want to improve their knowledge in floating offshore wind, whose development in
the years to come will be high.
The contributors have different professional proles, from academic professors
who are working at different universities to international scientists who are
developing their research at technological centers or national laboratories. Some
of them are experts in wind resource (offshore and onshore), others are focused on
marine structures (platforms, mooring and anchoring), others study maritime and

vi Preface

ocean aspects such as economic issues or spatial planning. But all of them made a
great contribution to this book, which is considered as interdisciplinary in the
floating offshore wind eld.

Ferrol, A Corua, Spain Laura Castro-Santos

Vicente Diaz-Casas

Present and Future of Floating Offshore Wind . . . . . . . . . . . . . . . . . . . 1

Sara Ferreo Gonzlez and Vicente Diaz-Casas
Life-Cycle Cost of a Floating Offshore Wind Farm . . . . . . . . . . . . . . . . 23
Laura Castro-Santos
Economic Feasibility of Floating Offshore Wind Farms. . . . . . . . . . . . . 39
Laura Castro-Santos
Floating Offshore Wind Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
E. Uzunoglu, D. Karmakar and C. Guedes Soares
CFD Applied to Floating Offshore Wind Energy . . . . . . . . . . . . . . . . . 77
M.I. Lamas and C.G. Rodrguez
Mooring and Anchoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Ral Rodrguez Arias, lvaro Rodrguez Ruiz
and Vernica Gonzlez de Lena Alonso
Resource Assessment Methods in the Offshore Wind Energy Sector . . . 121
N. Salvao and C. Guedes Soares
A Spatiotemporal Methodology for Deep Offshore Resource
Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Ana Estanqueiro, Antnio Couto and Luis Rodrigues Jr.
Tools for Ocean Energy Maritime Spatial Planning . . . . . . . . . . . . . . . 161
Paulo Costa, Teresa Simes and Ana Estanqueiro
Operation and Maintenance of Floating Offshore Wind Turbines . . . . . 181
Fernando P. Santos, ngelo P. Teixeira and Carlos Guedes Soares

Editors and Contributors

About the Editors

Laura Castro-Santos obtained her Ph.D. in Industrial Engineering in 2013 from

the University of A Corua (Spain), which has been awarded the research prize
Gonzlez-Llanos in the naval sector. Her nal degree project, about repowering
wind farms, has also been awarded rst prize at the College of Industrial Engineers
of Galicia. She has taken part in several national and international research projects
related to offshore renewable energies (offshore wind, waves and tides in xed or
floating single offshore structures or using hybrid systems). In addition, she has
collaborated with several international institutions such as the National Laboratory
of Energy and Geology (LNEG) and the Centre for Marine Technology and Ocean
Engineering (CENTEC), both in Lisbon (Portugal). She has written lots of papers in
international journals and she has also participated in international congresses
regarding offshore wind, which has allowed her to receive the prize of the best
poster prize in the DeepWind2013 Congress, in Trondheim (Norway). Currently,
she is working as lecturer in the Department of Naval and Oceanic Engineering at
the University of A Corua (Spain), where she does not stop of improving her
teaching skills and research activities in the eld of offshore renewable energies.
Vicente Diaz-Casas is currently Associate Professor of ship design and marine
renewable energies at the Department of Naval and Oceanic Engineering of the
University of A Corua (Spain). He received his Masters in Naval Architecture and
Marine and Offshore Engineering from that university and his Ph.D. in
Mathematical Methods and Numerical Simulation in Engineering and Applied
Sciences through an interuniversity program of University of Santiago de
Compostela, University of Vigo, and University of A Corua. He is a member
of the Committee of Marine Renewable Energy of the Spanish Association of Naval
Architects and member of the executive committee of the Western European
Graduate Education in Marine Technology. He has developed his research activity
in the Integrated Group for Engineering Research as coordinator in marine engi-
neering and offshore technology. He has participated in a high number of

x Editors and Contributors

multidisciplinary and borderline research projects. In his research he has combined

different approaches and knowledge areas with topics such as articial intelligent
(articial neural networks and evolutionary computation), computational fluid
dynamics, mechanical design, and control systems. Now his research focuses are
simulation and design of floating structures for marine renewable energies.

About the Contributors

Ral Rodrguez Arias had been the Head of R&D+i projects at CTC until
September 2015. He is Naval and Oceanic Engineer at the ETSIN (Polytechnic
University of Madrid). He started his professional career as an engineer in
NAVANTIA (the mayor Spanish Shipbuilding Company) in the Structural
Department at the Technical Ofce. For ve years, he was in charge of basic and
detailed structural design in several naval programs (Spanish Frigate F100,
Norwegian Frigate F310, BPE for the Spanish Navy, etc). Besides this work, he
was Shock Manager in the shipyard. In the following three years (20062009), he
worked in the mayor Spanish R&D institution (TECNALIA) as a Senior Researcher
in the wave energy converter project OCEANTEC. Among other responsibilities,
he was in charge of the hydrodynamic, mooring, and structural analysis of the
project. Finally, he was the manager for the design, construction and test program
of the 1:4 scaled prototype, which was tested in real conditions at sea during several
months in the northwest coast of Spain. He is also the Spanish delegate in the
IEC/TC 114 (Marine energyWave, tidal and other water current converters) and
PT 62600-10 (Assessment of mooring system for marine energy converters).
Antnio Manuel Vitoriano Couto obtained his M.Sc. in Meteorology and
Physics Oceanography from the University of Aveiro in 2009. Since 2010, he has
been a research fellow at UAER department of LNEG. He contributed actively in
many European projects (e.g., DemoWFloat, SeaEnergy), national projects (e.g.,
FCT FluctWind), and private consulting. In the FP7 DemoWFloat project, he
contributed operationally in many tasks namely (i) development of methodologies
for deep water wind resource assessment; and (ii) characterization of wave/wind
induced oscillations and their impact on the power performance of the wind turbine.
The main topics of his R&D are wind power forecast based on multivariate sta-
tistical techniques, wind resource assessment, and wind turbine power performance.
He has several technical reports published in the wind energy sector and recently
published articles in journals and conferences proceeding. Antnio is currently
participating in FP7 IRP.Wind and ERANET+/FCT NEWA.
Paulo Alexandre da Silva Costa holds an M.Sc. in Meteorology at the
Department of Physics of the University of Lisbon and is an advanced fellow
researcher at LNEG. His main research activities involve data assimilation tech-
niques and phase error correcting algorithms coupled with atmospheric mesoscale
models to improve wind speed forecast elds up to 48 h. He joined LNEG in 2001
and since then he has been a researcher in wind potential studies mainly focused on
Editors and Contributors xi

offshore and onshore purposes. He is also author of several wind and energy
resource atlases at national and international scales. He has several articles pub-
lished in the wind energy sector and in specialized conference proceedings. He is
also a scientic reviewer of the Journal of Applied Meteorology and Climatology.
Paulo was part of the team of the research projects: ESFRI WindScanner, FP7
DemoWfloat, FP7 Norsewind, IEE 2020 Seanergy and is currently participating in
Vernica Gonzlez de Lena obtained Naval and Oceanic Engineering degree
from the Polytechnic University of Madrid (ETSIN-UPM).
She began her career at PYMAR (small and medium shipyards in conversion)
where she performed the technical and economic monitoring tasks of ships under
construction. In 2010, she began to work in ASTANDER, a repairs and conversions
shipyard. For 2 years her work was focused on technical ofce and R&D+i projects,
involving economic evaluation studies. Since 2012, she has developed her career as
an early stage researcher in the Fundacin Centro Tecnolgico de Componentes
(CTC), in the Marine Renewable Energy sector. The projects performed up to now
include the developing of technical specications for wave energy converters,
marine operations associated to the marine energy converters, naval architecture
calculations, and national and international proposals for R&D as well. During last
year and a half, she participated in several projects focused on mooring and
anchoring systems and stability analysis of floating devices (one wave energy
converter and one floating wind turbine). The development of a methodology to
design specic anchoring systems for renewable marine energies offshore platforms
was the main work.
Ana Isabel Lopes Estanqueiro has a Ph.D. in Mechanical Engineering and a
Masters in Power Engineering. She is a Senior Researcher at the Portuguese gov-
ernment laboratory LNEGNational laboratory for Energy and Geology, I.P. Since
1987, she has been a coordinator of the Energy Analysis and Networks Unit, and
invited Professor at the University of Lisbon. Before, she was Director of the Wind
and Oceans Energy Unit between 2003 and 2009. She is Portuguese representative in
several international organizations, such as IEA wind Implementing Agreement,
TP-Wind Mirror Group, SET-Plan (European Industrial Initiative Wind and
EERA Wind. She is a member of the professional associations: IEEEInstitute
of Electrical and Electronics Engineers and OEOrdem dos Engenheiros. From
2007 to 2009 she was chair of the executive committee of the Implementing
Agreement in Wind Energy of the IEA-Wind. She is an expert in several wind
energy subareas and some of her research activities are in the area of grid integration
of variable renewable generation. More than 25 years of activity and experience in
this sector has led her to coordinate with several research projects (national and
international) in the wind energy area with the most recent: ERANET+/FCT NEWA,
FP7 IRP.Wind, FP7 DemWfloat, ESFRI WindScanner, FP7 Norsewind, IEE 2020
Seanergy, FCT Fluctwind, FCT Roadmap WW, REIVE, T.Urban DEMTEC (ADI).
xii Editors and Contributors

Teresa Maria Nunes Simes Esteves holds a Ph.D. in Energy and Environment
Energy and Sustainable Development from the University of Lisbon. She is an
advanced fellow researcher at LNEG, and has developed her research activity in the
wind energy group since 1998 with main focus on the area of wind resource
assessment and development of GIS methodologies for planning purposes. She is a
member of the COST Action TU 1304 Winer Cost Wind Energy Technology
Reconsideration to enhance the Concept of Smart CitiesWG1. Her main research
activities are: wind resource assessment for wind park installation onshore and
offshore; wind energy production estimates of wind parks in complex terrains; CFD
modeling of the wind in complex and urban environments; database planning and
development in standard and GIS environments; development of methodologies for
the identication of the sustainable wind and solar potential in national and regional
territories using GIS techniques. She was involved in the publishing of the rst
Portuguese databases on wind potential (1999 and 2004) and in the development
of the rst planning. Teresa was part of the team of the research projects: ESFRI
WindScanner; FP7 Norsewind and FCT Roadmap WW, and is currently partici-
pating in ERANET+/FCT NEWA project.
Mara Isabel Lamas Galdo is a Ph.D. in Industrial Engineering. Since 2008 she is
Associate Professor at the University of A Corua. She has written four books and
several papers in scientic journals, most of them related to energy and pollution
reduction. Besides, she has participated in several national and international con-
gresses. She also has professional experience in engineering projects.
Sara Ferreo Gonzlez received her MS Degree in Naval Architecture and
Marine and Offshore Engineering from the University of A Corua in 2010 and in
2012 received a Master of Research in Naval and Industrial Technologies also from
the University of A Corua. Since 2010, she has been working as researcher at the
Integrated Group for Engineering Research at the University of A Corua. Her
current research activities are related to naval and ocean design and maritime safety.
She has been participating in many research projects focused on naval and oceanic
technology (i.e., vessel and oceanic structures dynamical behavior) and renewable
offshore energy (i.e., floating offshore wind energy). She has several publications
related to offshore wind.
Debabrata Karmakar received his Ph.D. degree from the Department of Ocean
Engineering and Naval Architecture, Indian Institute of Technology Kharagpur,
India, in 2009. He has worked as a postdoctoral researcher in the Centre for Marine
Technology and Ocean Engineering (CENTEC), Instituto Superior Tcnico,
University of Lisbon, Portugal in several research projects funded by Portuguese
Foundation for Science and Technology since 2010. His main areas of research are
marine hydrodynamics, hydroelasticity of floating structures and offshore renewable
Luis Carlos Rodrigues Jr. is an MIT Portugal research fellow and a Ph.D. student
at the Faculty of Sciences at Lisbon University (FCUL) in the Doctoral Program of
Editors and Contributors xiii

Sustainable Energy Systems. He received his Electrical Engineering degree in 2009

at the Federal University of Santa Catarina (UFSC), Brazil and the M.Sc. degree in
Energy and Environmental Engineering at FCUL in 2011. He has been collabo-
rating with LNEG since 2009, where he contributed operationally in many
European projects (e.g., DemoWFloat, NORSEWinD), national projects, and pri-
vate consulting. Since 2012, he has been a teaching assistant at FCUL for the
courses, Wind Energy and Energy Networks. His recent research activities are
optimal aggregation of renewable generation for electricity markets participation
and he is currently being a researcher in FP7 IRP.Wind.
lvaro Rodrguez Ruiz is the Head of the Marine Renewable Energy Unit at
Centro Tecnolgico de Componentes (CTC). He received his Industrial
Engineering and Mechanical Engineering degrees and M.Sc. in Automotive
Industry from the University of Cantabria. He has obtained two extraordinary
awards while at his university: extraordinary award in mechanical engineering
and design, construction and test of a racing motorcycle prototype in a worldwide
competition. He has written scientic papers, participated in congresses and is
inventor (EU patent EP20110382092). In June 2014, he obtained a degree in
Industrial Engineering with high scores (three distinctions). One of the three dis-
tinctions he obtained along this degree was in the last subject, in which he
developed an algorithm to study the crack growth. The title of this nal work was
3D Numerical Modelling and Experimental Validation of Fatigue Crack Growth in
R5 Steel in Offshore Mooring Chains used in Moored Offshore Structures. He
began his career at APIAXXI (2006) working on the Renewable Energy Group,
participating in the design of a 250 m2 solar tracker, and calculation of several civil
structures. He started his career at Centro Tecnolgico de Componentes (CTC) in
2007. From 2007 to 2009, he has been working on design and structural assessment
applied for renewable energies. Mechanical design, structural calculations, and fluid
structure interaction have been the main tasks carried out. Since 2009, he has
participated in several projects designing the moorings of one meteorological mast,
and moorings and anchoring of one floating wind turbine and one wave energy
converter. He has acquired skills in project management and in summer 2013 he
was promoted as the Head of the Marine Renewable Energy Unit. Since 2013, he
has been leading several proposals of H2020 in the Low Carbon Energies call
(LC1-LC2) and he is managing two big proposals to be submitted in LCE3 call.
At the same time, lvaro has managed some FP7 projects such as ACORN,
IRPWIND, MAREWINT. He is an active participant in the most important associa-
tions regarding wind energy and wave energy, such as EERA, TP OCEAN, etc.
Recently, he has been positioned as expert in the IEC TC114 (AEN/CTN 206/SC 114).
Ndia Carina Pereira Salvao is a Research Assistant at the Centre for Marine
Technology and Ocean Engineering (CENTEC) at Instituto Superior Tcnico,
Lisboa, Portugal. She graduated in Meteorology, Oceanography, and Geophysics
from the Faculty of Sciences, University of Lisbon, where she also got her M.Sc.
degree in Geophysical Sciences. She has participated in several projects related to
xiv Editors and Contributors

the offshore renewable energies in an international context, and has attended some
congresses on the subject. Her research resulted in numerous papers presented at
international conferences or scientic journals. She is now pursuing her Ph.D. in
Naval Architecture and Marine Engineering and conducting studies focusing on the
implementation of regional atmospheric models in areas of offshore industrial
Fernando Pinto dos Santos received his degree and the M.Sc. degree in
Mechanical Engineering from Instituto Superior Tcnico (IST), University of
Lisbon, Lisbon, Portugal and from the Faculty of Science and Technology,
Universidade Nova de Lisboa, Lisbon, Portugal, in 2002 and 2012, respectively. He
also completed an Advanced Training Diploma in Risk Assessment, Safety and
Reliability at IST in 2007. He is a Research Assistant in the Centre for Marine
Technology and Ocean Engineering (CENTEC) of IST and is currently pursuing
his doctoral studies on modeling and optimization of offshore wind systems reli-
ability and maintenance.
Carlos Guedes Soares received the MS and Ocean Engineering degrees from the
Massachusetts Institute of Technology, Cambridge, MA, USA, in 1976, the Ph.D.
degree from the Norwegian Institute of Technology, Trondheim, Norway, in 1984,
and the Doctor of Science degree from the Technical University of Lisbon, Lisbon,
Portugal, in 1991. He is a Professor of Naval Architecture and Marine Engineering
and President of the Centre for Marine Technology and Ocean Engineering
(CENTEC), a research center of the University of Lisbon, Lisbon, Portugal, which
is evaluated as Excellent and funded by the Portuguese Foundation for Science and
Technology. Renewable Energies Offshore is one of the topics of the Strategic
Research Plan of CENTEC.
ngelo Palos Teixeira is Assistant Professor at Instituto Superior Tcnico (IST),
University of Lisbon, since 2004 and Principal Investigator of the research group on
Safety and Logistics of Maritime Transportation of the Centre for Marine
Technology and Ocean Engineering (CENTEC). He has received his Ph.D. degree
in Naval Architecture and Marine Engineering from the Technical University of
Lisbon in 2007 and the BS and MS degrees from the Technical University of
Lisbon in 1994 and by the Faculty of Engineering of Glasgow University in 1998,
respectively. His main areas of research are risk assessment, maritime safety,
structural reliability analysis of marine structures, and systems reliability and
maintenance. He has worked also with different human factor aspects and has
participated in the development of a common methodology for the investigation of
maritime accidents that improved the understanding of human elements as related to
Emre Uzunoglu obtained his B.Sc. from Yildiz Technical University in Istanbul,
Turkey in 2008, and M.Sc. from Instituto Superior Tcnico in Lisbon, Portugal in
2011, in Naval Architecture and Marine Engineering. His education includes an
exchange year in Universit degli Studi di Trieste, Italy. He joined Centre for
Editors and Contributors xv

Marine Technology and Ocean Engineering (CENTEC) in 2009 as a part of the

Marine Dynamics and Hydrodynamics Group. He authored articles in journals and
conferences and chapters in books. His areas of interest include design and per-
formance evaluation of offshore platforms, hydrodynamics of ships and moored
floaters, and renewable energies.
Carlos G. Rodrguez Vidal is a Master in Maritime Engineering, a marine
engineer, and a naval technical engineer. He has almost 20 years of professional
experience. Besides, he has participated as researcher in several projects at the
University of Corua and CIS Galicia related to energies and mechanical design. He
has authored several books and papers in scientic journals and participated in


Ral Rodrguez Arias Centro Tecnolgico de Componentes, Santander, Spain

Laura Castro-Santos Departamento de Enxeara Naval e Ocenica, Universidade
da Corua, Ferrol, A Corua, Spain
Paulo Luis Costa LNEGNational Laboratory for Energy and Geology, Lisbon,
Antnio Couto LNEGNational Laboratory for Energy and Geology, Lisbon,
Vernica Gonzlez de Lena Alonso Centro Tecnolgico de Componentes,
Santander, Spain
Vicente Diaz-Casas Departamento de Enxeara Naval e Ocenica, Universidade
da Corua, Ferrol, A Corua, Spain
Ana Estanqueiro LNEGNational Laboratory for Energy and Geology, Lisbon,
Sara Ferreo Gonzlez Integrated Group for Engineering Research, University of
A Corua, Edicio de Talleres Tecnolgicos, Ferrol, A Corua, Spain
C. Guedes Soares Centre for Marine Technology and Ocean Engineering
(CENTEC), Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
D. Karmakar Centre for Marine Technology and Ocean Engineering (CENTEC),
Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
M.I. Lamas Departamento de Enxeara Naval e Ocenica, Universidade da
Corua, Ferrol, A Corua, Spain
Luis Rodrigues Jr. LNEGNational Laboratory for Energy and Geology,
Lisbon, Portugal
xvi Editors and Contributors

C.G. Rodrguez Departamento de Enxeara Naval e Ocenica, Universidade da

Corua, Ferrol, A Corua, Spain
lvaro Rodrguez Ruiz Centro Tecnolgico de Componentes, Santander, Spain
N. Salvao Centre for Marine Technology and Ocean Engineering (CENTEC),
Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
Fernando P. Santos Centre for Marine Technology and Ocean Engineering
(CENTEC), Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
Teresa Simes LNEGNational Laboratory for Energy and Geology, Lisbon,
ngelo P. Teixeira Centre for Marine Technology and Ocean Engineering
(CENTEC), Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
E. Uzunoglu Centre for Marine Technology and Ocean Engineering (CENTEC),
Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
Present and Future of Floating
Offshore Wind

Sara Ferreo Gonzlez and Vicente Diaz-Casas

Abstract Europe, USA and Japan are positioning for offshore wind renewable
energy becomes an option in order to reduce their energy dependence on fossil fuels.
This is especially a key factor in countries that does not have its own fossil fuel
resources. Given the characteristics of our continental shelf, it is expected that these
platforms for harnessing the wind energy will be installed in locations away from the
coast (offshore) and in 50+ deeper waters. This will lead to a new scenario that
request new concepts and engineering solutions based on become floating structures
(moored and anchored to the seabed). These new requirements and the economical
constrains of this type of energies has made that nowadays we are in a stage of proof
of concepts without a clear candidate to become the dominant technology in the near
future. Therefore, in this chapter an overview of present and future trends in floating
offshore wind technology is presented. The objective of this overview is to structure
the different concepts and clarify the advantages and disadvantages of each.

Keywords Offshore wind  Floating wind  Offshore mooring  Wind platform

1 Introduction

Nowadays the European energy framework is constrained by the mandatory reg-

ulations on Climatic Change, like the Kyoto Protocol, and by the dependence from
foreign fossil fuels producers. In order to reduce the emissions of greenhouse gases
and increase its independence from fossil fuels producers, European Union
(EU) established in 2009 that 20 % of nal energy consumption should be from
renewable sources in 2020.

S.F. Gonzlez (&)

Integrated Group for Engineering Research, University of a Corua, Edicio de Talleres
Tecnolgicos, Campus de Esteiro, 15403 Ferrol, A Corua, Spain
V. Diaz-Casas
Departamento de Enxeara Naval e Ocenica, Universidade da Corua, C/Mendizbal, s/n,
15403 Ferrol, A Corua, Spain

Springer International Publishing Switzerland 2016 1

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_1
2 S.F. Gonzlez and V. Diaz-Casas

Due to the reduced available locations and the environmental constrains for
inland renewable energies, marine renewable energies will increase its presence in
the European energy framework. In this sense, offshore wind will make a substantial
contribution to meeting the EUs energy policy objectives through a very signicant
increasein the order of 3040 times by 2020 and 100 times by 2030in installed
capacity compared to today [1].
Therefore, offshore wind farms (see wind turbine in Fig. 1) are becoming one of
the most signicant sources of renewable energy. The principal world develop-
ments in this eld focuses on wind turbines with seabed foundations (transferring
the technologies used onshore, to the sea). The principal technological constrains
for current designs are the distance to shore and depth. Fixed designs are restricted
to operate at 5060 m depth, however much of the regions with the highest
available energy request to operate deeper.
For many countries, xed offshore wind is not an option because the continental
shelf drops away suddenly and steeply making installation of xed structures much
more difcult than in the waters of northern Europe.
In places where the seabed geography means coastal waters are relatively
shallow, the use of conventional techniques to install offshore wind farms are
commonplace. Foundations, either monopile or jacket, are embedded directly into
the sea floor. Going into deeper water means pushing foundation technology to its
very limits as xed structures are only suitable for water up to 50 metres deep.
Thus, the next step is developing floating structures that can operate in deep
Two great examples of how far floating wind has come in recent years are
Statoils Hywind Demonstrator project in Norway and Principle Powers
WindFloat in Portugal.
Hywind is designed around a spar anchored in 200 m of water and is testing how
wind and waves affect the floating structure, ahead of a small pilot project. It has
already produced some enlightening results about how the design could work
commercially and has been instrumental in opening the door for the advancement of
this novel technology.

Fig. 1 Wind turbine

Present and Future of Floating Offshore Wind 3

WindFloat uses a catenary mooring system, similar to what you see in many oil
and gas designs, and can be fully assembled onshore and towed to its nal desti-
nation. It has been generating power since 2011 and a number of projects based on
this design are in development the world over with much larger turbines than the
2 MW prototype.
It might seem counter intuitive to have lots of different options for how floating
wind could work, but as it is still a relatively infant technology and there are lots of
ideas coming into the market, the competition helps to drive the technology to its
greatest potential to realise maximum power and efciency at the best price.

2 Offshore Wind Market

As of the end of 2014, worldwide, total cumulative installed capacity from wind
power amounts to 369,553 MW and increased by 16 % compared to the previous
year (318,106 MW). In 2014, world leader China installed almost half of the
worlds added wind power capacity. Global wind power installations increased by
35,467 megawatts (MW) and 51,447 MW in 2013 and 2014, respectively (Fig. 2)
[2, 3].
At the end of 2014 a total of 8.05 GW of installed capacity in offshore wind
energy had been installed in 74 parks across 11 European countries [4]. These
facilities are mainly in the North Sea, with the UK, Denmark and the Netherlands,
countries with higher installed power [5]. It is expected to reach in 2050 between 20
and 55 GW of offshore wind power installed [6].
Nowadays, offshore wind farms are concentrated in shallow areas (up to 60 m)
and up to 20 km from the coast, being xed technologies [4], of which:
78.8 % are monopiles
10.4 % gravity structures
4.7 % jackets,
4.1 % tripods and
1.9 % tripiles

Global Wind Power Cumulative Capacity.

1997-2014 (MW)

Fig. 2 Worldwide installed capacity (19972014) (GWEC, 2015)

4 S.F. Gonzlez and V. Diaz-Casas

However, most of the future offshore wind farms will present greater capacity
and will move into deeper waters and far from the coast, so that other technology
adapted to greater depths is required.
This is the main reason why research in the coming years will be focused on
offshore energy generation with floating systems.

3 Offshore Wind Fixed Foundations

With the exception of two turbines, Europes grid connected offshore wind turbines
rely on xed foundations, and the majority of those on monopile foundations.
Gravity based substructures are the second most common foundation type, followed
by space frame structures.
Typical congurations of xed offshore foundations are showed in Fig. 3:
1. Monopile: Monopile foundation is a structure made of a cylindrical steel tube. Is a
relative simple design by which the tower is supported by the monopile (directly
or through a transition piece). The monopile continues down into the seabed.
The pile penetration depth is adjustable to suit the actual environment and
seabed conditions.
2. Gravity base foundation: GBF are normally concrete based structures, which
can be constructed with steel or concrete. The ballast required to anchor the
foundation consist on sand, iron ore or rock lled into the base of the structure
with adjustments in the designed base width to suit the soil conditions.
3. Tripod: The tripod structure is a relatively lightweight steel jacket (with three
legs) compared to a standard lattice structure. Under the steel central column,
which is below the turbine, there is a steel frame which transfers the forces from
the tower into the three steel piles. These piles are installed at each leg position
to anchor the tripod to the seabed.
4. Jacket: Jacketor latticestructures typically consists of corner piles inter-
connected with bracings with diameters up to 2 m. The soil piles are driven
inside the pile sleeves to the required depth to gain stability for the structure.

Fig. 3 Typical xed offshore foundations

Present and Future of Floating Offshore Wind 5

5. Tripile: Tripile structures is a three-legged jacket structure in the lower section,

connected to a monopole in the upper part of the water column, all made of
cylindrical steel tubes. The base width and the pile penetration depth can be
adjusted to suit the sites geological conditions.
Monopiles are used in shallow depthsup to 20 mdue to their simplicity and
minimal design developments required. An alternative to the monopile is the
gravity base foundation, These foundations can overcome the flexibility issues of
monopiles but will grow in cost very rapidly with water depth as well, although the
use of concrete may provide some advantage in extending favorable economics.
Moreover, it can be installed in depths between 0 and 30 m.
Monopiles and GBF are also used in shallow waters.
Transitional substructure technology can be deployed up to depths 60 m.
Tripods, jackets or tripiles are included in this category (transitional water
At some water depth (60 m or more), floating substructures may be the best
option. A floating structure must provide enough buoyancy to support the weight of
the turbine and to restrain pitch, roll, and heave motions within acceptable limits
(Fig. 4).
It is expected that xed-bottom offshore wind farms will continue to dominate
up to 2030, but the next 510 years will be an important development period for
floating technology, with more prototype demonstrations and pilot arrays to prepare
the technology for commercial projects from 20202025.

Fig. 4 Description of shallow, transitional and deep water technology

6 S.F. Gonzlez and V. Diaz-Casas

4 Offshore Wind Floating Foundations

Floating foundation typologies for floating wind turbines, and its associated com-
ponent technologies (mooring, anchors), are based on designs used in oil and gas
industry. Having initially exploited oil wells in shallow waters using xed bottom
structures, the identication of signicant oil reserves further offshore in deep
waters created the need for floating structures. This need was met with the devel-
opment of spar, semisubmersible, and TLP structures, which could access
deep-water locations and extract signicant volumes of oil. However, while the
typologies are similar, the structures themselves are very different and must meet
different needs. For example whereas oil and gas industry requires bigger but fewer
structures, offshore wind will require installation of a large number of smaller
structures, which impacts greatly on the design, fabrication, installation, and
operational characteristics of the structures.

4.1 Typologies of Floating Platforms for Offshore Wind

The concept of a floating wind turbine has existed since the early 1970s, but the
industry only started researching it in the mid-1990s.
In 2008, Blue H technologies installed the rst test floating wind turbine off the
Italian coast. The turbine had a rated capacity of 80 kW and after a year of testing
and data collection it was decommissioned.
A year later the Poseidon 37 project followed, a 37 m wide wave energy plant
and floating wind turbine foundation tested at DONGs offshore wind farm at
In 2009, Statoil installed the worlds rst large scale grid connected floating
wind turbine, Hywind, in Norway, with a 2.3 MW turbine.
The second large scale floating system, WindFloat, developed by Principle
Power in partnership with EDP and Repsol, was installed off the Portuguese coast
in 2011. Equipped with a 2 MW Vestas wind turbine, the installation started
producing energy in 2012.
2011 was the best year on record for deep offshore development with two
floating substructures tested, SeaTwirl and SWAY, in addition to the grid connected
Windfloat project.
There are numerous congurations of floating platforms to support wind turbines
at sea. The conguration of the floating support platform will contribute greatly to
achieve platform-tower-turbine system stability. Because of that, a classication
system that divides all platforms into three general categories has been followed,
based on the physical principle or strategy that is used to achieve static stability
(Fig. 5) [7, 8]:
Semi-submersible platformBuoyancy stabilized (Platform 1 in Fig. 5): plat-
form which floats semi-submerged on the surface of the ocean whilst anchored
Present and Future of Floating Offshore Wind 7

Fig. 5 Floating wind turbine concepts

to the seabed with catenary mooring lines. Those platforms that achieve stability
through the use of distributed buoyancy, taking advantage of weighted water
plane area for righting moment. Often requires a large and heavy structure to
maintain stability, but a low draft allows for more flexible application and
simpler installation. Examples: WindFloat (by Principle Power); Damping Pool
(by IDEOL); SeaReed (by DCNS).
Spar buoyBallast stabilized (Platform 2 in Fig. 5): A cylindrical ballast-
stabilised structure which gains its stability from having the centre of gravity lower
in the water than the centre of buoyancy. Those platforms achieve stability by
using ballast weights in the low part of the floater which creates a righting moment
and high inertial resistance to pitch and roll, and usually enough draft to offset
heave motion. The simple structure of the spar-buoy is typically fairy easy to
fabricate and provides good stability, but the large draft requirement can create
logistical challenges during assembly, transportation, and installation, and can
constrain deployment to waters >100 m depth. Examples: Hywind (by Statoil);
Sway (by Sway); Advanced Spar (by Japan Marine United).
Tension leg platform (TLP)Mooring line stabilized (Platform 3 in Fig. 5):
Semi-submerged buoyant structures, anchored to the seabed that achieve sta-
bility through the use of mooring line tension. The shallow draft and tension
stability allows for a smaller and lighter structure, but this design increases
stresses on the tendon and anchor system. There are also challenges with the
installation process and increased operational risks if a tendon fails. The TLP
relies on mooring line tension for righting stability. Examples: PelaStar (by
Glosten); Blue H TLP (by Blue H Group); Eco TLP (by DBD Systems);
8 S.F. Gonzlez and V. Diaz-Casas

Table 1 Advantages and disadvantages of three types of floating wind platforms

Type Advantages Disadvantages
Spar buoyballast Simple design is amenable to Constrained to deep water
stabilized serial fabrication processes locations
Few moving parts (no active Offshore turbine assembly
ballast required) requires dynamic positioning
Excellent stability vessels and
Heavy-lift cranes
Large draft limits ability to tow
the structure back to port for
major repairs
Tension leg platform Low structural mass High loads on the mooring and
(TLP)mooring line Onshore turbine assembly anchoring system
stabilized Few moving parts (no active Challenging installation process
ballast required) Bespoke installation barge often
Excellent stability required
Semi-submersible Flexible application due to the High structural mass to provide
platformbuoyancy ability to operate in shallow sufcient buoyancy and stability
stabilized water depths Complex steel structures with
Low vessel requirementonly many welded joints can be
basic tug boats required difcult to fabricate
Onshore turbine assembly Potentially costly active ballast
Amenable to port-side major systems

Table 1 shows principal advantages and disadvantages of the three types of

floating support platforms described above [5]:
In practice, all floating concepts are actually hybrid designs that gain static
stability from all three methods, although generally relying on one primary source
for stability.

4.2 Mooring and Anchoring Systems

We can classify the possible moorings for floating platforms used in offshore wind
in three main types:
Spread mooring: In a typical spread mooring system, groups of mooring lines
are nished at the corners of the platform, holding a stable platform heading.
Some examples are:
Catenary Mooring
Multi-Catenary Mooring
Taut Spread Mooring
Present and Future of Floating Offshore Wind 9

Single point mooring. Single point moorings are used primarily for ship shaped
platforms. They allow the platform to weather vane. There is a wide variety in
the design of single point moorings:
Turret Mooring
Catenary Anchor Leg Mooring (CALM)
Single Anchor Leg Mooring (SALM)
Articulated Leg Column
Single Point Mooring and ReservoirSpar
Fixed Tower Mooring
Dynamic positioning (DP): The last type of station-keeping system is DP. DP is
a technique for automatically maintaining the position of a floating plat-form,
within a specied tolerance, by controlling on board thrusters.
Active Mooring
In practice not all the systems that have been presented have been incorporated
into real platforms, for example, the use of DP is not cost effective for use in these
The most common mooring congurations are taut spread mooring systems
(which are used in TLPs) and catenary mooring systems (which are used in spar
buoys and semi-submersibles platforms). Some concepts will also adopt a semi-taut
mooring system. [6, 9].
Taut spread mooring: Made by synthetics bres or wire, which use the buoy-
ancy of the floater and rm anchor to the seabed to maintain high tension for
floater stability.
Catenary mooring: Steel chains and/or wires whose weight and curved shape
holds the floating platform in place. Lower section of mooring chain rests on the
seafloor, supporting the anchor and acting as a counterweight in stormy
Semi-taut mooring: Synthetic bres or wires usually incorporated with a turret
system, where a single point on the floater is connected to a turret with several
semi-taut mooring lines connecting to the seabed.
Principal characteristics of these three systems are given in Table 2 [6].
There are a number of anchoring solutions available, depending on the mooring
conguration, seabed conditions, and holding capacity required. Catenary mooring
congurations will often use drag-embedded anchors to handle the horizontal
loading, though piled and gravity anchors are still applicable, while taut-leg
moorings will typically use either drive piles, suction piles, or gravity anchors to
cope with the large vertical loads placed on the mooring and anchoring system. The
size of the anchor is also variable, with larger and heavier anchors able to generate a
greater holding capacity.
Ultimately, anchor choice will be project and site specic, often dictated by the
seabed conditions.
10 S.F. Gonzlez and V. Diaz-Casas

Table 2 Principal characteristics of taut spread, catenary and semi-taut moorings

Type Characteristics
Taut spread Small footprint
mooring Vertical loading at anchoring point
Large loads placed on the anchorsrequires anchors which can withstand
large vertical forces
Very limited horizontal movement
High tension limits floater motion (pitch/roll/heave) to maintain excellent
Challenging installation procedure
Minimal disruption to the seabed (small footprint)
Catenary Large footprint
mooring Horizontal loading at anchoring point
Long mooring lines, partly resting on the seabed, reduce loads on the
Some degree of horizontal movement
Weight of mooring lines limits floater motion, but greater freedom of
movement than taut-leg
Relatively simple installation procedure
Lower section of chain rests on the seabed, resulting in more disruption
(large footprint)
Semi-taut Medium footprint
mooring Loading typically at *45 to anchoring point
Medium loads on the anchors
Limited horizontal movement, but full structure can swivel around the
turret connection
Single connection point makes the platform susceptible to wave induced
Relatively simple installation procedure
Low level of disruption (medium footprint)

Higher holding capacities are usually generated in sands and hard clays than in
soft clays, although where penetration is difcult in rm soils, gravity base or piled
solutions might be required. A summary of the main anchor types is detailed below,
but there is great variety even within these typologies. All are proven concepts
which have been used extensively in the marine and oil & gas industries.
In Fig. 6, we can see principal anchors employed in floating offshore wind:
1. Drag anchor.
2. Driven pile.
3. Suction pile.
4. Gravity anchor.
The type of anchor to be used for floating offshore wind platforms depends on the
directional variation of the forces acting on the anchor and on type of soil of the sea
bottom. Drag embedment anchors only can be used on soft beds (gravel, sand, silt or
clay) and when forces do not change too much in direction. Gravity anchors can be
deployed in all type of beds and their performance is nearly independent on the force
angle. Finally, pile anchors can be very expensive in rock (in terms relative to the cost of
Present and Future of Floating Offshore Wind 11

Fig. 6 Anchors employed in

floating offshore wind

platforms) at the water depths where those devices are deployed while gravity plunge
piles or suction piles could be affordable in soft bottoms. Otherwise, pile anchor per-
formance can be made nearly independent on the changes on the force action angle.
Table 3 shown principal characteristics of tree typologies of anchors [6].

4.3 Existing Floating Wind Concepts

There are a number of floating wind concepts currently under development, but
there is no clear winner with regard to which is most likely to be deployed at scale

Table 3 Characteristics of three principal typologies of anchors

Type of anchor Characteristics
Drag anchor Best suited to cohesive sediments, though not too stiff to impede penetration
Horizontal loading
Simple installation process
Recoverable during decommissioning
Driven pile Applicable in a wide range of seabed conditions
Vertical or horizontal loading
Noise impact during installation (requires hammer piling)
Difcult to remove upon decommissioning
Suction pile Application constrained by appropriate seabed conditionsnot suitable in
loose sandy soils or stiff soils where penetration is difcult
Vertical or horizontal loading
Relatively simple installation, less invasive than other methods
Easy removal during decommissioning
Gravity anchor Requires medium to hard soil conditions
Usually vertical loading, but horizontal also applicable
Large size and weight can increase installation costs
Difcult to remove upon decommissioning
12 S.F. Gonzlez and V. Diaz-Casas

in the future. In all likelihood there will not be a single winning concept, but a range
of leading devices suitable for different site conditions, and influenced by local
infrastructure and supply chain capabilities. This section contains an overview of
the various concepts currently under development which are vying for market
leader position in the race to commercialise their designs (Tables 4 and 5).
This floating technology is in an early stage of its development. Almost all
existing floating structures are in design phase, with the following exceptions,
different prototypes that are testing in the marine environment:
Hywind (Table 6): Consists of a slender, ballast-stabilised cylinder structure. The
spar-type floater has a low water plane area that minimises wave induced loading,
and a simple structure that minimises production cost. It can be used with any

Table 4 Existing floating wind concepts

Type Name Developer Typology
Buoyancy stabilized WindFloat Principle Power Semi
platforms Damping Pool IDEOL Barge
VERTIWIND Technip/Nenuphar Semi
SeaReed DCNS Semi
Tri-Floater GustoMSC Semi
Nautilus Semi-Sub Nautilus Floating Semi
Nezzy SCD Aerodyn Engineering Semi
TetraFloat TetraFloat Ltd. Semi
VolturnUS DeepCWind Semi
Compact Semi-Sub Mitsui Engineering Semi
V-Shape Semi-Sub Mitsubishi Heavy Semi
Ballast stabilized platforms Hywind Statoil Spar
Sparbuoys Sway Sway A/S Spar
WindCrete U.P. Catalunya Spar
Hybrid Spar Toda Construction Spar
Advanced Spar Japan Marine United Spar
SeaTwirl SeaTwirl Engineering Spar
DeepWind Spar DeepWind Consortium Spar
Mooring line stabilized PelaStar Glosten Associates TLP
platforms Blue H TLP Blue H Group TLP
Tension leg platforms
Eco TLP DBD Systems TLP
TLPWind Iberdrola TLP
AFT Nautica TLP
Present and Future of Floating Offshore Wind 13

Table 5 Existing floating wind concepts

Name Depth range Full-scale prototype Pre-commercial array
WindFloat 40100 m 2011 2018
Damping Pool 35 m+ 2015
VERTIWIND 50 m+ 2016 2018
SeaReed 50200 m 2018 2020
Tri-Floater 50300 m
Nautilus Semi-Sub 60150 m
Nezzy SCD 35200 m
TetraFloat 30200 m
VolturnUS 2018
Compact Semi-Sub 2013
V-Shape Semi-Sub 2015
Hywind 100500 m 2009 2017
Sway 55300 m
WindCrete 1501000 m
Hybrid Spar Concrete-Steel 100 m+ 2013
Advanced Spar 80 m+ 2013
DeepWind Spar
PelaStar 70200 m
Blue H TLP 50250 m 2018 2020
GICON-SOF 40250 m 2015 2017
Eco TLP 100 m+ 2018
AFT 40700 m

Table 6 Spar buoy of Statoil

Company Statoil
Manufacturer Siemens
Type of floater Spar-buoy
Turbine capacity 2.3 MW (prototype)
37 MW (commercial)
Prototype installed 2009, Norway
Commercial 20162017
Origin Norway
14 S.F. Gonzlez and V. Diaz-Casas

qualied offshore wind turbine. The mooring system consists of three mooring
lines connected to the hull by means of bridles that prevent excessive rotation
about the vertical axis (yaw motion). The mooring system has inherent design
redundancy, with adequate reserve strength in case of a mooring line failure.
The 2.3 MW Hywind demo was installed in Norway in 2009the worlds rst
full scale floating offshore wind turbine. The unit is located at a water depth of
200 m, 10 km off Norways west coast. It has been thoroughly inspected after the
rst and second years in service, and no signs of deterioration, damage, or wear
connected to being on a floater have been reported. Statoil now considers the design
to be technically veried. The floater design has been optimised and up-scaled for
deployment with multi-MW turbines in the 37 MW range. The next step will be to
test the design in a pilot farm with ve units (actually, in fabrication phase in
Spanish shipyards).
WindFloat (Table 7): The worlds second full-scale floating wind turbine is the
WindFloat, operating at rated capacity (2 MW) approximately 5 km offshore of
Pvoa de Varzim, Portugal since 2012. The WindFloat design consists of a
semi-submersible floater tted with patented water entrapment (heave) plates at
the base of each column. The plate improves the motion performance of the
system signicantly due to damping and entrained water effects. This stability
performance allows for the use of existing commercial wind turbine technology.
In addition, WindFloats closed loop hull trim system mitigates average wind
induced thrust forces. This secondary system ensures optimal energy conversion
efciency following changes in wind velocity and directions.

Table 7 Semisubmersible platform of Principle Power

Company Principle Power, EDP, Repsol
Manufacturer Vestas
Type of floater Semi-submersible
Turbine capacity 2 MW (prototype)
Prototype installed 57 MW (commercial)
Commercial 2011, Portugal
Origin 2017
Present and Future of Floating Offshore Wind 15

The mooring system employs conventional components such as chain and

polyester lines to minimise cost and complexity. Through the use of pre-laid drag
embedded anchors, site preparation and impact is minimised.
In 2011, WindFloat was installed, equipped with a 2 MW Vestas wind turbine,
the installation started producing energy in 2012. The next step will be to build a
27 MW array off Portugal, with the support of the NER300 funding. Another
30 MW demonstration project is also planned off Oregon in the Pacic Ocean.
Rest of developments are in design phase, more of them are ballast stabilized
Damping PoolIDEOL: The IDEOL platform is a square concrete hull with a
central opening to create a patented damping pool system that uses the
entrapped water to minimise floater motions, resulting in strong hydrodynamic
performance. The design can also be constructed using steel, but delivers greater
cost reduction and higher local content when using concrete.
IDEOL have an initial 2 MW demonstration slated for installation at the
SEMREV test site in 2015, part funded by the European Commissions
FLOATGEN and French Government (ADEME) OCEAGEN initiatives.
IDEOL have also secured a partnership with Hitachi Zosen to deploy the
IDEOL concept in Japan.
VERTIWINDTechnip/Nenuphar: A floating wind structure which the
designers claim can reduce the aerodynamic loads on the structure, with the low
centre of gravity improving stability and minimising the gyroscopic effects.
An onshore prototype of the vertical axis wind turbine is currently in operation
at Fos-sur-Mer, and a 2 MW unit is expected to be deployed in the
Mediterranean in 2016, with support from the European Commission through
the INFLOW project. Following this initial pilot, there are plans for a 13
turbine 34 MW wind farm in 2018.
SeaReedDCNS: The concept is a semi-submersible composed of a central
cylindrical column, which supports the turbine, plus 3 external cylindrical
columns linked to the central column by pontoons.
A pilot demonstration is expected to be installed at Le Croisic, Brittany, in 2018
before scaling up to pre-commercial and commercial scale from 2020.
Tri-FloaterGustoMSC: The GustoMSC Tri-Floater is a 3 column braceless
semi-submersible unit.
The concept has undergone advanced simulations and model testing to verify its
performance and further optimise the design.
SPINFLOATEOLFI/GustoMSC: SPINFLOAT is a variation on the
GustoMSC Tri-Floater, adopting a vertical axis turbine on the floater platform
instead of a conventional horizontal axis turbine. The vertical axis design is
being developed by EOLFI, working closely with GustoMSC to optimise an
integrated solution.
Nautilus Semi-SubNautilus Floating Solutions: The NAUTILUS technology
consists of a four column ring pontoon semi-submersible unit with heave plates
and a catenary mooring system. The wind turbine is located centrally relative to
16 S.F. Gonzlez and V. Diaz-Casas

the columns which provide buoyancy to support the turbine and allow sufcient
water plane inertia to maintain stability. The horizontal plates at the bottom and
in between the columns increase the added mass, thus shifting the natural period
away from the wave motions and increasing the viscous damping in roll, pitch,
and heave. The water ballast, which is located inside the bottom of the columns,
acts as a static ballast to lower the platform to its target operational draft. An
active ballast system is then used to compensate changes in wind speed and
directions, taking water into and off each column to compensate for the wind
loading on the turbine.
Nezzy SCDAerodyn Engineering: The floater consists of a concrete y-shaped
floating unit with a cylindrical plastic- composite buoy at the end of each leg to
provide added buoyancy. A semi-taut turret mooring system at the end of the
longest leg which can rotate to align the turbine with the wind flow. The tower is
angled at 10 downwind and the drivetrain is mechanically adapted to accel-
eration forces caused by floating wind movements.
TetraFloatTetraFloat Ltd.: TetraFloat is a floating platform comprising
structural members on the edges of a tetrahedron. The concept has been basin
tested with scaled prototypes.
VolturnUSDeepCWind Consortium: The VolturnUS is a semi-submersible
platform developed at the University of Maines Advanced Structures and
Composites Centre through the DeepCWind Consortium. A 1:8 scale 0.02 MW
prototype was installed in the Gulf of Maine in 2013 and there are plans for a
full- scale demonstration with two 6 MW turbines deployed off the coast of
Compact Semi-SubMitsui Engineering and Shipbuilding: Four column
semi-submersible platform using a standard catenary mooring conguration.
The structure was installed off the coast of Fukushima, Japan, in 2013 sup-
porting a 2 MW Hitachi turbine.
V-Shape Semi-SubMitsubishi Heavy Industries: V-shape semi-submersible
platform with the turbine located on one of three columns which provide
additional buoyancy for the structure. A rst full-scale demonstration is slated
for installation in 2015 supporting the 7 MW MHI SeaAngel turbine, which will
make it the worlds largest floating wind device.
There are also some ballast stabilized platforms/spar designs:
SwaySway A/S: Sway consists of a continuous tower/foundation which is
anchored to the seabed using a tension- torsion leg which is equipped with a
passive subsea yaw swivel. A 0.15 MW prototype was installed off the coast of
Norway in 2011, and Sway is now seeking support for a full-scale
WindCreteUniversitat Politcnica de Catalunya: The WindCrete solution is a
conventional spar-buoy floater designed for application in deep waters >150 m.
The spar consists of a monolithic concrete design, including both the tower and
the floater, to reduce steel content. To avoid cracking and fatigue in the marine
Present and Future of Floating Offshore Wind 17

environment, the structure is completely post-tensioned with steel tendons in a

longitudinal direction.
Hybrid Spar Concrete-SteelToda Construction: Conventional spar-buoy with
a slender body and deep draft, but a hybrid design in which the spar is split into
a steel top section and concrete lower section. After an initial 100 kW prototype,
a full-scale 2 MW unit was installed off Kabashima Island, Japan, in 2013, and
has so far survived the severe typhoons which hit southern Japan.
Advanced SparJapan Marine United: Alternative spar design with a shorter
body to allow for more flexible application in shallower waters, to minimise
sway and heave. The concept was chosen to support the worlds rst floating
substation off the coast of Fukushima, installed in 2013, and it will support a
5 MW Hitachi turbine at the Fukushima site, expected in 2016.
SeaTwirlSeaTwirl Engineering: SeaTwirl is a novel spar design that uses a
vertical axis wind turbine to capture the wind energy and a torus ring stabilise
the energy output. A 1.5 kW prototype was installed in Swedish waters in 2014,
with plans to scale up to 30 kW in 2015, before multi-megawatt demonstrations
from 2017.
DeepWind SparDeepWind Consortium: The DeepWind spar consist of a long
vertical tube that rotates in the water using a vertical axis rotor and bottom-based
generator. The structure is anchored to the seabed using a catenary mooring
And nally, mooring lines stabilized platforms/TLP designs:
PelaStarGlosten Associates: The Glosten PelaStar is a 5-arm submerged hull
with a single central column, anchored to seabed with one cable tendon per arm.
The PelaStar was chosen by the ETI for a 6 MW demonstration at the WaveHub
test site in Cornwall, supporting Alstoms 6 MW Haliade turbine.
Blue H TLPBlue H Group: TLP with a minimum of 3 taut-leg mooring lines,
each connected to a gravity anchor system suitable for a range of seabed con-
ditions. The turbine is located in the centre of a buoyant lightweight platform,
which is semi-submerged during operation and held in position by the uplifting
force and tension of the mooring tethers. An initial 80 kW prototype was
installed off the coast of Italy in 2008, and the concept has since been rened to
reduce structural mass. Blue H is currently seeking partners to nance and
construct a full-scale proof-of-concept unit with a 57 MW turbine.
GICON-SOFGICON: The TLP design consists of 4 linked columns anchored
to the seabed with 4 vertical taut-leg mooring lines and an additional 4 support
mooring lines to provide added stability that the designers anticipate to be
equivalent to conventional xed-bottom foundations. A a full-scale demon-
stration unit is planned for 2015 in the German Baltic Sea.
Eco TLPDBD Systems: The Eco TLP is a concrete gravity anchor and floater
coupled with taut tendons. Tank testing has been completed and DBD Systems
are now targeting a full-scale deployment at Dounreay Test Centre, off the north
coast of Scotland.
18 S.F. Gonzlez and V. Diaz-Casas

TLPWindIberdrola: Iberdrolas TLPWind concept consists of a central

cylindrical column and four pontoons symmetrically distributed on its bottom.
Each of the outer ends of the four pontoons incorporates porches which allow
the connection of two tendons per pontoon, which provide a level of redundancy
against potential tendon failures. At the top of the central column, a conical
frustrum allows a smooth transition between the main cylinder and the wind
turbine. The concept has performed well in tank testing and is now looking at
further optimisations to the structure and bespoke installation vessel ahead of a
potential deployment off the coast of Scotland by 2018.
Advanced Floating Turbine (AFT)Nautica: nauticas AFT is a hybrid of TLP
and semi-submersible characteristics. This deployment methodology can be
reversed to allow access for major work on the turbine. A 1:10 scale prototype is
planned for 2016.

4.4 Offshore Wind Turbines

Actually, almost all offshore wind turbines are tri-blade horizontal axis wind tur-
bines. Over recent years, the average size of wind turbines installed has increased
signicantly. As Fig. 7 shows, since year 2000 to the present the power of the
turbines has been doubled, from 2 MW to almost 4 MW [10].
As with average turbine size, the trends toward increasing hub heights and large
blade design appear to be slowing, though likely only temporarily. Figures 8 and 9
shows the hub height and rotor diameters, respectively, of global offshore wind
projects over time [10].
Increasing hub heights and larger blade designs accompany the trend toward
larger turbine sizes to provide for increased energy capture per turbine. While larger
blades increase each turbines swept area, the towers on which those turbines are
installed must also grow in order to accommodate the required blade-tip clearance
between the turbine and the sea surface.

Fig. 7 Average wind turbine size, in MW

Present and Future of Floating Offshore Wind 19

Fig. 8 Global offshore wind

plant hub heights

Fig. 9 Global offshore wind

plant rotor diameter

Virtually all current developments on floating platforms are designed with about
2.5 MW turbines. Although it is expected that in the coming years, when carry out
the nal launch of the floating wind, the turbines are similar to those employed on
xed offshore wind platforms.
As the newly announced 7, 8 MW and larger machines reach commercial
deployment in the next few years, the trend toward taller towers and larger blades is
likely to resume as well. This includes the following three turbine and blade designs
currently being prototyped:
The Vestas 8 MW prototype turbine installed in early 2014 has a rotor diameter
of 164 m, greater than any other turbine currently slated for construction through
2015 [11, 12].
Mitsubishi completed testing of the 82 m blade (167-m rotor diameter) for its
7 MW turbine in May 2014, and expects to complete the development of a
full-scale prototype turbine in summer 2014 [13]; however, this represents a
year-long delay from its original plans.
20 S.F. Gonzlez and V. Diaz-Casas

The Samsung 7 MW prototype turbine was commissioned in June 2014 at the

Fife Energy Park in Scotland [14]. The blade, developed by SSP Technology,
also uses carbon and holds the current record for the longest blade ever pro-
duced at 83.5 m (171-m rotor diameter). The blade is part of Samsungs 7-MW
turbine, which is expected to be deployed in 2015 in South Koreas rst offshore
wind plant [15].
Mitsubishi also plans to deploy the 7 MW turbine on a floating,
semi-submersible platform at the Fukushima Demonstration project in Japan [16].

5 Market Analysis

More than thirty concepts have been identied in the market from around ten
countries. These constitute concepts that have openly published material on their
design, and there may be more concepts currently under development that have not
been considered in this analysis. The majority of concepts assessed are
semi-submersibles, likely due to the flexibility of their application to site locations
with shallow water depths and the lower infrastructural requirements for installa-
tion. The more site constrained TLPs and spars are also well represented, with a
handful of multi-turbine platforms and hybrid wind/wave devices also under
By region, Europe leads the way in device development, representing two-thirds
of the concepts on the market; however, by country, Japan has the greatest number
of concepts, followed closely by the USA. There is also a discernible trend in the
typologies developed by different countries. The flexible application of
semi-submersibles means that they are a popular choice for development in most
markets, with France in particular targeting semi-submersible concepts that can be
deployed in shallower waters. Conversely, spar-buoys are generally limited to
geographies where very deep water is prevalent; namely, in Japan and Norway.
Meanwhile, TLP designs have principally emanated from the USA, as well as in a
handful of European countries.
Following a handful of successful full-scale demonstrations, the scale of floating
wind projects is beginning to increase as concepts prepare for commercial
deployment, both in terms of total project capacity and turbine rating. For concepts
such as Hywind and WindFloat, this is a natural progression from their single
prototype demonstrations of *2 MW towards precommercial arrays using 6 MW
units. A number of concepts are likely to follow this path, with single prototype
demonstrations planned for 20152018 (Table 8) [6, 17].
Despite just 8 MW of floating wind installed to date, build out is expected to
increase sharply over the next few years. With a handful of demonstrations
expected in 2015 and 2016, installed capacity is expected to spike in 2017 and 2018
as a number of pre-commercial arrays are constructed. While the phasing may alter,
with some projects experiencing delays or falling away and new developments
Present and Future of Floating Offshore Wind 21

Table 8 Floating wind prototype installations planned for 20152018

Year Name Country
Europe 2009 Hywind Demonstrator Norway
2011 WindFloatPhase 1 Portugal
2015 GICON-SOF Pilot Germany
2016 VERTIWIND France
2017 WindFloatPhase 2 Portugal
2017 Hywind Pilot Park Scotland
2018 Kincardine Offshore Wind Farm Scotland
2018 Dounreay Floating Offshore Wind Development Centre Scotland
2018 SEAREED France
Japan 2013 Kabashima
2013 Fukushima FORWARDPhase 1
2015 Fukushima FORWARDPhase 2
USA 2017 WindFloat Pacic
2018 Maine Aqua Ventus I
WindFloat Hawaii

coming online, it is expected that the cumulative capacity of *240 MW is certainly

attainable by 2020, and the success of these projects will likely determine build out
rates in the subsequent decade.


1. European Commission (2012) Blue growth opportunities for marine and maritime sustainable
2. Global Wind Energy Council (2015) Global wind statistics 2014
3. Japan Wind Power Association (2014) Target and roadmap for Japanese wind power
4. European Wind Energy Association (2015) EWEA: the European offshore wind industry
key trends and statistics 2014
5. European Wind Energy Association (2013) Deep water, July
6. James R, Costa Ros M (2015) Floating offshore wind : market and technology review, June
7. Buttereld CP, Musial W, Jonkman J, Sclavounos P, Wayman L (2007) Engineering
challenges for floating offshore wind turbines
8. Jonkman J, Matha D (2010) A quantitative comparison of the responses of three floating
platforms. In: European offshore wind 2009 conference and exhibition, vol 1, p 21, March
9. C.O. Technology and E. Enforcement (2013) Design guideline for stationkeeping systems of
floating offshore wind turbines, June
10. Navigant Consulting (2013) Offshore wind market and economic analysis-Annual market
assessment, p 126
11. Lewandowski CM (2013) The future of offshore wind
12. Smith A, Stehly T, Musial W (2015) 20142015 offshore wind technologies market report
technologies market report
22 S.F. Gonzlez and V. Diaz-Casas

13. Snieckus D (2013) Hunterston delay for Mitsubishi 7 MW. Available: http://www. Accessed 11 Nov 2015
14. PE (2013) Giant 7 MW Fife offshore turbine completed. Available:
news/engineering/giant-7mw-fe-offshore-turbine-completed. Accessed 11 Nov 2015
15. CompositesWorld (2013) DIAB core going into 83.5 m long wind turbine blade. Available:
Accessed 11 Nov 2015
16. Foster M (2014) Fukushima 7 MW platform complete. Available http://www. Accessed 11
Nov 2015
17. Pineda I, Azau S, Moccia J, Wilkes J (2014) Wind in power: 2013 European statistics.
Technical Report, pp 112, February
Life-Cycle Cost of a Floating Offshore
Wind Farm

Laura Castro-Santos

Abstract This chapter describes a general methodology in order to calculate the

costs of a floating offshore wind farm. It is based on the analysis of its life-cycle
cost system (LCS). In this sense, several phases have been dened: conception and
denition, design and development, manufacturing, installation, exploitation and
dismantling. The calculation of costs of each of these steps gives the total cost of a
floating offshore wind farm. The method proposed has been applied to the particular
case of the Galician coast, where a floating offshore wind farm could be installed
due to the great offshore wind potential and the depth of its waters. Results indicate
that the most important cost in the life cycle of a floating offshore wind farm is the
manufacturing cost. It is due to the fact that the floating offshore wind platforms and
the offshore wind turbines have a high cost. The methodology proposed can be used
by investors in the future to know the real costs of a floating offshore wind farm.

Keywords Life-cycle cost  Floating offshore wind farm  Ocean energy

1 Introduction

Offshore wind energy is an emerging technology which can be subdivided into two
main types of devices: xed (tri-piles, monopiles, etc.) and floating [semisub-
mersible, spar and tensioned leg platform (TLP)]. Nowadays, xed structures
compose the offshore wind farms in the North Sea. On the other hand, floating
offshore wind devices are new technologies, the study of which will be developed
in the following years.
In this sense, it is important to know the life cycle of a floating offshore wind
farm. It can be studied considering two views [1, 2]: environmental and economic.
Nevertheless, this chapter will focus on the economic issue.

L. Castro-Santos (&)
Departamento de Enxeara Naval e Ocenica, Universidade da Corua, C/Mendizbal, s/n,
15403 Ferrol, A Corua, Spain

Springer International Publishing Switzerland 2016 23

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_2
24 L. Castro-Santos

The main objective of this chapter is to dene a method to calculate all the costs
involved in a floating offshore wind farm. It is composed of two main steps: the
Economic Maps Tool (EMT) and the Restrictions Maps Tool (RMT). In addition,
the EMT includes the models selection (MS), the technical study (TS), the maps of
the costs (MC) and the maps of the economic indexes (MEI) [Net present value
(NPV), internal rate of return (IRR), discounted payback period (DPBP), Levelized
cost of energy (LCOE) and cost of power]. It is the rst step in analysing a floating
offshore wind location where a floating farm could be developed.

2 General Methodology to Calculate the Life-Cycle Cost

of a Floating Offshore Wind Farm

2.1 Methodology

The method carried out for calculating the costs of a floating offshore wind farm is
based on the phases of two different methods of life-cycle cost calculation [1, 2]: the
environmental and the economic view. This new methodology will be named as
life-cycle cost system of a floating offshore wind farm, LCSFOWF, and it will be
composed of several steps:
Economic Maps Tool (EMT).
Restrictions Maps Tool (RMT).
The EMT is supported by the following:
Models selection (MS): MS denes each of the models which will be taken into
consideration in the study according to offshore wind turbines, floating offshore
wind platforms, mooring lines, anchors, electric system, installation, accom-
modation, maintenance, seabed and dismantling.
Technical study (TS): TS consists in all the engineering calculation related to
floating offshore wind energy platforms technology, electrical cables, mooring
and anchoring dimensions and feasibility of mooring lines.
Maps of the costs (MC): Considering MS and TS the total cost of a floating
offshore wind farm can be calculated (LCSFOWF).
Maps of the economic indexes (MEI): All the economic indexes (NPV, IRR,
DPBP, LCOE and Cost of power) are calculated considering the costs obtained
in MC, the floating offshore wind energy production and the value of the electric
Results obtained from EMT will be processed with the RMT, which is devel-
oped using a geographic information system (GIS) software, the results of which
are the allowed areas considering the geographical restrictions of the site.
Consequently, not only EMT results but also RMT results will be used to
determine the resulting maps for a particular geographic case.
Life-Cycle Cost of a Floating Offshore Wind Farm 25

Fig. 1 General methodology

A detailed description of the model has been presented in [3]. A general scheme
could be seen in Fig. 1 [4].

2.2 Models Selection (MS)

The MS takes into consideration all the models taken into account in the study. In
this context, the variables whose costs are dependent on will be classied taking
into account three criteria [5]: level 1 variables, level 2 variables and level 3
variables. Level 1 variables are dependent on the general characteristics of the
floating offshore wind farm, for instance, wind scale parameter, wind shape
parameter, depth, height of waves, electric tariff, distance from farm to shore,
number of offshore wind turbines, etc. The second ones are dependent on the
particular aspects of the floating offshore wind farm. They will be subdivided taking
into account the main components of a floating offshore wind farm: offshore wind
turbine, floating offshore wind platform, mooring, anchoring and electric system.
Examples of this type of variables are power of wind turbine, diameter of wind
turbine, cost per unit of wind turbine, etc. Finally, level 3 variables are independent
of the characteristics of the farm. They are focused on the general constants
(gravity, air density, sea water density, etc.) or variables dependent on the shipyard
where the floating platform will be constructed (cost per hour of labour, salaries,

2.3 Technical Study (TS)

The TS consists in calculating all the dimensions that will be used to calculate the
total cost of a floating offshore wind farm.
26 L. Castro-Santos

In this sense, the floating offshore wind energy platforms are dened to their
main dimensions (weight, length, draft, etc.), but also the main characteristics of the
electrical cables, anchoring and mooring.
Furthermore, the location where the floating offshore wind farm will be installed
is dened considered its distance from shore, its depth, its offshore wind resource
and its wave resource (period and height of waves).

2.4 Maps of the Costs (MC)

The IEC 60300-3-3:2004 [7] proposes several models to calculate the life-cycle
cost. However, the present study will only take into account the model based on the
life-cycle phases because it is the most representative of the whole process.
In this context, the categories of the life-cycle cost of a floating offshore wind
farm are as follows [68]:
Capital Expenditure (CAPEX): initial investment cost. These costs will be taken
into consideration only once.
Operational Expenditure (OPEX): operation or exploitation cost. These costs
will be considered throughout the life cycle of the project.
If we apply this classication to the total costs of a floating offshore wind farm,
the CAPEX will include the rst four phases (conception and denition, design and
development, manufacturing, installation) and the last phase (dismantling), and the
OPEX will be composed by the phase 5 of exploitation, as Fig. 2 shows.

2.5 Maps of the Economic Indexes (MEI)

The maps of the economic indexes are based on obtaining the main variables on
which the economic indexes are dependent on. These economic indexes will be
NPV in , IRR in %, DPBP in years, the LCOE in /MWh and Cost of Power in

Fig. 2 CAPEX and OPEX components

Life-Cycle Cost of a Floating Offshore Wind Farm 27

M/MW. However, this chapter does not include their calculation, which is taken
into account in the next chapter.

2.6 Restrictions

It is important to know that phase MEI can have as a result a good location in
economic terms, but not in terms of the real installation process of the floating
offshore wind farm.
For instance, it can have a good IRR but it can be a navigation area, environ-
mental area, seismic zone, electrical cables, etc., where laws do not allow the
installation of the farm.
This is the main reason why the restrictions appear. Thus, the real locations can
be selected to carry out the floating offshore wind farm. However, this chapter does
not include this step.

3 Life Cycle of a Floating Offshore Wind Farm

3.1 Denition of the Life-Cycle Process

The life-cycle process has been dened modifying the recommendations of IEC
60300-3-3:2004 [7] because this normative is focused more on a product than on a
process. Therefore, the main phases of the life cycle of a floating offshore wind
farm are as follows:
Phase 1: Conception and denition.
Phase 2: Design and development.
Phase 3: Manufacturing.
Phase 4: Installation.
Phase 5: Exploitation.
Phase 6: Dismantling.
They are all shown in Fig. 3.

3.2 Breakdown Structure of the Process

The process breakdown structure establishes which are the main stages and sub-
stages of the process. A floating offshore wind farm is composed of ve main

P1: Conception and P2: Design and P3: P4:

P5: Exploitation P6: Dismantling
definition development Manufact. Installation

Fig. 3 Life cycle of a floating offshore wind farm

28 L. Castro-Santos

Fig. 4 Main components of a floating offshore wind farm

P1: Conception and P2: Design and P3:

Manufacturing P4: Installation P5: Exploitation P6: Dismantling
definition development

P31. Offshore wind P41. Offshore wind P61. Offshore wind

P11. Market study P21. Engineering project P51. Taxes
turbines manufacturing turbines installation turbines dismantling

P32. Floating platforms P42. Floating platforms P62. Floating platforms

P12. Law factors P52. Assurance
manufacturing installation dismantling

P33. Mooring P43. Mooring P63. Mooring and

P13. Design of the farm P53. Administration
manufacturing installation anchoring dismantling

P34. Anchoring P44. Anchoring P54. Electrical

P54. O&M
manufacturing installation elements dismantling

P35. Electrical elements P45. Electrical

installation P65. Cleaning

P46. Start up
P66. Materials disposal

Fig. 5 Breakdown structure of a floating offshore wind farm

components (Fig. 4): offshore wind turbines, floating offshore platforms, moorings,
anchorages and electrical elements.
In this context, the majority of the phases of the life-cycle process denition will
be developed for each of these elements, as Fig. 5 shows.

3.3 Initial Cost Breakdown Structure and Calculation

of Costs

3.3.1 Denition of the Cost Breakdown Structure

Initial cost breakdown structure (ICBS) of a floating offshore wind farm is based on
the disaggregation of the main costs of its life cycle. In this context, the main costs
Life-Cycle Cost of a Floating Offshore Wind Farm 29

Fig. 6 Initial cost breakdown structure of a floating offshore wind farm

are (Fig. 6) C1 is the cost of conception and denition, C2 is the cost of design and
development, C3 is the cost of manufacturing, C4 is the cost of installation, C5 is
the cost of exploitation and C6 is the cost of dismantling.
Therefore, the LCSFOWF k could be formulated as [9]:

LCSFOWF k C1 C2 C3k C4k C5k C6k 1

As Eq. (1) shows, the total cost will be a geo-referenced map of a floating
offshore wind farm, because each k value is a point of the geography where the
farm is installed. However, costs C1 and C2 are independent of the location k factor
and constant, because they are mainly related to onshore activities.
In addition, in order to obtain their main dependences, each of these costs should
be subdivided into sub-costs that should be analysed separately. The following
sections are focused on developing each of these sub-costs. Nevertheless, in order
to give a notion of the main dependences in costs, the following parameters could
be considered [4]:
Number of wind turbines.
Power of wind turbines.
Cost (in ) per MW of wind turbine.
Mass of the floating platform.
Mass of the wind turbine.
Cost of steel necessary to build the floating platforms at the shipyard.
Cost of direct labour at shipyard.
Cost of direct materials at shipyard.
Cost of no direct activities (management, ofce materials, amortization of the
machines, etc.) at shipyard.
Height and period of waves.
Wind speed at anemometer height.
Wind shape and wind scale parameters.
Weight of anchoring and mooring.
Anchoring and mooring cost per kilogram.
Number of mooring lines.
Cost per section of electrical cables.
Number of electrical cables.
Wind turbine diameter.
Distance to shore.
30 L. Castro-Santos

Grid and cable voltages.

Distance to port.
Distance to shipyard.
Number, speed and fleet of vessels used in installation phase.
Failure probability.

3.3.2 Conception and Denition Cost

The conception and denition cost takes into consideration studies, which have
been taken in advance to develop the floating offshore wind farm. Aspects as the
offshore wind resource or the environmental issues, which are very important to
know the main location of the farm, have been considered. In this sense, three
sub-costs have been taken into consideration: the market study cost (C11), leg-
islative factors cost (C12) and the farm design cost (C13), as Eq. (2) shows. The
market study cost considers the feasibility study, which has been developed to
know if the farm is feasible. The legislative factor cost considers the taxes that the
investor should pay in a previous phase. It is dependent on the country where the
farm is located. The farm design cost considers aspects as the energy resource study
or the seabed study.
C1 C11 C12 C13 2

3.3.3 Design and Development Cost

The design and development cost takes into account the engineering cost of the
floating offshore wind farm. In this sense, the engineers should consider the number
of floating offshore wind turbines which will be involved in the farm, the calculation
of the offshore and onshore electric cables, the calculation of the offshore substation
and the calculation of the moorings and anchors. This value has been calculated
considering the number of wind turbines (NWT) and the power of each of them
(PWT), as Eq. (3) shows:
C2 f NWT; PWT 3

3.3.4 Manufacturing Cost

The manufacturing cost takes into consideration the manufacturing of each of the
main components of a floating offshore wind farm: the offshore wind turbines
manufacturing (C31), the floating wind platforms manufacturing (C32), the
mooring manufacturing (C33), anchoring manufacturing (C34) and electric systems
manufacturing (C35), as equation shows:
Life-Cycle Cost of a Floating Offshore Wind Farm 31

C3 C31 C32 C33 C34 C35 4

The cost of manufacturing the offshore wind turbines is dependent on the cost of
each generator, the number of wind turbines and the power per unit of each wind
generator. The manufacturing cost is based on the activity-based cost
(ABC) proposed of a traditional shipyard, where the floating offshore wind plat-
forms are constructed. In this context, the cost of materials, direct labour and the
activities cost are the main parts in the calculation process of this manufacturing
cost. The mooring manufacturing cost is dependent on the mass per metre of the
mooring system, its length and the cost per kilogram of this mooring. In addition,
the number of lines of mooring per floating platform and the number of floating
platforms are also considered. The anchoring manufacturing cost is very similar to
the mooring manufacturing cost, but taken into account the mass of the anchor and
the price of the anchors per kilogram. Finally, the cost of manufacturing the electric
system is dependent on the number of wind turbines, the number of electric cables,
their length, the cost per metre of electric cable and the cost of the offshore

3.3.5 Installation Cost

The installation cost is composed of the cost of installing the offshore wind turbine
(C41), the cost of installing the floating offshore wind platforms (C42), the cost of
installing the mooring and anchoring (C43), the cost of installing the electric system
(C45) and the cost of the start-up, as equation shows:

C4 C41 C42 C43 C44 C45 5

The costs of installing the wind turbine and the floating platform are dependent
on the shipyard or port operations, the transport of the wind turbine or the floating
platform from the port to the farm and the offshore installation of the wind turbine
or the floating platform. It is important to know that the cost of transporting the
offshore wind turbines is included in the cost of transporting the floating platform
when the transport process is carried out considering a tug which tows the platform
from the shipyard or the port to the farm. The process of installing the mooring and
the anchoring is carried out considering an anchor handling vehicle (AHV).
Therefore, the cost is dependent on the direct labour cost per day, the cost of pumps
and divers, the number of anchors or moorings considered and the number of wind
turbines. Finally, the cost of installing the electric system takes into account the
installation of the electric cables and the installation of the offshore substation.
Therefore, variables as the cost of installing the electric cable, the metres of electric
cables installed per day or the number of lines of wind turbines have been
32 L. Castro-Santos

3.3.6 Exploitation Cost

The exploitation cost is composed of the assurance cost (C51), the business and
administration cost (C52) and the operation and maintenance cost (C53), as Eq. (6)
C5 C51 C52 C53 6

The assurance cost is dependent on the previous costs (conception and deni-
tion, design and development, manufacturing and installation). The business and
administration cost is composed of the number of years of the life cycle of the farm,
the cost of administration of the farm and the legal costs. On the other hand, the cost
of operating and maintenance considers the corrective and the preventive aspects
during the number of years of the life cycle of the floating offshore wind farm. In
this sense, aspects as inspections of equipments, calibration and adjust of sensors,
substitution of equipments, cleaning and visual reviews, among others, have been
considered. In the case of the corrective maintenance the probability of failure is
considered for each of the pieces of the offshore wind turbine, the floating platform,
the mooring and anchoring and the electric system.

3.3.7 Dismantling Cost

The floating offshore wind farm should be dismantled in order to clean and recover
the area where it has been installed. First, the farm is disassembled and, then, the
main material obtained (steel, copper, etc.) can be sold. This income is considered
as a negative cost. In addition, the cost of dismantling is dependent on the cost of
dismantling the offshore wind turbines (C61), the cost of dismantling the floating
platforms (C62), the cost of dismantling the mooring and anchoring systems (C63),
the cost of dismantling the electric system (C64), the cleaning cost of the location
(C65) and the removal cost of the materials (C66), as equation shows:

C6 C61 C62 C63 C64 C65 C66 7

4 Case of Study

The case of study considers a floating offshore wind farm of 21 wind Repower 5 M
turbines (NWT) of 5 MW of power (PWT) and located in Galicia (North-West of
Spain). This region has been selected because it has deep waters, which is a con-
dition to the installation of floating offshore wind platforms, it has a good offshore
wind resource, which will give better economic results in terms of protability of
the investment, and, nally, this area has a great experience in onshore wind and
naval sector. In this sense, the port and the shipyard considered are located in Ferrol
(A Corua, North-West of Galicia).
Life-Cycle Cost of a Floating Offshore Wind Farm 33

Fig. 7 Geo-referenced variables regarding waves (Hw, Tw), wind (cw, kw), depth (D), distances
from farm to shore (d), the storage area (dstorage) and the construction area (dconstruction)
34 L. Castro-Santos

Fig. 7 (continued)

In addition, the methodology has been applied to each of the points of the
geography where the floating offshore wind farm can be installed (k). Therefore,
several variables are geo-referenced: shape parameter (kw ) and scale parameter (cw )
of the wind [10], period (Tw ) and height of the waves (Hw ) [11, 12], depth (D),
distance to shore (d), distance from the farm to the storage area (dstorage ) and
distance from the farm to the platform construction area (dconstruction ), as Fig. 7
shows [13].
Moreover, a semisubmersible platform based on the Dutch tri-floater [14] has
been considered for the study. It is composed of three steel columns joined using
several steel pontoons. The offshore wind turbine is located in the centre of them.

5 Results

The conception and denition cost and the design and development cost, as
mentioned before, are independent of the area selected to install a floating offshore
wind farm. Therefore, their maps do not exist. Their constant values are 6.79 and
0.24 M, respectively.
Life-Cycle Cost of a Floating Offshore Wind Farm 35

On the other hand, costs of the other phases of the life cycle of a floating offshore
wind farm are dependent on the location. In this sense, aspects as distance to shore
and depth have influence on costs. It means that they can be drawn as independent
maps of the geography selected, in this case the Galician coast.
The cost of manufacturing goes from 215 M for the closest areas to the Galician
shore to 406 M for the most remote areas. The installation cost goes from 19 to
392 M. As Fig. 8 shows, these costs are dependent on the distance from shore,
mainly due to the fact that the installation cost is calculated considering the
transport from the port or the shipyard to the farm, which is developed using a
particular vessel whose trip depends on the distance.
The exploitation cost goes from 108 to 114 M and dismantling cost values from
0.0058 to 31 M. As we can see in Fig. 9, the cost of exploitation, which is mainly
composed of the cost of operation and maintenance, is less dependent on the
distance from shore than the cost of dismantling, whose value depends on the trips
of the vessels which are decommissioning the floating offshore wind farm.
Finally, the total life-cycle cost of a floating offshore wind farm of 21 wind
turbines of 5 MW values is from 366 to 946 M, as Fig. 10 shows.
Figure 11 shows the main cost of a floating offshore wind farm: the manufac-
turing cost, with the 62 % of the total cost. It is mainly due to the offshore wind

Fig. 8 Manufacturing and installation cost of a floating offshore wind farm

Fig. 9 Exploitation and dismantling cost of a floating offshore wind farm

36 L. Castro-Santos

Fig. 10 Total life-cycle cost

(LCS) of a floating offshore
wind farm

Fig. 11 Percentages of the

life-cycle cost of a floating
offshore wind farm

turbine cost and the floating platform cost, which is built in a shipyard. It is
followed by the exploitation cost (31 %), which includes the operation and main-
tenance cost (corrective and preventive). The other signicant cost is the installation
cost, with a 5 %. However, costs of conception and denition, costs of design and
development and costs of dismantling are irrelevant in comparison with the other
On the other hand, considering the CAPEX and the OPEX, values of 241 and
108 M are, respectively, obtained, as Fig. 12 shows.
Life-Cycle Cost of a Floating Offshore Wind Farm 37

Fig. 12 CAPEX and OPEX

of the life-cycle cost of a
floating offshore wind farm

6 Conclusions

This chapter describes a general methodology to calculate the costs of a floating

offshore wind farm. It is based on the analysis of its life-cycle cost system (LCS). In
this sense, several phases have been dened: conception and denition, design and
development, manufacturing, installation, exploitation and dismantling. The cal-
culation of costs of each of these steps gives the total cost of a floating offshore
wind farm.
The method proposed has been applied to the particular case of the Galician
coast, where a floating offshore wind farm could be installed due to the great
offshore wind potential and the depth of its waters.
Results indicate that the most important cost in the life cycle of a floating
offshore wind farm is the manufacturing cost. It is due to the fact that the floating
offshore wind platforms and the offshore wind turbines have a high cost.
The methodology proposed can be used by investors in the future to know the
real costs of a floating offshore wind farm. However, not only the economic aspects
are important. In this sense, you can be also limited by the area where you want to
install the farm. Therefore, a complementary study using GIS and considering
environment, sheries, navigation and fault lines, among others, should be


1. European Committee for Electrotechnical Standardization (2009) IEC 60300-3-3:2004.

Dependability management. Part 3-3: Application guide. Life cycle costing, pp 170
2. Fabrycky WJ, Blanchard BS (1991) Life-cycle cost and economic analysis. Prentice Hall,
Englewood Cliffs, NJ
3. Castro-Santos L, Ferreo Gonzlez S, Martnez Lpez A, Diaz-Casas V (2012) Design
parameters independent on the type of platform in floating offshore wind farms. Renew Energy
Power Qual J 10:15
4. Castro-Santos L, Prado G, Diaz-Casas V (2013) Methodology to study the life cycle cost of
floating offshore wind farms. In: 10th deep sea wind R&D conference, pp 179186
38 L. Castro-Santos

5. Castro-Santos L, Ferreo Gonzlez S, Martnez Lpez A, Diaz-Casas V, Ferreo Gonzalez S

(2012) Design parameters independent on the type of platform in floating offshore wind farms.
In: International conference on renewable energies and power quality (ICREP12), 2012, vol
10, no 10, pp 15
6. Barringer H (2003) A life cycle cost summary. In: International conference of maintenance
societies, pp 110
7. Levitt AC, Kempton W, Smith AP, Musial W, Firestone J (2011) Pricing offshore wind power.
Energy Policy 39(10):64086421
8. Tong K (1998) Technical and economic aspects of a floating offshore wind farm. J Wind Eng
Ind 76:112
9. Castro-Santos L, Diaz-Casas V (2014) Life-cycle cost analysis of floating offshore wind farms.
Renew Energy 66:4148
10. Hunt GL (2009) Maine offshore wind energy. Wind resources, technologies and energy
production, Maine
11. Cerda Salzmann DJ (2004) Ampelmann, development of the access system for offshore wind
turbines, TU Delft University
12. Alari V, Raudsepp U (2012) Simulation of wave damping near coast due to offshore wind
farms. J. Coast. Res. 28(1):143148
13. Meteogalicia (2012) Web page Meteogalicia. Available:
14. ECN, MARIN, L (2002) The Windmaster, TNO, TUD, and MSC, Study to feasibility of
boundary conditions for floating offshore wind turbines, Delft
Economic Feasibility of Floating Offshore
Wind Farms

Laura Castro-Santos

Abstract This chapter develops a methodology to determine the economic feasi-

bility of a floating offshore wind farm in a particular location. It is composed by
inputs (electric tariff, energy produced in the offshore location, the technical study)
and outputs [Internal Rate of Return (IRR), Net Present Value (NPV), Discounted
Pay-Back Period (DPBP), Levelized Cost Of Energy (LCOE) and Cost of power
ratio (Cpower)]. The method has been developed for a particular case of study: the
Galician area, located in the North-West of Spain, where the offshore wind resource
is high. Furthermore, this area has wide experience in onshore wind energy and in
shipbuilding. Therefore, technicians of both industries can be introduced in this new
sector. Results indicate the best areas where a floating offshore wind farm can be
installed in the Galician coast. Nevertheless, the feasibility study should be fol-
lowed by a sensitivity study, where one can research how modications in the input
variables considered carry out transformations in the output parameters. In this
sense, the most important variables are the scale wind parameter and the electric
tariff. The method can be useful for future investors to determine the feasibility of a
floating offshore wind farm in a particular location.

Keywords Net present value (NPV) Internal rate of return (IRR) Discounted 

Pay-Back period (DPBP) Levelized cost of energy (LCOE) Cost of power  

Feasibility study Floating offshore wind farm

1 Introduction

Nowadays, xed offshore wind structures (monopile, tripod, jackets, etc.) are being
installed in the shallow waters of the North Sea in Europe [1]. However, a greater
part of this planet Earth is composed of deep waters, which need other types of
offshore substructures.

L. Castro-Santos (&)
Departamento de Enxeara Naval e Ocenica, Universidade da Corua, C/ Mendizbal, s/n,
15403 Ferrol, A Corua, Spain

Springer International Publishing Switzerland 2016 39

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_3
40 L. Castro-Santos

In this sense, several prototypes of floating offshore wind platforms are being
developed. There are mainly three types of floating offshore wind technologies:
semisubmersible, spar and Tensioned Leg Platform (TLP). The most important
examples of these floating offshore wind devices around the world are the
semisubmersible WindFloat [2, 3], developed by Principle Power, and installed in
Portugal, and the spar platform called Hywind [4], developed by Statoil, and
installed in Norway.
These floating structures are in their development process. Therefore, it is
important to know their feasibility in order to improve their characteristics in the
future. In this sense, there are platforms that can be directly towed from the ship-
yard to the floating offshore wind farm location, which involves less cost than using
an offshore crane, the hiring of which is very expensive. Other platforms need less
steel for their construction. All of these examples are aspects that should be taken
into consideration.
In this context, the present chapter will carry out the methodology to determine
the economic feasibility of a floating offshore wind farm in a particular location.
Aspects as Net Present Value (NPV), Internal Rate of Return (IRR), Discounted
Pay-Back Period (DPBP), Levelized Cost Of Energy (LCOE) and Cost of power
ratio (Cpower) are taken into consideration. The method has been developed for a
particular case of study: the Galician area, located in the North-West of Spain,
where the offshore wind resource is high. Furthermore, this area has wide experi-
ence in onshore wind energy and in shipbuilding. Therefore, technicians of both
industries can be introduced in this new sector. Results indicate the best areas where
a floating offshore wind farm may be installed in this region.

2 Methodology

2.1 General Procedure

The general procedure of the economic feasibility study of a floating offshore wind
farm is composed by several inputs and several outputs, as Fig. 1 is shown.
In this context, electric tariff, energy produced in the offshore location, the
technical study and the costs are the inputs. The electric tariff will vary depending
on the country selected and its electrical laws. The energy produced is mainly
dependent on the wind speed of the location, the height of the tower of the offshore
wind turbine and the height of the part of the platform located over the sea. The
costs of the floating offshore wind farm are calculated taking into consideration the
technical study (calculation of electric cable dimensions, mooring and anchoring,
number of vessels required to transport the floating offshore wind platforms, etc.)
Economic Feasibility of Floating Offshore Wind Farms 41

Fig. 1 General procedure of the feasibility study of a floating offshore wind farm

On the other hand, the outputs of the method are composed by all these
parameters that indicate if the floating offshore wind farm is or not feasible. In this
sense, IRR, NPV, DPBP, LCOE and Cpower are calculated [6].
Nevertheless, the feasibility study should be followed by a sensitivity study,
where you can research how modications in the input variables considered carry
out transformations in the output parameters.

2.2 CAPEX and OPEX

Capital Expenditures or CAPEX is composed by the costs related to the concept

and denition cost, the design and development cost, the manufacturing cost, the
installation cost and the dismantling cost of the floating offshore wind farm [7].
Operating Expenditures or OPEX includes the exploitation cost (operating and
maintenance cost, taxes, among others) [7].
All of them, CAPEX and OPEX, have been calculated in the total life-cycle cost
of a floating offshore wind farm (LCSFOWF).

2.3 Net Present Value (NPV)

The Net Present Value (NPV) is the net value of the expenses (cash outflow) and
revenues (cash inflows) of a particular project, both of them discounted from the
42 L. Castro-Santos

beginning of the investment. Cash inflows are composed by incomes from the sale
of the electricity considering a specic floating offshore wind tariff. This tariff
changes depending on the country considered. On the other hand, Cash outflows are
composed of the maintenance, operation and nancial costs of the floating offshore
wind farm [6].
The investment decision on the floating offshore wind energy project depends on
the value of the NPV. In this sense:
If NPV > 0, the project will occur.
If NPV = 0, the project will be indifferent for investors.
If NPV < 0, the investor must discard the project.
Alternatively, if an investor has to choose several types of projects, he will tend
to select the floating offshore wind project with the highest NPV, because this
option will provide greater benets.
NPV is calculated considering Eq. (1):

NPV G0 1
t1 1 r t

CFt: cash flow on year t.
t: number of years of the project.
G0: initial investment costs.
r: discount rate.

2.4 Internal Rate of Return (IRR)

The Internal Rate of Return (IRR) is the value of r that makes the NPV zero. In this
sense, its calculation is more complicated that the NPV and requires an iterative
method. It is a polynomial equation of N degree; therefore, there are N different
solutions to the equation. Situations where there is only one value are easy to analyse.
Nevertheless, when results do not contain an approximate value rather multiple
positive solutions, the IRR analysis should be dismissed and other economic indi-
cators should be studied [6, 8]. IRR is calculated taken into account Eq. (2)

G0 0 2
t1 1 IRRt

CFt: cash flow on year t.
t: number of years of the project.
Economic Feasibility of Floating Offshore Wind Farms 43

G0: initial investment costs.

IRR: Internal Rate of Return.
The investment decision on the floating offshore wind energy project also
depends on the value of the IRR. In this sense:
If IRR > WACC, the project will occur.
If IRR = WACC, the project will be indifferent for investors.
If IRR < WACC, the investor must discard the project.
The Weighted Average Cost of Capital (WACC) is the discount rate of capital. If
we considered a project which has been nanced, then the WACC value is given by
Eq. (3) [6]:

MVe  Re MVd  Rd  1  T

MVe: total equity.
Re: cost of equity.
MVd: total debt.
Rd: cost of debt.
(1T): tax shield.

2.5 Discounted Pay-Back Period (DPBP)

The Discounted Pay-Back Period (DPBP) considers the cash flow of each year
with the respective discount rate and adds it to all the previous cash flows with their
respective discount rate, accumulating its NPV [6]. When this sum is equal or
greater than the initial investment, this is the year of the DPBP, as Eq. (4) is shown:

 G0 4
t1 1 r t

t: Discounted Pay-Back Period (years).
CFt: cash flow on year t.
G0: initial investment costs.
r: discount rate.
44 L. Castro-Santos

2.6 Levelized Cost of Energy (LCOE)

The Levelized Cost Of Energy (LCOE) is used to compare several energy

alternatives, whose values of investment and operational costs are different, but
which are installed in the same location. It depends on the total life-cycle cost of the
project in the t period LCSFOWFt , which goes from year 1 to the number of
years of the life-cycle of the project (Nfarm), the capital cost of the project (r) and the
energy produced by the floating offshore wind farm (Et) in kWh/year, as Eq. (5) is
shown [9]:
t0 1 rt
LCOE PNfarm Et
t0 1 rt

However, the total cost (LCSFOWFt ) depends on the year which has been taken
into consideration. In this sense, the total cost of year 0 is the cost of the investment
without the costs of exploitation (C5) and dismantling (C6). The total cost of years
from 1 to t is the exploitation cost (C5) divided by the number of years of the farm
(Nfarm). Finally, when the last year of the life-cycle of a floating offshore wind farm
is achieved, the total cost also includes the dismantling cost (C6):


Year 1  n LCSFOWF1t NC5


Year NFarm LCSFOWFNfarm NC5

C6 8

The objective of the investors in terms of LCOE is to obtain the lowest value of it.

2.7 Cost of Power Ratio

The location cost of power ratio (Cpower), in M/MW, is calculated dividing the
total cost of a floating offshore wind farm (LCSFOWF) in M, by the total power
installed in MW, as Eq. (9) is shown

Cpower 9

NWT: Number of floating offshore wind turbines.
PWT: Power of the floating offshore wind turbines (in MW).
The objective for investors in terms of Cpower is to obtain the lowest value of it.
Economic Feasibility of Floating Offshore Wind Farms 45

2.8 Sensitivity

The study of sensitivity in the feasibility of floating offshore wind farms indicates
how much the output variables (NPV, IRR, DPBP, LCOE, Cpower) can vary when
the input variables are changing their value [10]. It can be developed in several
ways: deterministically or using the Montecarlo simulation.
Traditional methods are deterministic models use single-point estimations,
obtaining the range of values that can have a variable to be studied [11]. On the
other hand, Montecarlo method allows considering the probability of occurrence of
each of these values, describing the behaviour of the variables in terms of a
probability distribution of each input variable. It generates thousands of possible
outcomes [12].
The main objective of these methods is managing variability and uncertainty,
quantifying risk and being useful for decision-making.

3 Case of Study

Galicia, the North-Western region of Spain, is the location considered to install a

hypothetic floating offshore wind farm of 21 wind Repower 5M offshore turbines
(NWT) of 5 MW of power (PWT).
The conception and denition costs and the design and development costs have
the constant values of 6.79 and 0.24 M, respectively. The cost of manufacturing
goes from 215 M for the closest areas to the Galician shore to 406 M for the most
remote areas. The installation cost goes from 19 to 392 M. As Fig. 2 shows, these
costs are dependent on the distance from shore.

Fig. 2 Manufacturing and installation costs of a floating offshore wind farm

46 L. Castro-Santos

Fig. 3 Exploitation and dismantling cost of a floating offshore wind farm

Fig. 4 Total life-cycle cost of

a floating offshore wind farm

The exploitation cost goes from 108 to 114 M and dismantling cost values from
0.0058 to 31 M, as shown in Fig. 3.
Finally, the total LCSFOWF of 21 wind turbines of 5 MW values from 366 to
946 M is shown in Fig. 4.
On the other hand, due to the instability of the Spanish electric tariff for
renewable energies, the value of which has changed several times in the last three
years, the value considered for electric tariff (PkWh) has been calculated taking into
account the minimum value which generates feasibility for a floating offshore wind
farm (IRR > r, NPV > 0 and DPBP < 20 years) [13, 14]. In this context, an initial
value of 190 /kWh has been considered.
Finally, a floating offshore semisubmersible platform based on the Dutch
Tri-Floater [15] has been taken into account. It is composed of three steel columns
joined using several steel pontoons. The offshore wind turbine is located in the
centre of them.
Economic Feasibility of Floating Offshore Wind Farms 47

4 Results

Considering the semisubmersible platform and the minimum electric tariff, the IRR
varies from 12.06 to 14.23 % (see Fig. 5), depending on the location considered.
Obviously, the best value is dependent on the offshore wind energy resource and
other factors as distance from shore, depth, among others.
Figure 6 shows negative values for NPV for far and near points of the geog-
raphy. However, these values are positive (the project would be feasible), for
intermediate areas. Therefore, the NPV varies from 383.1 to 366.79 M.
The best value for the DPBP is approximately 7 years, as Fig. 7 shows.
On the other hand, the best value for the LCOE of the semisubmersible platform
considered is 75.11 /MWh, as Fig. 8 shows.

Fig. 5 IRR of a
semisubmersible platform for
the minimum tariff

Fig. 6 NPV of a
semisubmersible platform for
the minimum tariff
48 L. Castro-Santos

Fig. 7 DPBP of a
semisubmersible platform for
the minimum tariff

Fig. 8 LCOE of a
semisubmersible platform for
the minimum tariff

Finally, the Cpower varies from 3.34 to 8.87 M/MW for the semisubmersible
platform taken into account, as Fig. 9 shows.
Finally, the sensitivity study was carried out using the Oracle Crystal BallTM
software, where you introduce all the input variables dening their type of prob-
ability distribution and their modication during the years; then you characterize the
output variables (IRR, NPV, DPBP, LCOE, Cpower) [16]. The software determines
the influences on the output parameters taking into account the possible changes in
the input parameters. Results obtained determine risks of a floating offshore wind
energy farm in a particular location.
In this case, 15 % of variation has been considered for all the input variables
taken into account: cw (wind scale parameter), PkWh (electric tariff in /kWh), CMW
(cost of the offshore wind turbine in /MW), Knancing (percentage of nancing),
r (discount rate), NWT (number of floating offshore wind turbines). In this sense,
results are shown in Table 1:
Economic Feasibility of Floating Offshore Wind Farms 49

Fig. 9 Cpower of a
semisubmersible platform for
the minimum tariff

Table 1 Sensitivity analysis [16]

Input variables Output variables
IRR (%) NPV (%) DPBP (%) LCOE (%) Cpower (%)
cw 76 73.8 76.7 92.1 64.3
PkWh 17.5 16.9 17.2
CMW 2.3 2.3 2.4
Knancing 1.6 1.1
r 0.8 2.6
NWT 6.9 0.9 18

As Table 1 shows, the wind scale parameter, which is related to the offshore
wind speed, is the variable which has more influence on results, followed by the
electric tariff. Therefore, variations in these aspects will generate that the outputs
vary considerably (until 92.1 % in the case of the scale parameter and the LCOE).

5 Conclusions

This chapter has carried out a methodology to determine the economic feasibility of
a floating offshore wind farm in a particular location. It is composed by inputs and
In this context, electric tariff, energy produced in the offshore location, the
technical study and the costs are the inputs. The electric tariff will vary depending
on the country selected and its electrical laws. The energy produced is mainly
dependent on the wind speed of the location, the height of the tower of the offshore
wind turbine and the height of the part of the platform located over the sea. The
costs of the floating offshore wind farm are calculated taking into consideration the
50 L. Castro-Santos

technical study (calculation of electric cable dimensions, mooring and anchoring,

number of vessels required to transport the floating offshore wind platforms, etc.).
On the other hand, the outputs of the method are composed by all these
parameters that indicate if the floating offshore wind farm is or not feasible. In this
sense, IRR, NPV, DPBP, LCOE and Cpower are calculated.
The method has been developed for a particular case of study: the Galician area,
located in the North-West of Spain, where the offshore wind resource is high.
Furthermore, this area has lots a wide experience in onshore wind energy and in
shipbuilding. Therefore, technicians of both industries can be introduced in this new
sector. Results indicate the best areas where a floating offshore wind farm can be
installed on the Galician coast.
Nevertheless, the feasibility study should be followed by a sensitivity study,
where you can research how modications in the input variables considered carry
out transformations in the output parameters. In this sense, the most important
variables are the scale wind parameter and the electric tariff.
The method can be useful for future investors to determine the feasibility of a
floating offshore wind farm in a particular location.


1. Nichols J, Camp T, Jonkman J, Buttereld S, Larsen T, Hansen A, Azcona J, Martinez A,

Munduate X, Vorpahl F, Kleinhansl S, Kohlmeier M, Kossel T, Bker C, Kaufer D (2009)
Offshore code comparison collaboration within IEA wind annex XXIII : phase III results
regarding tripod support structure modeling. In: 47th annual aerospace sciences meeting
2. Roddier D, Cermelli C (2009) Windfloat: a floating foundation for offshore wind turbines. Part
I: design basis and qualication process. In: ASME 28th international conference on ocean,
offshore and arctic engineering (OMAE2009), pp 19
3. Cermelli C (2009) Windfloat: a floating foundation for offshore wind turbines. Part II:
hydrodynamics analysis. In: ASME 28th international conference on ocean, offshore and arctic
engineering (OMAE2009), no 1, pp 19
4. Myhr A, Bjerkseter C, gotnes A, Nygaard TA (2014) Levelised cost of energy for offshore
floating wind turbines in a life cycle perspective. Renew Energy 66:714728
5. Castro-Santos L, Ferreo Gonzlez S, Martnez Lpez A, Diaz-Casas V, Ferreo Gonzalez S
(2012) Design parameters independent on the type of platform in floating offshore wind farms.
In: International conference on renewable energies and power quality (ICREP12), vol 10, no
10, pp 15
6. Short W, Packey D, Holt T (1995) A manual for the economic evaluation of energy efciency
and renewable energy technologies
7. Levitt AC, Kempton W, Smith AP, Musial W, Firestone J (2011) Pricing offshore wind power.
Energy Policy 39(10):64086421
8. Hertenstein JH, Mckinnon SM (1997) Solving the puzzle of the cash flow statement. Bus
Horiz 40(1):6976
9. Castro-Santos L, Diaz-Casas V (2015) Economic influence of location in floating offshore
wind farms. Ocean Eng 107:1322
10. Castro-Santos L, Diaz-Casas V (2015) Sensitivity analysis of floating offshore wind farms.
Energy Convers Manag 101:271277
Economic Feasibility of Floating Offshore Wind Farms 51

11. Cooke RM (1993) Graphical methods for uncertainty and sensitivity analysis. Math Stat
Methods Sensit Anal Model Output, pp 121
12. Charnes J (2007) Financial modeling with crystal ball and excel
13. Castro-Santos L (2013) Methodology related to the development of the economic evaluation
of floating offshore wind farms in terms of the analysis of the cost of their life-cycle phases,
Universidade da Corua
14. Castro-Santos L (2014) Metodologa para la evaluacin econmica de parques elicos offshore
flotantes. Eolus 68:2026
15. ECN, MARIN, L (2002) The Windmaster, TNO, TUD, and MSC, Study to feasibility of
boundary conditions for floating offshore wind turbines, Delft
16. Castro-Santos L, Diaz-Casas V (2015) Sensitivity analysis of floating offshore wind farms.
Energy Convers Manag 101:271277
Floating Offshore Wind Platforms

E. Uzunoglu, D. Karmakar and C. Guedes Soares

Abstract The chapter introduces the platform types designed for supporting the
tower, nacelle and turbine assembly of floating offshore wind turbines. Basic
information on hydrodynamics is presented to provide an understanding of platform
behaviour in waves. The classication method that distinguishes the platform types
is stabilization. Covering this topic, buoyancy, mooring and ballast stabilized
platform types are illustrated and briefly explained. The advantages and disad-
vantages of each approach are considered and summarized. The areas in which the
benets of each type are the most evident are claried through various comparative
studies. Most platform models are currently still at demonstration or conceptual
stages. For this reason, computer codes that aim to capture their motions become
particularly signicant in design stages, and they are discussed next. A signicant
number of projects at the initial stages of planning and application are being
developed in Japan, Europe and the United States. The platform types observed in
these projects are presented. The chapter concludes with newly evolving design
standards and a brief discussion on the optimization of platform shapes.

Keywords Offshore wind  Floating wind platforms

1 Introduction

The platform is an essential element of floating offshore wind turbines. Eventually,

it is the motions and loads due to the moving platform that separates the floating
wind turbines from xed-bottom structures. This chapter presents an outlook on the
platforms and aims to provide information on floater characteristics.
To understand why platforms differ from each other, a knowledge of the laws
that govern their motions is essential. The chapter contains a brief summary of this

E. Uzunoglu  D. Karmakar  C. Guedes Soares (&)

Centre for Marine Technology and Ocean Engineering (CENTEC), Instituto Superior
Tcnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon, Portugal

Springer International Publishing Switzerland 2016 53

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_4
54 E. Uzunoglu et al.

broad topic that explains the hydrodynamics of floating bodies. This information
helps to understand the characteristic properties of major platform types. In this
regard, recent research and demonstration projects frequently focus on semisub-
mersibles, TLPs and single point anchor reservoir (SPAR) platforms. These three
forms are evaluated in detail. Combinations of these main types and multi-turbine
setups are other possibilities presented briefly. In addition, a short discussion on
various projects undertaken globally on the design and installation of the offshore
floating wind turbine is provided.
The oil-and-gas industry provided the basis for most of the initial information
that currently applies to offshore platforms. In moving onto the newer eld of wind
power, additional aerodynamic loading signies that the behaviour of similar
platform types will differ in between the two industries. Consequently, the
knowledge needs to be updated and re-evaluated to focus on this combined loading.
Two ways of obtaining data are real-life measurements and experimental pro-
grammes. In these earlier phases of floating wind turbine installations, the number
of real-world turbines is limited. In this light, current experimental studies that
feature identical turbines tted on different platforms become crucial in under-
standing platform related differences in motions, loads and performance. The sec-
tion on comparative studies aims to summarize recent works in this aspect.
The additional aerodynamic loading combined with hydrodynamic loading also
reflects in estimating the platform motions through numerical models. Differences
in motion dynamics signify that tools that are capable of capturing the complete
response are required. Regarding the hydrodynamics of platforms, multiple tools
are already available and are subject to validation. A short discussion on these tools
and their approaches to hydrodynamic modelling is described next in the chapter.
The section also aims to integrate the requirements recently in development by the
class societies for floating counterparts of offshore wind turbines.

2 Hydrodynamics of Offshore Wind Turbine Platforms

The hydrodynamic response of the platform denes the overall stability of the
system and makes it one the most important factors in floating wind turbine design.
This section aims to provide brief information on the terminology and the theories
used in this aspect, starting with the static stability of the platform and continuing
into the dynamics of motions.
The static stability of the platform in still water is a result of the relative position
of its centre of gravity in comparison to its centre of buoyancy. The structures
centre of gravity is located at the centre of mass (denoted by G). The geometric
centre of the volume of the fluid displaced by the floating system is the centre of
buoyancy (indicated by B). The point of intersection between the line of action of
the buoyancy force and the centreline of the platform denes a point termed as the
metacentre (denoted by M). Figure 1 explains their locations in a stable floating
structure dislocated from its equilibrium position. The platforms static stability is
Floating Offshore Wind Platforms 55

Fig. 1 Placement of the G, B

and the metacenter

evaluated using the distance between the centre of gravity and the metacentre (GM).
A positive GM value denes the platform as initially stable in its upright position
(i.e. the metacentre is located above the centre of gravity). Unless stability is
provided by other means (e.g. mooring lines), this condition is fullled for all
floating structures. The method of attaining static stability is also used in classifying
floating offshore wind turbine platforms [1]. The platforms that depend on the
displaced volume to stay afloat are considered buoyancy stabilized. When addi-
tional weight is necessary to lower the centre of gravity to arrive at a positive GM
value, they are regarded as ballast stabilized. Mooring stabilized platforms form the
third category.
The platform afloat in water is subject to periodic wind and wave loads that
result in forces and moments that destabilize it. These dynamic loads result in
motions described in six modes, with the addition of the coupling between these
modes. Three transitional and three rotational motions are presented in Fig. 2. When
there is symmetry present as in the gure, the wave direction may be used as a
reference to differentiate between the modes such as roll and pitch.
Focusing on solely the hydrodynamic forces, a wave consisting of a single
sinusoidal component is the simplest form that excites the structure into harmonic
motion. This wave is represented using its amplitude, frequency and the phase. The
frequency of the wave relates to its wavelength as a function of the water depth and

Fig. 2 Six modes of platform

56 E. Uzunoglu et al.

the wave propagation speed. The wavelength is dened as the horizontal distance
between any two successive wave crests. The ratio between the wave height (i.e.
double the amplitude) and wavelength represents the wave steepness that plays a
signicant role in dening the wave characteristics.
In the presence of incoming waves, the total force acting on the floating body is
described by breaking it down into components. One component considers the body
as xed in water, and evaluates the forces on it due to incoming waves. The second
component considers the oscillating body in water without the incoming waves and
evaluates the forces resulting from this motion.
The incident waves that arrive at the structure form the wave exciting forces.
This force is a result of the integration of water pressures on the body and it is
termed the FroudeKrylov force. Additionally, the presence of the body diffracts
the fluid around it and forms the diffraction component.
The second part of total forces considers a structure that oscillates in water
without the incoming waves. Oscillations force the water to radiate away from the
body. This motion denes the radiation forces that relate to important hydrody-
namic characteristics of the platform. The acceleration in the fluid results in motions
that would be expected from a body with a higher mass. In other words, as the body
moves, it also acts as if it moves a certain volume of the surrounding fluid. This is
termed as added mass. The velocity of the motion factors in determining the
potential damping. Additionally, due to the viscosity of the water and depending
on the size of the structure, there may be signicant viscous damping.
Different methods are used in evaluating these forces and coefcients, and their
application depends on certain assumptions. The potential theory and Morrisons
equation are two approaches that are applied in most cases. Small wave steepness
combined with frictionless and irrotational flow points to potential theory.
Morrisons equation presumes that the diameter of the body in the water is small in
comparison to wavelength and the diffraction component is negligible. In this case,
the total force acting on the body is a summation of FroudeKrylov, inertial and
drag components. The motions of the platform have to be dominated by drag or
inertia for Morrisons approach to be considered valid. It is used ubiquitously for
offshore platforms in both oil and gas and offshore wind industries.
Superposition principle makes it possible to extend the case of a single sinu-
soidal wave into irregular seas. It was introduced by St. Denis and Pearson [2], who
dened irregular waves as a linear sum of multiple sinusoidal components, forming
the basis of motions in irregular waves. In reality, the summation is not entirely
linear, and the platform is also subject to loads that are outside the wave frequency
(WF, typically considered to reside between ranges of 425 s). Particularly in cases
that the structure is moored or water depths are shallow, platform behaviour suffers
from nonlinear (second-order) effects and linear theory does not represent the loads
effectively. The resulting motions are in lower frequencies (LF) than the wave
frequencies. Resonant conditions occur when they coincide with natural frequencies
of the platform or the mooring system. Consequently, avoiding resonance becomes
an important design decision.
Floating Offshore Wind Platforms 57

The solution of the second-order problem is computationally expensive.

However, it is possible to evaluate the rst-order forces and the second-order forces
separately and bring them together to determine the total response. In addition to
motions in wave frequencies, the platform may exhibit ringing and springing type
high frequency (HF) responses, and low-frequency (LF) responses of slowly
varying nature.

3 Main Offshore Wind Turbine Platform Concepts

As the offshore wind turbine industry evolves to meet the demand, the number of
proposed platforms increased signicantly, and most of them employ a slightly
different approach in terms of conguration. As of 2013, 80 different types were in
evaluation for production [3]. Despite this variety, it is possible to classify most of
the platform types into three major categories depending on their method of sta-
bilization [1]. These three types are buoyancy stabilized platforms, ballast sta-
bilized platforms and mooring stabilized platforms. In practice, a platform uses a
combination of these methods. Regardless, the prominent factor in stability is used
to identify the type of the foundation.
Semi-submersibles and barges provide examples to buoyancy stabilized
structures as they depend on the volume of the submerged body. The tension leg
platform (TLP) is not stable without its mooring setup, which classies it as a
mooring stabilized structure. SPARs are column-like structures that require heavy
ballasting at the bottom of the platform to overcome tipping. Barge-type platforms
behave similarly to semi-submersibles as they both depend on buoyancy combined
with catenary mooring. However, the barge has a larger waterplane area while the
semi-submersible mainly utilizes columns. For this reason, semi-submersibles,
TLPs, and SPARs were preferred as the representative forms in this section.

3.1 Buoyancy Stabilized Platforms

Semi-submersibles have been in use by the oil and gas industry since the 1960s.
They were seen favourably in terms of stability in waves, particularly against heave
and pitch motion. The platform sizes and encountered loads differ between the
offshore wind and oil gas industry. Regardless, the platform type is considered to be
one the major concepts.
The semi-submersible concept comprises of columns that provide the main
volume under water and connecting members that provide structural integrity to the
system as a whole. The oil-and-gas industry uses large volume pontoons under the
water and columns that pierce the water level to carry the deck; a setup that has its
origins in submersible hull forms. In passage to wind turbine platforms, this concept
evolved to suit the new specic needs. Considering that the payload of wind
58 E. Uzunoglu et al.

turbines is lower and the required deck space is equally reduced, reducing the
volume of the platform became essential in achieving economic feasibility.
The number of columns of wind turbine platforms changes with the concept and
the distance between columns becomes a factor that increases the stiffness of the
structure. Additionally, the distance changes the moments and forces coming from
each column, affects the platform bracing loads and the structural integrity. The
multi-body setup provides different ballasting options to counter pitch and roll
motions. Heave motion may be reduced by adding plates of large radii at their lower
extremity of the structure [4]. The platform offers relatively small water plane area;
therefore the natural frequencies in heave usually fall outside the wave frequencies
except for extreme weather conditions. While heave, pitch and roll are counteracted
by the restoring forces, surge, sway and yaw are dealt with by catenary mooring
Design decisions include column properties (e.g. rectangular vs. cylindrical) as
well as the placement of the tower and turbine. In Fig. 3, the WindFloat [5] platform
is illustrated, which is one of the early adopters of floating offshore wind turbine
technology. It brings together three columns connected by braces and places the
wind turbine on one of the columns. DeepCwind semi-submersible that was
designed by the DeepCWind consortium features a different solution, utilizing four
columns where one is placed at the centre and supports the wind turbine (Fig. 4).
DeepCWind and the WindFloat feature similar characteristics in terms of column
shapes and connecting braces. Alternatively, models without braces that make use
of entirely different congurations are possible. Fukushima Shimpuu is a concep-
tual semi-submersible that utilizes three columns and a larger pontoon connected by
pontoons that are set out in a V-shape. The turbine, in that case, is placed on the
bottom tip of the V.

Fig. 3 WindFloat
semi-submersible platform
Floating Offshore Wind Platforms 59

Fig. 4 Platform geometry of


One of the main advantages of semi-submersibles comes from their installation

procedure. As a hydrostatically stable structure, it may be towed to the deployment
site after being built onshore. Mooring system complexity and design are also lower
in comparison to TLPs, as they do not factor in stabilizing the structure. The variety
in possible geometry options, therefore the draft, suggests these platforms would
not need very deep waters to be deployed in.

3.2 Mooring Stabilized Platforms

The oil-and-gas industry initially developed TLP concept as a cost-effective way of

exploiting deeper water, dating back to 1970s. The rst working TLP was suc-
cessfully deployed by Conoco in the North Sea in the 1980s. The main feature of a
TLP lies in its stabilization characteristics. Instead of relying on the buoyancy
resulting from the underwater geometry, the stability of the structure is provided by
taut mooring lines. They have been mainly preferred for their resistance to motions
due to this extra stiffness. The tendons restrict motions in certain modes, providing
a stable platform and reducing their effect on the performance of the wind turbine.
However, mooring-related complications arise.
The typical TLP concept consists of a central column that carries the wind
turbine with arms that support the tendons extending from the main body. The
tendons are tensioned to provide stability. The number of arms and the angle
between them may vary, usually being three or four. Figure 5 represents the
underwater geometry of a representative TLP platform.
The lack of buoyancy provided by the volume of the platform is not enough to
keep the structure afloat. Therefore, extra stiffness is provided by increasing the
tension of the mooring lines and submerging the platform deeper. As a result, roll,
pitch, and heave motions become highly restricted. TLP platforms behave more like
a xed structure in these modes. The methods to connect the tendons to the seabed
include tension piles, gravity anchors and suction buckets.
60 E. Uzunoglu et al.

Fig. 5 Underwater geometry

of a typical TLP offshore
wind turbine

In the optimization of the platform, the shape of the body and the arms (e.g.
rectangular vs. cylindrical), the placement of extra weight, and the general form
may be taken as parameters. The distance of the tendons from the main body
signicantly alters the moment provided by the mooring. In comparison to catenary
mooring used in other platforms, the tendon length is shorter as they feature direct
distance to the seabed. The tendons counteract the aerodynamic forces coming from
the turbine and wave exciting forces. Therefore, they are subject to increased
loading. The rigid-like behaviour of the platform also reflects in its natural fre-
quencies. The added stiffness leads to higher frequencies in modes that are not
allowed to move due to tendon tension. Consequently, resonant cases have to be
treated accordingly in the design of offshore wind turbine platforms.
In installation phases, it is possible to provide the platform temporary stability by
ballasting the main body to tow the structure to its deployment site. In this case, the
structure is ballasted at the pre-installation phase, and ballast is removed following
the installation of the anchors on-site. If the structure cannot be towed, it will need
to be carried to the deployment site by other means. Similarly, maintenance plan-
ning also becomes an issue, as the turbine cannot be moved onshore by towing.
Regarding the wave loads on the main body, the volume of the column is lower
in comparison to buoyancy stabilized platforms. This fact effectively reduces the
loads by incoming waves, and it is an advantage in terms of corrosion as it typically
occurs at the water line level. One disadvantage is the loss of a mooring line, which
should be paid attention to at design stage. Unlike catenary mooring solutions, the
loss of the stability providing mooring line may result in losing the platform instead
of drifting.
For the reasons outlined above, the design of a TLP platform requires high
coupling between the platform and the mooring system. In return, it delivers a
platform that behaves rigidly in heave, roll and pitch and is compliant in surge,
sway and yaw modes. While TLP provides the rigidity for efcient operation of
wind turbines, installation and maintenance complexity shows to be its primary
Floating Offshore Wind Platforms 61

3.3 Ballast Stabilized Platforms

SPAR platforms are the main ballast stabilized platforms. Originally designed as
buoys that gather oceanographic information, they were put into use for production
by the oil and gas industry in the 1990s by the Oryx Neptune eld. Since then,
alternative forms have come into use, focusing on slightly different applications but
the main idea stayed unchanged. SPAR resembles a slender cylinder. To provide
stability, heavy ballast is used at the lower extremity of the platform, shifting the
centre of gravity to below the centre of buoyancy. Station keeping is provided by
catenary mooring.
Figure 6 illustrates the underwater geometry of a representative SPAR platform.
SPARs are intended as deep-water platforms due to the necessity of placing the
ballast conveniently deep to achieve desired restoring moment. The heavy weight at
the bottom of the cylinder also counters the rotational motions of pitch and roll. The
restoring forces in heave are provided by the surface area at the water line. This
small water plane area characterizes the SPAR as a platform that has low stiffness in
heave. Due to symmetry, hydrodynamic excitation enforces only negligible yaw
motion on the body of the cylinder. On the other hand, aerodynamic loads induce
yaw motion through the applied moments on the blades. The mooring lines have to
counteract these forces. When catenary mooring lines are attached directly to the
body of the platform, they introduce a limited amount of yaw restoring moment. As
the platforms are intended for higher depths, the mooring length increases, and this
may reflect in costs. In contrast, the hull represents a simple-to-produce form that
may turn out to be economical. At the design stage, these considerations have to be
The simplicity of the hull form makes the SPAR susceptible to a phenomenon
called vortex-induced vibrations (VIVs). As the cylinder heaves, it creates vortices
that alter the pressure distribution along the surface, and the flow becomes irregular.
This pressure change leads to low-frequency vortex-induced vibrations, causing
higher mean current forces and leading to fatigue [6]. Various methods are available
to eliminate the cause of VIVs. An example to common practice is to employ
strakes at the cost of increased drag and total mass. Water currents may need to be

Fig. 6 Underwater geometry

of a SPAR platform
62 E. Uzunoglu et al.

taken into account when designing a SPAR, considering the signicant underwater
body of the platform. In terms of corrosion effectiveness, the water plane area is
signicantly small compared to semi-submersibles and other buoyancy stabilized
platforms. As a result, corrosion at water level is also low. In summary, SPAR
provides a simple-shaped platform that has slow and small motions due to its small
waterplane area. It may become disadvantageous economically due to its higher
depth requirement.

3.4 Hybrid Platforms and Multi-turbine Concepts

In most cases, any platform type ts into one of the classications given above.
However, alternatives that will accommodate the turbine and tower better are
always in consideration. Such structures aim to employ a combination of the
methods listed above to reap the advantages of different platforms and bring them
together. Heave restrained TLP, that is compliant in other modes but is restrained in
heave, and the heave restrained SPAR, a concept that limits heave motion of the
buoy are these kinds of hybrid structures [7]. There are also concepts that aim to
integrate other forms of energy (such as wave) into a single system [8].
An alternative to single turbine platforms, integrating multiple turbines together
on a single platform brings its advantages and disadvantages. The benets include
using a single grid connection and a mooring system for multiple turbines and the
possibility having a common service area. The difculties arise mainly due to the
proximity of multiple turbines. As in xed-bottom offshore wind turbines, floating
offshore wind turbines also have to be installed as farms to provide a signicant
amount of energy. Their placement effects performance of the surrounding turbines
due to the wake created behind the turbine. Any multi-turbine solution has to be
aware of such effects and accommodate it in design. This may also mean that
nacelle yaw motion may sufce to mitigate the wake effect, and motion of the
platform may be necessary to face the wind.
The Swedish company Hexicon proposes a concept (Fig. 7) that hosts four
turbines on a single platform. Another concept of three wind turbines on the
semi-submersible platform is developed by Shimiju Corporation in collaboration
with University of Tokyo (Fig. 8).

4 Evaluation of Platform Concepts

Major questions of platform design include how to choose the optimal platform
type that provides the most efcient performance and how to optimize the selected
shape. The current knowledge about global platform motions mostly derives from
the previous experiences of oil-and-gas-related projects. However, the installation
Floating Offshore Wind Platforms 63

Fig. 7 Hexicon H4-24 MW

multi-turbine platform

Fig. 8 Shimiju corporation

multi-turbine platform

of the turbine and the tower changes the responses signicantly. For this reason, it
is important to conduct studies that concern wind turbine platforms specically.
According to their global responses, the platforms can be divided into categories
where they are classied as compliant or restrained in their respective modes of
motion. A summary that denes these boundary conditions is presented in
ISSC2012 [9]. Table 1 couples this information with the primary means of stabi-
lizing the platforms. Compliant modes respond to the exciting forces coming from
the environment while restrained modes show small responses. Restrained motions
are usually measured in centimetres, limited by the mooring line elasticity. While
this information provides a brief understanding of motion characteristics of the
platforms, further probing is necessary for more details.

Table Platform types according to their stability and motion compliance

Platform Stability Surge and sway Heave Roll and pitch Yaw
SPAR Ballast C C C C
Semi sub Buoyancy C C C C
TLP Mooring C R R C
C indicates compliant modes and R indicates restrained modes
64 E. Uzunoglu et al.

Due to the coupled hydrodynamics and aerodynamics, the platform behaviour

cannot be evaluated separately from the turbine and tower. In the same light, to
isolate the effect of the platform on system dynamics, the studies need to keep all
other variables (turbine, blade and tower assembly) identical and change only the
platform type.
Experimental works of this nature have already been carried out and are discussed
in the following sections. Two particular studies were chosen. Both investigate a
5 MW turbine mounted on different types of platforms to obtain a comparison of
characteristics. They feature collaborative works bringing multiple groups together.

4.1 DeepCWind Consortium

The USA-based DeepCWind consortium carried out testing of three platforms

coupled with the scaled model of the NREL 5 MW turbine [10]. The setup included
a TLP, a semi-submersible and a SPAR. The study aimed to exclude proprietary
information to provide the researchers open access to data and the experimental
programme was carried out at the Maritime Institute of the Netherlands (MARIN).
This study is invaluable in terms of isolating and understanding the effects of
platform behaviour on system dynamics.
The semi-submersible was designed by the University of Maine led DeepCWind
Consortium [11]. It features a four-column design with the turbine placed in the
middle column and three extra columns placed to form a triangle (see Fig. 4).
The TLP concept shows similarities to Glostens PelaStar [12]. The SPAR platform
is based on the currently afloat Hywind SPAR, which was also the subject of an
earlier code comparison study [13]. They are presented in Fig. 9. Full results of the
study were made available by Robertson et al. [14], and summary of the ndings
are provided below.
A scale of 1/50 was utilized to prepare the model and the environment, using
Froude scaled wind, and wave loads. Scaling laws play a signicant role in
experimental setups, and they are discussed in Jain et al. [15] and Martin et al. [16].
The test procedure followed a methodology explained in Koo et al. [17]. According
to this methodology, initially the natural frequencies of the platforms are identied
using hammer tests, followed by the determination of static equilibrium position.
Hydrodynamic characteristics such as viscous damping are determined through free
decay tests. After the platform is examined in the absence of external forces, the
effect of wind and wave forces are evaluated separately. Hydrodynamic studies
include regular and irregular waves. Wind responses are examined for steady and
dynamic wind. The last stage brings the conditions together to perform tests under
combined wind and wave loading.
The natural frequencies of the platforms are summarized in Fig. 10 [18].
Comparing this data to the WF region (425 s [19]) and between them provides an
idea of the platform dynamics. Surge natural periods are signicantly longer for the
semi-submersible in comparison to TLP and SPAR. Due to symmetry, sway is
Floating Offshore Wind Platforms 65

Fig. 9 The DeepCWind semi-submersible, TLP, and the spar-buoy used in scaled model tests

Fig. 10 Natural periods of the platforms in six modes of the DeepCWind plaforms

almost identical to surge for all platforms. The TLP clearly identies itself in heave,
pitch and roll by proving to be a very stiff structure with short natural periods (under
5 s). Other platforms deliver periods closer to 30 s. Yaw period is different for all
platforms. The SPAR provides very short yaw periods, resembling a restrained
structure. TLPs yaw period is slightly over the WF region, and the semisub-
mersible delivers signicantly long periods in comparison to the other two
66 E. Uzunoglu et al.

platforms. These results clarify the methods of avoiding resonance, either by going
above or staying below the exciting frequencies. In terms of resonant motions, it
was also noted that the platform affects the bending frequencies of the tower as it
changes the total stiffness of the system.
The results have shown that under steady wind, surge responses of the TLP and
the semi-submersible platform turned out to be similar and delivered larger motions
in comparison to the SPAR. In the absence of waves, the surface area facing the
wind in addition to the turbine thrust may factor in surge, making SPAR advan-
tageous. The semi-submersible has delivered the highest surge motion overall,
which may be reasoned with the same disadvantage.
Pitch motion was the highest for the SPAR-buoy, as pitching and rolling are only
counteracted by the ballast. The TLP showed almost no pitching due to its
restrictions forced by the mooring system. For this set of experimental models, the
semi-submersible delivered pitching motion that equals roughly half of SPARs
pitching motion amplitude. At this point, it is important to add as a reminder that
going beyond certain angles of pitch motion limits the aerodynamic performance of
the turbine considerably. Additionally, higher pitch angles reflect as larger tower
base bending moment due to the weight of the nacelle-turbine assembly on the top
of the tower.
Similar ndings were made available for the combined wind and wave loading
case. Experimental data showed that the response amplitude operators of combined
wind/wave loading and waves without the wind were similar. According to these
ndings, waves are the primary driver of platform motions in all modes, as opposed
to wind. Additionally, the work presented an experimental conrmation of the
presence of motions outside the WF, identifying higher order wave effects to be a
consideration for all platform types. However, the semi-submersible delivered by
far the highest responses outside the WF region.

4.2 Collaborative Comparison Studies in Japan

Japan shifted their focus to clean energy after the Fukushima nuclear disaster in
2011. With the aim of speeding up this process, a comparative study was carried out
with the joint effort of Osaka Prefecture, Yokohama National, Nihon and Osaka
Universities. Contributors provided their platform design, to be coupled with a
5 MW scaled turbine and tower of 90 m. A TLP, two semi-submersibles and a
SPAR type platform with a scale of 1/100 were evaluated (see Fig. 11). The
leftmost model shows similarities to the DeepCWind TLP with three mooring lines
of 120 in between, using a longer fairlead distance. Two alternative semisub-
mersible platforms were studied. One of the suggested forms uses a single point
mooring system and places the turbine on one of the three columns. The second
model employs a 4-column approach by placing the turbine on top of the central
column. The SPAR platform brings an alternative to the standard cylindrical
Floating Offshore Wind Platforms 67

Fig. 11 The TLP, two semi-submersibles, and the spar-buoy in their respective order, used in
comparative studies in Japan

approach, with the addition of an extra floating column in the form of a disc
placed at 15 % of the total draft below the water level.
The comparative results of this study were discussed in Nihei et al. [20] and the
ndings concur with DeepCWind studies in terms of platform characteristics. Their
study classied TLP favourably for providing stability in pitch, roll and heave
motion. SPAR was identied to show highest acceleration values in most envi-
ronmental conditions. Considering that high acceleration affects the crew at the
nacelle level, they have proposed that the platform may induce difculties main-
tenance. In addition to wave and wind induced motions of the platforms, the work
has reported signicant yaw motion of the SPAR that they have attributed to the
gyroscopic effect of the rotor combined with the small moment of inertia of the
platform. Their conclusion concerning the platforms agrees with the previous
ndings and shows waves as the main driver of platform motions as opposed to
wind. Single point moored semi-submersible provided an exception to this general

5 Numerical Analysis of Platform Motions

The dynamics of floating offshore wind turbines involves signicant coupling

between the aerodynamics of the turbine and the hydrodynamics of the platform.
The motions from the turbine, waves, and the moorings all contribute to the global
response of the system. Consequently, the number of parts that come together to
form the structure imply that the analysis can only be carried out through a coupled
approach. It should also consider the interaction between all components and their
relation regarding the natural frequencies to avoid resonance and fatigue. There are
two ways to approach this problem. A validated hydrodynamic tool may be
extended to include aerodynamics of the turbine, or an aerodynamic code may be
extended to include the hydrodynamics of the platform and mooring. In both cases,
to capture the nonlinearity of the responses in wind turbine platforms, usually
68 E. Uzunoglu et al.

time-domain analysis is preferred as opposed to frequency domain approaches [21].

In hydrodynamic calculations, the main difference between codes lies in their basis
of theories employed to estimate the motions. Most of the codes consider the
Morrison equation, the potential theory, or a combination of both [22].
One of the most prominent works in the eld of numerical tool development is
the offshore code comparison collaboration studies that bring different codes to
provide comparative results. The IEA Task 23 (OC3), Phase IV involved a study on
the Hywind SPAR platform [21] as joint work of multiple groups using a total of
seven different codes. On a continuation study, a semi-submersible was analysed in
Task 30 (OC4) Phase II [23] with an increased number of collaborators. The results
have been published in [24]. These studies have contributed to the understanding of
the validity of various modelling approaches and their application.
Code-to-code studies are signicant in understanding the effect of theories on the
motions; however experiment-to-code studies are essential in validating the tools.
Consequently, recent results that compare numerical and experimental data have
been published [25]. These works are important in identifying and evaluating areas
where the codes need improvement, such as the mooring models [26]. The recent
iteration of code comparison studies, OC5, is ongoing as of this date and aims to
provide comparative results for a wider range of tools with the purpose of validating
them against experiments.

6 Floating Offshore Wind Turbine Projects

The research on the floating offshore wind turbine by various project teams and
consortia is progressing, and the majority of them have invested in R&D projects.
Globally different floating foundation projects are underway and some of the
companies such as Vestas, Siemens, Areva, Mitsubishi and Fuji Heavy Industries
working on the wind turbine technology are involved in the project team for the
development of floating wind turbine foundations. In addition, Acciona, Alstom,
and Gamesa are also working on the projects as consortium members and Samsung
of Korea is involved in the development of floating foundation as well as for the
development of the turbine. The involvement of these global companies shows that
the floating offshore wind turbine foundation technology has emerged from pure
R&D status and is moving towards deployment of a number of pilot plants in
Europe, Japan US and other countries. Some of the offshore floating wind turbine
foundation projects underway are as follows.

6.1 Projects in Japan

In Japan, various design concepts of the floating foundation are developed. The
design concept on the SPARs, semi-submersible, TLP, and barge are widely under
Floating Offshore Wind Platforms 69

production, and the majority of the Japanese research projects are government
funded. The nuclear accident in 2011 at Fukushima has changed the vision of the
Japanese government. Most of the funding are involved in the projects related to
renewable energy and in 2011, the rst scale model was launched in Hakata Bay in
Kyushu. In 2012, the SPAR design was deployed off the Kabashima Island in
Kyushu. In Figs. 12ac and 13ac, some of the floating foundation for offshore
wind turbine under development and operation in Japan are presented.
In 2013, a 2 MW full-scale SPAR wind turbine was developed and deployed off
Choshi at the entrance of Tokyo Bay. The floating foundation technology on
semi-submersible and TLP under development by Mitsubishi and Mitsui Zosen at
Fukushima are funded by the Japanese government. These floating foundation

Fig. 12 a Wind lens, Kyushu, b Semi-submersible at Mitsui off Fukushima and c Japan marine
united advanced SPAR [3]

Fig. 13 a Mitsui Zosen semi-submersible, Fukushima, b Mitsubishi semi-submersible Fukushima

and c Mitsui Zosen TLP, Fukushima [3]
70 E. Uzunoglu et al.

concepts in progress are proposed to be of full commercialization, which makes

Japan a global leader with regards to full-scale floating pilot projects.

6.2 Projects in Europe

In Europe, the offshore wind energy potential is in abundance in deep and inter-
mediate water depths. Most of the offshore floating foundation technologies under
development are led by the countries having deep-water offshore potential. The
floating foundation concepts under development by various companies, such as
Blue H from the Netherlands, Nass et Wind of France, GICON of Germany,
Poseidon of Denmark, IDEOL of France, HiPR Wind of Spain, Winflo of France
and Hexicon of Sweden. The development of the offshore technology is due to the
positive global market condition on offshore renewable energy. Some of the
floating foundation projects undertaken by these companies in Europe are presented
in Figs. 14ac and 15ac.
Most of the European projects on the offshore floating foundation technology
were started by medium size companies, investing own capital in the R&D and
some government funding. The technologies developed by these companies are
now involved in the prototype construction globally and successfully nding
investors for necessary funds to develop the foundation concepts commercially.

6.3 Projects in the United States

In the US, the government plans to develop wind turbine technology for power
generation of about 20 % by 2030. The National Renewable Energy Laboratory
(NREL) is playing a major role in the development of offshore wind turbine

Fig. 14 a GICON, Germany, b Winflo, France and c IDEOL, France [3]

Floating Offshore Wind Platforms 71

Fig. 15 a HiPR wind, Spain, b WindSea, Norway and c Gusto trifloater, the Netherlands [3]

technology and is also dedicated to the advancement of renewable energy concept

for commercial application. The US Department of Energy is also involved in
various R&D projects on offshore floating foundation technology. The focus is on
the cost reduction of the novel technology over existing offshore wind technology.
In the intermediate and deep-water depth, the company Principle Power, based in
Seattle, WA is involved in the development of the WindFloat semi-submersible
floating foundation concept. The WindFloat foundation provides for placing the
offshore wind turbines independent of water depth. The floating concepts developed
for the advancement of wind power technology is presented in Fig. 16a, b.
In early 2011, Principle Power, EDP, InovCapital, Vestas and other partners
deployed the rst full-scale 2 MW WindFloat off the coast of Portugal. The
full-scale pilot was launched into the water in late 2011 and has been successfully
operating using a Vestas 2 MW turbine. In order to reduce the cost of offshore wind
energy, the Glosten PelaStar engineering farm of Seattle, WA was selected in 2011

Fig. 16 a PelaStar TLP and b Nautica windpower, USA [3]

72 E. Uzunoglu et al.

through targeted technology advancement. Now, various technology developments

are in progress, and these will help into benchmark the initial foundation of offshore
wind turbine technology.

7 Design and Optimization of Floating Offshore Wind

Turbine Platforms

7.1 Design Standards

Design standards evolve according to the needs and the status of the industry. When
this is considered, it is as expected that the standards dealing with offshore wind
turbines are signicantly new. Considering the developments of standards for
floaters, it is important to note that offshore wind turbines differ signicantly from
their onshore counterparts regarding the loads that they encounter. They experience
signicantly different environment-related conditions such as hydrodynamic loads,
sea ice and longer periods of standby. Additionally, they differ from offshore oil rig
platforms due to wind loads and shallow water effects, such as steeper waves, and
they operate unmanned. For these reasons, they have to be considered separately,
addressing all of these issues to improve condence in their implementation.
To address these concerns, DNV and GL both published the rst editions of their
standards for offshore wind turbines in 2004 and 2005, which have been subse-
quently updated [19, 27]. The IEC standard IEC-61400-3 was rst published in
2009 [28] as a result of 10 years of study, starting with the development process in
1999 [29]. ABS then followed its rules with the ABS standard #176 in 2010 [30].
Following the trend in the industry, the previous set of rules was mainly
applicable to xed-bottom offshore structures. However, they have served as a
precursor to evaluating their application for floating wind turbines as in the case of
DNV standard OS-J101 [31]. Bureau Veritas issued a note concerning offshore
floating wind turbines in 2010 to address floating platforms [32] mainly addressing
IEC-61400-3 and dening load cases to be tested out.
From 2012 onwards, the focus shifted to floating platforms and class societies
have undertaken work to develop standards concerning floating offshore wind
turbines. DNV introduced OS-J103 [10] in 2013 with ABS providing ABS #195
guide [33]. Following, the events of the Fukushima nuclear disaster in 2011,
Nippon Kaiji Kyokai followed with their set of rules in 2012 [34]. The work of
DNV is a result of a Joint Industry Project that was initiated in September 2011 with
Statoil, Navantia, Iberdrola, Alstom Gamesa, Sasebo Heavy Industries, Nippon
Steel Corporation, STX, Principle Power and Glosten Associates to bring together a
wider range of experience in the eld [35]. IEC has also accepted the standard to
dene design requirements for floating offshore wind turbines in an ongoing work
[36]. The expanded standard aims to address the differences between floating off-
shore wind turbines and xed-bottom wind turbines. Considerations such as
Floating Offshore Wind Platforms 73

hydrostatic stability and the effect of second-order hydrodynamics on platforms are

taken into account [37].
One of the challenges in floating offshore wind turbines continues being the high
costs associated. In order to lower costs, the need for more efcient designs
becomes inevitable. For this reason, the classication standards try to incorporate
cost-benet analysis and reliability into the design process [35]. Without a doubt,
life-cycle assessment is also an important part of the process with cyclic loads
playing important roles in fatigue and other aspects. The system suffers from
multiple types of loads. The platform has to accommodate these loads and provide
stability. Consequently, class societies have to consider a number of factors in
platform design different from the topics that are covered in bottom-xed platforms.
The hydrodynamics of the platform has to be evaluated thoroughly in installation,
operation, and accidental cases and set within certain standards. Shallow water
effects factor heavily in floating offshore structures. Therefore, the design has to
consider higher order wave effects on the platform. Unimodal spectrums become
insufcient for floaters excited by swell, forcing two peaked spectral density models
to be considered. Longer simulation times will be needed to capture nonlinearities
and slowly varying responses of the platform.
At the earlier stages of their development, the rules concerning the platforms are
currently relatively new. As the industry gathers more experience in the eld, they
will become subsequently updated to address the concerns of safety and operation.

7.2 Shape Optimization

The type of the platform reveals only a general idea of how it will behave under load.
In addition to differences due to types, each design will provide different responses
depending on the shape and hydrodynamic properties. Recent studies address
optimization problems and provide the form that conrms with the best perfor-
mance. These studies involve various methods and are linked to design standards of
platforms. To follow a standard optimization process, certain environmental and
load conditions will have to be chosen to base the decisions on shape optimization.
Additionally, the number of possible congurations (i.e. connecting brace cong-
uration of a semi-submersible) is limitless, making optimization a broad topic.
Despite these challenges, the problem has been tackled by various authors.
Parametric studies have been considered one of the solutions to optimize the
structure [38]. These studies chose a number of parameters to be optimized, such as
the platform diameter and depth and create a number of geometries in order to
obtain the desired performance characteristics. Using simplications such as
omitting the connecting bracings of multi-body structures, the geometry may be
modelled rapidly [39, 40], however, time-domain simulations of turbine motions
are computationally expensive. Therefore in this approach, evaluation of design
space requires considerable computing power. To facilitate the process, genetic
algorithms [41] and other nonlinear approaches [42] are used.
74 E. Uzunoglu et al.

Acknowledgements This work was performed within the Strategic Research Plan of the Centre
for Marine Technology and Ocean Engineering, which is nanced by Portuguese Foundation for
Science and Technology (Fundao para a Cincia e Tecnologia-FCT).


1. Buttereld CP, Musial W, Jonkman J, Sclavounos P, Wayman L (2007) Engineering

challenges for floating offshore wind turbines, National Renewable Energy Laboratory
2. St Denis M, Pierson WJ Jr (1953) On the motions of ships in confused seas. Soc Nav Archit
Mar Eng Trans 61:280357
3. Bossler A (2013) Floating offshore wind foundations: industry consortia and projects in the
United States, Europe and Japan, Maine International Consulting LLC
4. Cermelli C, Roddier D, Aubault A (2009) WindFloat: a floating foundation for offshore wind
turbinespart II: hydrodynamics analysis. In: ASME 2009 28th international conference on
ocean, offshore and arctic engineering. American Society of Mechanical Engineers,
pp 135143
5. Roddier D, Cermelli C, Aubault A, Weinstein A (2010) WindFloat: a floating foundation for
offshore wind turbines. J Renew Sustain Energy 2:033104
6. Hirdaris S, Bai W, Dessi D, Ergin A, Gu X, Hermundstad O, Huijsmans R, Iijima K,
Nielsen U, Parunov J (2014) Loads for use in the design of ships and offshore structures.
Ocean Eng 78:131174
7. DNV (2013) Design of floating wind turbine structures, DNV-OS-J103. Det Norske Veritas,
Hovik, Norway
8. Guedes Soares C, Bhattacharjee J, Karmakar D (2014) Overview and prospects for
development of wave and offshore wind energy. Brodogradnja 65:87109
9. Brennan FP, Falzarano J, Geo Z, Lendet E, Le Boulluec M, Rim CW, Sirkar J, Sun L,
Suzuki H, Thiry A, Trarieux F, Wang CM (2012) Committee report on offshore renewable
energy. International Ship and Offshore Structures Congress (ISSC) 2012, Rostock/Germany,
pp 153200. GL 2012. Rules and guidelines, IV industrial services, Part 2: guidelines for the
certication of offshore wind turbines, Germanischer Lloyd, Hamburg, Germany
10. Jonkman J, Butteeld S, Musial W, Scott G (2009) Denition of a 5 MW reference wind
turbine for offshore system development. National Renewable Energy Laboratory (NREL),
11. Thiagarajan K, Dagher H (2014) A review of floating platform concepts for offshore wind
energy generation. J Offshore Mech Arct Eng 136:020903
12. Moon III WL, Nordstrom CJ (2010) Tension leg platform turbine: a unique integration of
mature technologies. In: Proceedings of the 16th offshore symposium. A25A34
13. Jonkman JM (2010) Denition of the floating system for phase IV of OC3. National
Renewable Energy Laboratory, Technical report. NREL/TP-500-47535
14. Robertson AN, Jonkman JM, Goupee AJ, Coulling AJ, Prowell I, Browning J, Masciola MD,
Molta P (2013) Summary of conclusions and recommendations drawn from the DeepCWind
scaled floating offshore wind system test campaign. In: ASME 2013 32nd international
conference on ocean, offshore and arctic engineering. American Society of Mechanical
Engineers, V008T09A053-V008T09A053
15. Jain A, Robertson AN, Jonkman JM, Goupee AJ, Swift R (2012) FAST code verication of
scaling laws for DeepCwind floating wind system tests. In: Proceedings of 22nd international
offshore and polar engineering conference, pp 355365
16. Martin HR, Kimball RW, Viselli AM, Goupee AJ (2014) Methodology for wind/wave basin
testing of floating offshore wind turbines. J Offshore Mech Arct Eng 136:020905
17. Koo BJ, Goupee AJ, Kimball RW, Lambrakos KF (2014) Model tests for a floating wind
turbine on three different floaters. J Offshore Mech Arct Eng 136:020907
Floating Offshore Wind Platforms 75

18. Kimball R, Goupee A, Coulling A, Dagher H (2012) Model test comparisons of TLP,
spar-buoy and semi-submersible floating offshore wind turbine systems. In: Proceedings of
2012 SNAME annual meeting and expo
19. DNV (2011) Design of offshore wind turbine systems. DNV-OS-J101, Det Norske Veritas,
Hovik, Norway
20. Nihei Y, Iijima K, Murai M, Ikoma T (2014) A comparative study of motion performance of
four different FOWT designs in combined wind and wave loads. In: ASME 2014 33rd
international conference on ocean, offshore and arctic engineering. American Society of
Mechanical Engineers, V007T05A025-V007T05A025
21. Cordle A, Jonkman JM, Hassan GG (2011) State of the art in floating wind turbine design
tools. National Renewable Energy Laboratory
22. Jonkman J, Musial W (2010) Offshore code comparison collaboration (OC3) for IEA task 23
offshore wind technology and deployment. Contract 303:2753000
23. Robertson A, Jonkman J, Masciola M, Song H, Goupee A, Coulling A, Luan C (2012)
Denition of the semisubmersible floating system for phase II of OC4. Offshore code
comparison collaboration continuation. IEA Task 30
24. Robertson A, Jonkman J, Vorphal F, Popko W, Qvist J, Froyd L, Xiaohong C, Azcona J,
Uzunoglu E, Guedes Soares C, Luan C, Yutong H, Pengcheng F, Yde A, Larsen T, Nichols J,
Buils R, Lei L, Nygaard TA, Manolas D, Heege A, Ringdalen Vatne S, Ormberg H, Duarte T,
Godreau C, Hansen HF, Nielsen AW, Riber H, Le Cunff C, Beyer F, Yamaguchi A, Jin Jun K,
Shin H, Shi W, Park H, Alves M, Guerinel M (2014). Offshore Code Comparsion
Collaboration Continuation Within IEA WIND Task 30: Phase II Results Regarding a Floating
Semisubmersible Wind System. 33rd International Conference on Ocean, Offshore and Arctic
Engineering (OMAE 2014), San Francisco, California, USA. OMAE2014-24040
25. Goupee AJ, Koo BJ, Kimball RW, Lambrakos KF, Dagher HJ (2014) Experimental
comparison of three floating wind turbine concepts. J Offshore Mech Arct Eng 136:020906
26. Uzunoglu E, Guedes Soares C (2015) Comparison of numerical and experimental data for a
DeepCwind type semi-submersible floating offshore wind turbine. In: Guedes Soares C
(ed) Renewable energies offshore. Taylor & Francis Group, Lisbon, pp 747754
27. GL, Germanischer Lloyd (2012) Rules and guidelines, IV industrial services, part 2: gudelines
for the certication of offshore wind turbines. Hamburg, Germany
28. IEC (2009) Wind turbinespart 3: design requirements for offshore wind turbines.
IEC61400-3, International Electrotechnical Commission
29. Quarton D (2005) An international design standard for offshore wind turbines. Garrad Hassan,
p 13
30. ABS (2010) Guideline for building and classing offshore wind turbine installations, ABS
#176. American Bureau of Shipping, Houston, Texas
31. Ronold KO, Hansen VL, Godvig M, Landet E, Jorgensen ER, Hopstad ALH (2010) Guideline
for offshore floating wind turbine structures. In: ASME 2010 29th international conference on
ocean, offshore and arctic engineering (OMAE), Shanghai, China, OMAE2010-20344
32. BV, Bureau Veritas (2010) Classication and certication of floating offshore wind turbines.
BV Guidance Note NI 572, Bureau Veritas
33. ABS (2014) Guide for building and classing floating offshore wind turbine installations, ABS
#195. American Bureau of Shipping, Houston, Texas
34. NK (2012) ClassNK guidelines for offshore floating wind turbine structures. Nippon Kaiji
Kyokai, Tokyo, Japan
35. Hopstad ALH, Ronold KO, Sixtensson C, Sandberg J (2013) Standard development for
floating offshore wind turbine structures. EWEA Offshore 2013, Frankfurt, Germany
36. IEC (2014) IEC 61400-3-2 Ed 1.0 Wind turbines part 3-2: design requirements for floating
offshore wind turbines
37. Sirnivas S, Musial W, Bailey B, Filippelli M (2014) Assessment of offshore wind system
design, safety, and operation standards. National Renewable Energy Laboratory, USA
76 E. Uzunoglu et al.

38. Sclavounos P, Tracy C, Lee S (2008). Floating offshore wind turbines: responses in a seastate
pareto optimal designs and economic assessment. In: ASME 2008 27th international
conference on offshore mechanics and arctic engineering. American Society of Mechanical
Engineers, pp 3141
39. Uzunoglu E, Guedes Soares C (2015) Parametric modelling of multi-body cylindrical offshore
wind turbine platforms. In Guedes Soares C, Santos TA (eds) Maritime technology and
engineering. Taylor & Francis Group, London, pp 11851193
40. Uzunoglu E, Guedes Soares C (2015) Influence of bracings on the hydrodynamic modelling of
a semi-submersible offshore wind turbine platform. In: Guedes Soares C (ed) Renewable
energies offshore. Taylor & Francis Group, pp 755762
41. Hall M, Buckham B, Crawford C (2013) Evolving offshore wind: a genetic algorithm-based
support structure optimization framework for floating wind turbines. OCEANS-Bergen, 2013
42. Kim JH, Hong SY, Kim HJ (2013) The shape design and analysis of floating offshore wind
turbine structures with damper structure and shallow draft. In: The twenty-third international
offshore and polar engineering conference. International Society of Offshore and Polar
CFD Applied to Floating Offshore
Wind Energy

M.I. Lamas and C.G. Rodrguez

Abstract Nowadays, it is well known that offshore is a promising alternative for

energy production. Many researchers have built different kinds of machines. The
main problem is that it is very important to fully understand the hydrodynamics of
wind turbines to design an optimal mechanism. For this reason, CFD
(Computational Fluid Dynamics) has become a very powerful technique because it
solves the governing equations of conservation of mass and momentum so as to
obtain the fluid flow characteristics. CFD analyses provide details of the velocity
and pressure elds, as well as the hydrodynamic forces, cruising velocity, power
and efciency. This is very useful to identify the advantages and disadvantages of
new designs of wind turbines. In this regard, this chapter illustrates how to realize a
3D CFD model of a wind turbine, from the initial stage of CAD designing and
meshing to the nal stage of analysis of the results.

Keywords CFD  Computational fluid dynamics  Numerical  Offshore

1 Introduction to CFD

The term CFD comes from Computational Fluid Dynamics. This is a eld of
Fluid Mechanics which employs numerical procedures to solve the governing
equations of the flows. The application eld of CFD is very wide and many books
have been written, such as [15]. Some examples of CFD simulations are indicated
below. Figure 1 indicates the mesh and simulation results of an aluminium sample
which is being melted by a laser beam [6].
Figure 2 indicates the computational mesh and velocity eld of an innovative
marine propeller which consists on an undulating n [7].
Figure 3 indicates the mesh and results of an internal combustion engine, [8].

M.I. Lamas (&)  C.G. Rodrguez

Departamento de Enxeara Naval e Ocenica, Universidade da Corua, C/Mendizbal, s/n,
15403 Ferrol, A Corua, Spain

Springer International Publishing Switzerland 2016 77

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_5
78 M.I. Lamas and C.G. Rodrguez

Fig. 1 Simulation of melting by a laser beam; a computational mesh; b temperature eld

Fig. 2 Simulation of a marine propeller based on an undulating n; a computational mesh;

b pressure eld of the middle section

In addition to the examples indicated in Figs. 1, 2 and 3, CFD can be applied to a

wide variety of mechanical fluids problems, such as heat transfer, turbulence,
chemistry, compressible flows, combustion, electromagnetism, etc.
Nowadays there are many CFD softwares. The most employed ones are indi-
cated below:
OpenFoam: This is a free, open source CFD software package produced by
OpenCFD Ltd. Basically, it consists of a flexible set of C++ modules for dif-
ferent engineering applications. This is the most famous free software about
CFD. As it allows access to the whole source code, it is easy to modify them and
simulate any problem that the user need. On the contrary, commercial softwares
do not allow access to the source code so it is not possible to modify the content
that these include by default.
Fluent: This is a very famous commercial software developed by ANSYS Ltd. It
is easy to understand and it allows to simulate a huge variety of CFD problems.
StarCD: Developed by CD-adapco.
Converge: Developed by CEI software. This is a curious CFD software since
automatically generates the mesh at runtime.
KIVA: This is a commercial software developed by Los Alamos National
Laboratory. It was specially designed to simulate internal combustion engines.
CFD Applied to Floating Offshore Wind Energy 79

Fig. 3 Simulation of an engine; a computational mesh; b pressure eld of the middle section

Flow3D: Developed by Flow Science.

Briefly, the CFD method consists on splitting the computational domain into a
lot of discrete elements constituting a mesh. The governing equations are solved in
these elements. The differential equations are discretized in algebraic equations
before being solved.
The discretization process is the approximation of a continuous problem into
discrete quantities. In CFD, three discretizations must be done:
Spatial discretization: this consists on splitting the computational domain into a
lot of smaller elements, which constitute a mesh.
Equation discretization: this consists on generating a system of algebraic
equations from the governing differential equations.
Temporal discretization: this is only applied to transient problems (steady
problems do not need a temporal discretization). It consists of splitting the time
in intervals called time steps.
The rst step in all CFD procedures is splitting the computational domain into a
lot of small elements called mesh or grid, Figs. 1a, 2a and 3a. Nodes are located at
the centre of each element. The mesh cannot have holes and the elements cannot
overlap. The selection of the mesh size is a critical process in all CFD procedures. If
the mesh is too ne, high computational recourses are needed. On the contrary,
coarse meshes may conduct to wrong results.
The discretization process of the governing equations provides an algebraic
equation, Eq. (1), for each control volume. This equation can be re-written as:
80 M.I. Lamas and C.G. Rodrguez

aP /nP aN /nN Su 1

where /nN is the value of the adjacent cells to the node P. Therefore, the value of /nP
depends on the adjacent cells, creating a system of linear algebraic equations which
can be expressed as:

A/ S 2

where [A] is a matrix with a lot zero values, [] is a vector of the dependent
variables and [S] is a vector with the source terms.
The system of Eq. (2) provides the values of for each node. The solution can
be done by direct methods and indirect or iterative methods. Simple examples of
direct methods are Cramers rule matrix inversion and Gaussian elimination. On the
other hand, iterative methods are based on the repeated application of a relatively
simple algorithm leading to eventual convergence after a number of repetitions.
Well-known examples are the Jacobi, Gauss-Seidel, TDMA, etc.

2 CFD Simulation of a Wind Turbine

Once the basic performance of CFD was described, the methodology to simulate a
wind turbine will be treated in this section.

2.1 Governing Equations

The governing equations of the flow around the wind turbine are the Navier-Stokes
ones. The energy equation is not needed since the thermal problem can be
neglected. In Cartesian tensor form, the continuity equation is given by:

@q @
qui 0 3
@t @xi

where is the density and u the velocity. It is very common to consider the air
incompressible, so the continuity equation results:

0 4

The momentum conservation equation is given by:

CFD Applied to Floating Offshore Wind Energy 81

@q @ @p @sij
qui qui uj  5
@t @xj @xi @xj

where ij is the stress tensor. If the fluid is treated as Newtonian, the stress tensor
components are given by:
@ui @uj 2 @uk
sij l  dij 6
@xj @xi 3 @xk

2.2 Movement of the Vanes

When simulating a wind turbine, one of the main difculties is the movement of the
vanes. It is possible to realize moving mesh simulations in OpenFOAM and many
works about this have been published until now. Nevertheless, a problem involving
a moving mesh is too complicated and the simulation needs high computational
resources and thus much time to run.
The movement of the vanes can be easily solved in a xed mesh using a
coordinate system which rotates with the vanes. Common CFD problems use a
xed coordinate system, but a rotating coordinate system allows to solve the
problem using a xed mesh. To this end, a Coriolis acceleration must be applied to
the momentum equation. In this regard, the velocity is relative to the rotating
reference system, so the equations of continuity and momentum results:

0 7

@ @   @p @sij
quri quri urj q2x  ur x  x  r  8
@t @xj @xi @xj

where 2x  ur is a momentum source which represents the Coriolis acceleration

and x  x  r the centripetal acceleration. The viscous stress srij is identical to
ordinary problems except that relative velocity derivatives are used.

2.3 Calculation of the Hydrodynamic Forces

As the wind passes through the turbine, a force is produced. The components of the
force, can be evaluated by integrating the projection of the pressure and the shear
stress in the x, y and z directions. The total thrust can be computed by adding the
pressure and viscous forces contributions, Eq. (9):
82 M.I. Lamas and C.G. Rodrguez

Fx Fp Fv 9

where Fp is the pressure force and Fv is the viscous force.

The pressure force is given by:
Fp  pndA 10

where n is the component of the unit normal vector on dA.

The viscous force is given by:
Fv  sj nj dA 11

where sj is the viscous stress tensor.

2.4 Turbulence

Todays standard in wind turbine simulations are Reynolds Averaged

Navier-Stokes (RANS) methods. Another approach consists on Large Eddy
Simulation (LES) techniques. LES and RANS techniques differ in the way they
address the present impossibility to resolve all the scales present in engine flows.
RANS simulations are based on a statistical averaging to solve only the mean flow.
This implies that modelling concerns the whole spectrum of scales. In LES, a
spatial or temporal ltering is used to represent the large turbulent scales of the
flow, which are directly resolved, while the small scales are modelled. In LES,
modelling thus concerns a much smaller part of the spectrum, which leads to an
improvement of predictavility as compared to RANS. LES inherently allows to
address large scale unsteady phenomena, and thus has a good potential to predict
engine unsteadiness. The problem is that LES would lead to a CPU time that is way
beyond reach of present supercomputers. Therefore, the use of LES is not very
In the eld of RANS methods, the two-equation model standard k- is the most
used to simulate wind turbines. The RNG k- model is also widely employed
because it is recommended for rotating flows.
As indicated above, the momentum conservation equation for a turbulent flow is
given by Eq. (8). A common method to model the Reynolds stresses, quri urj , is
the Boussinesq hypothesis to relate the Reynolds stresses to the mean velocity
CFD Applied to Floating Offshore Wind Energy 83

@ui @uj 2 @uk
quri urj ut  qk lt dij 12
@xi @xi 3 @xk

where ij is the Kronecker delta (ij = 1 if i = j and ij = 0 if i j), which is included

to make the formula applicable to the normal Reynolds stresses for which
i = j (Versteeg and Malalasekera [1]) and t is the turbulent viscosity. The k- model
includes two differential equations, corresponding to the turbulent kinetic energy
(k) and its dissipation rate (), given by Eqs. (13) and (14) respectively.
@ @ @ @k
qk qkui ak lt Gk Gb  qe  YM 13
@t @xi @xi @xj

@ @ @ e e2
qe qeui C1e Gk G3e Gb  C2e q 14
@t @xi @xj k k

In the above equations, Gk represents the generation of turbulence kinetic energy due
to the mean velocity gradients; Gb is the generation of turbulence kinetic energy due to
buoyancy; YM represents the contribution of the fluctuating dilatation in compressible
turbulence to the overall dissipation rate. C, C1, C2, C3, k and are constants and
the terms k and represent the inverse effective Prandtl numbers for k and
The turbulent viscosity, t, is computed by combining k and as follows:

lt qCl 15

2.5 Numerical Procedure

In this section, the generation of the mesh and other numerical details will be

2.5.1 CAD Design

The rst step is to create a 3D CAD design. Many softwares can be employed for
this purpose. For example, commercial softwares are SolidWorks, SolidEdge,
Rhinoceros. Free softwares are FreeCAD, Blender, etc. For simple geometries, this
step can be omitted and the CAD design can be generated using the meshing
software; nevertheless, it is recommended to employ these softwares, especially for
the design of the blades. As the geometry of the blades includes too curve forms,
84 M.I. Lamas and C.G. Rodrguez

Fig. 4 Computation domain

the softwares Blender (free software) or Rhinoceros (commercial software) are

highly recommended.
The computation domain must be a box with the wind turbine inside, as shown
in Fig. 4. Several tests must be performed to determine the adequate extent of the
numerical domain, dimensions a, b and c in the gure, in such a way to eliminate
any potential effect of the outer boundaries, theoretically at innity, on the flow
close to the turbine.

2.5.2 Mesh

As indicated above, the principle of operation of CFD codes is subdividing the

domain into a number of smaller, non-overlapping sub-domains. The result is a grid
(or mesh) of cells (or elements). Too many grid generation programs can be found.
Regarding commercial softwares, Gambit, ANSYS Icem, Star-CCM are too
appropriate. Regarding free softwares, Salome, GMSH, Engrid, etc.
Hexahedral elements provide better accuracy and stability, so a structured
hexahedral mesh must be adopted if possible. Sometimes it is not possible or it is
too difcult to employ hexahedral elements in the totality of the control volume. In
these cases, tetrahedral elements must be employed in complicated regions. The
critical zones, i.e. near the vanes, must be meshed with a ner size.

2.5.3 Boundary and Initial Conditions

All CFD models require initial and boundary conditions. In the case of a wind
turbine, the appropriate boundary conditions are indicated in Fig. 5. Downwards,
the boundary condition must be an inlet and upwards an outlet. The exterior
CFD Applied to Floating Offshore Wind Energy 85

Fig. 5 Boundary conditions

boundary conditions must be declared taking into account that the influence of the
exterior boundary conditions indicated in the gure must be as neglecting as pos-
sible. A slip boundary condition is adequate to this end.

2.5.4 Calculation Parameters

A good choice is a rst or second order differencing scheme in time. A rst or

second order differencing scheme in space can also be used. Implicit procedures are
recommended. Pressure velocity coupling of the continuity equation may be solved
using both SIMPLE or PISO procedures. The grid size and the time step sensibility
must be studied in order to verify that both of them are adequate to obtain accurate
enough results.

2.5.5 Resolution of the Governing Equations

As indicated previously, several solvers are available in the literature to solve the
governing equations.

2.5.6 Analysis of the Results

The last step is to analyse the results obtained by the CFD solver. Some solvers
include analysis tools and there are other softwares specic for this purpose. For
instance, ParaView. As example, Fig. 6 indicates the velocity eld around a wind
86 M.I. Lamas and C.G. Rodrguez

Fig. 6 Velocity eld around

a wind turbine

3 Conclusions

The present chapter has indicated how to realize a CFD simulation of a wind
turbine. The pressure and velocity elds can be obtained, as well as the involving
forces. CFD models are very useful to design new wind turbines.
Finally, it is very important to mention the disadvantages of CFD. First of all, 3D
CFD models are usually very tedious due to the large computational resources. If a
moving mesh is employed, this is too computationally expensive. Other disad-
vantage is that CFD cannot not be applied blindly as it has the capability to produce
non-physical results due to erroneous modelling. The process of verication and
validation of a CFD model is necessary to ensure the numerical model accurately
captures the physical phenomena present. By comparing numerically obtained
results with experimental results, condence in the numerical model is achieved.
Once thoroughly validated, a numerical model may be used to accurately predict
the effect of design changes and experimentally unobservable phenomena.

Acknowledgements The authors would like to express their gratitude to Norplan Engineering,
S.L. & Technical Courses, for its support about the software
OpenFOAM and other CAD/CAE softwares, specially the following online courses: OpenFOAM
online course:,
C++ applied to OpenFOAM online course:
cursos_cha.php?curso_id=15, Rhinoceros marine design online course: http://www.
CFD Applied to Floating Offshore Wind Energy 87, SALOME online course:, Blender online


1. Versteeg HK, Malalasekera W (2007) An introduction to computational fluid dynamics. The

nite volume method. Pearson Education, Harlow
2. Anderson JD (1995) Computational fluid dynamics: the basics with applications. McGrawhill
Inc., New York
3. Anderson DA, Pletcher RH, Tannehil JC (1985) Computational fluid mechanics and heat
transfer. Hemisphere Publishing Co., London
4. Date AW (2005) Introduction to computational fluid dynamics. Cambridge University Press,
5. Ferziger JH, Milovan P (2002) Computational methods for fluid dynamics. Springer, Berlin
6. Lamas MI, Rodrguez JD, Rodrguez CG, Gonzlez PB (2011) Three-dimensional CFD
analysis to study the thrust and efciency of a biologically-inspired marine propulsor. Polish
Maritime Research 68:1016
7. Tobar MJ, Lamas MI, Yez A, Snchez-Amaya JM, Boukha Z, Botana FJ (2010)
Experimental and simulation studies on laser conduction welding of AA5083 aluminium
alloys. Phys Procedia 5:299308
8. Lamas MI, Rodrguez CG (2012) CFD analysis of the scavenging process in the MAN B&W
7S50MC two-stroke diesel marine engine. Journal of Ship Research 56(3):154161
Mooring and Anchoring

Ral Rodrguez Arias, lvaro Rodrguez Ruiz

and Vernica Gonzlez de Lena Alonso

Abstract Floating offshore wind turbines (FOWT) need to maintain their position
even during the most extreme events or energetic storms. The mooring and
anchoring systems are the responsible to provide the station-keeping of the FOWT
so, the effect of their behaviour affects directly to the dynamics of the floating
platform and it has to be carefully assessed during all the design stages. Although the
station-keeping of vessels and offshore structures has been carried out for centuries,
FOWT represent a relatively recent eld of application with specic requirements
and challenges. Existing experience from the offshore oil and gas industry, as for-
mulated in a number of design criteria and guidelines published by American
Petroleum Institute (API), International Organization for Standardization (ISO) and
classication societies, can be adapted for this specic application but taking into
consideration the existing differences. A general overview regarding mooring and
anchoring systems applied to FOWT is provided in this chapter, taking into account
the most usual system congurations and their associated types of anchor, likewise
materials. A range of requirements that should be considered for the design of
FOWT mooring and anchoring systems is included. Finally, methods for structural
analysis and the main specic mooring system guidelines are identied.

Keywords Mooring  Anchoring  Floating offshore wind

1 Introduction

Floating Offshore Wind Turbines (FOWT) need to be secured to the seafloor. The
purpose of a mooring and foundation or anchoring system is to provide offshore
equipment with a means of station-keeping that is sufciently robust to resist
environmental loading (e.g. tide, wind, wave, current and ice), impact and opera-
tional procedures.

R.R. Arias (&)  .R. Ruiz  V.G. de Lena Alonso

Centro Tecnolgico de Componentes, Isabel Torres 1, Santander, Spain

Springer International Publishing Switzerland 2016 89

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_6
90 R.R. Arias et al.

The function and type of mooring and anchoring system is determined by a

number of factors including: cost, site characteristics, expected environmental
loading and environmental or legislative constraints. In terms of cost, recent studies
have shown that the Levelized Cost of Energy (LCOE) values are strongly
dependent on depth and distance from shore, due to mooring costs and export
electrical cable length, respectively.
Although the station-keeping of vessels and offshore structures has been carried
out for centuries, FOWT represents a relatively recent eld of application with
specic requirements and challenges. Existing experience from the offshore oil and
gas industry, as formulated in a number of design criteria and guidelines published
by American Petroleum Institute (API), International Organization for
Standardization (ISO) and classication societies, can be adapted for application to
FOWTs, provided the differences between station-keeping systems for FOWTs and
those for floating oil and gas offshore structures are appropriately addressed.
In this chapter, a general overview of the mooring and anchoring systems which
are currently used is provided, taking into account the most usual system cong-
urations and their associated types of anchors, likewise materials. Also, a brief
picture of the design considerations since the functional point of view is provided.
The analysis methods and the main specic mooring and anchoring system guid-
ance are identied.

2 Mooring System Congurations and Types of Anchors

The objective of this section is to give a brief overview about the possible mooring
congurations and anchor types that can be applicable to FOWT. At the end of this
section, a guidance of the applicability of each mooring conguration and anchor
type is presented.

2.1 Conguration of the Mooring System

The most common mooring congurations are either catenary mooring systems,
which are used with spar-buoy and semi-submersible concepts either taut-leg
mooring systems, which are used with tension-leg platform (TLP) concepts. Some
concepts will also adopt a semi-taut mooring system, which is a mix between both
characteristics, even though this is less common. An overview of these congu-
rations is included in this section.
Mooring and Anchoring 91

2.1.1 Spread Mooring Systems

This system consists of several lines attached to the floating body, which limit the
horizontal movement (excursion) providing a damping to the system. These sys-
tems do not allow the floating body to turn or to auto orient itself to the prevailing
wind, wave or current directions (weathervane). They are often used in structures
whose response to external loads is not affected signicantly by its direction, such
as semi-submersible or spar type platforms. Usually the following types are

Catenary Mooring

Catenary mooring consists of several lines hanging freely as classic catenary from
the structure to the seabed. The lines reach the bottom horizontally so the anchor
point receives only horizontal loads. The restoring forces on the structure are
mainly due to the weight of the catenary and the scope (mooring length-to-depth
Over the past fteen years, many design concepts for FOWT have been pro-
posed. Since 2009, several small scale prototype FOWT have been installed for
eld testing and three full-scale FOWT have been deployed for concept demon-
strations, including the Hywind FOWT installed offshore Norway in 2009 [1, 2],
the WindFloat FOWT installed offshore Portugal in 2011 [35] or Fukushima
FORWARD in Japan [6].


This mooring consists of several lines which incorporate buoys and/or weights in
order to take more complex shapes. In this case the anchor point may receive
vertical loads.
The mooring lines of a free hanging catenary mooring arrive horizontal to the
seabed so that the anchor point is only subject to horizontal forces. So, a typical
catenary mooring system uses heavy chains that form a catenary shape from the
point of attachment to the seafloor where mean horizontal forces of the platform are
reacted. Catenary moored anchors have predominately horizontal mooring forces
and require less precision in their placement. Therefore, installation is less
expensive. Platforms using catenary mooring systems experience greater motions in
every direction compared to taut-leg or vertical tension-leg systems. The restoring
forces are mainly generated by the weight of the mooring lines returning the system
to equilibrium position. The biggest advantages of floating platforms with catenary
moorings are the relatively low cost of the anchors, and the potential to be deployed
in shallower water. The biggest problem is that the vertical tension of the anchor
92 R.R. Arias et al.

line is generally insufcient to maintain platform stability against overturning,

especially for a wind turbine where the weight and horizontal forces act quite far
above the centre of buoyancy [7].

2.1.2 Taut Mooring

In this conguration the mooring lines arrives straight on the seabed with a specic
angle (45 approximately) which means that they must be able to resist horizontal
and vertical forces (Fig. 1). The restoring forces are generated mainly by the
elasticity of the mooring lines. A particular case is the TLP, when the mooring lines
are orthogonal to the seabed (roughly vertical). In this conguration, the restoring
forces are generated mainly by the change in buoyancy of the topside structure [8].
Taut-leg mooring systems become advantageous over catenary systems as water
depth increases, because they have a smaller footprint and shorter mooring line is
needed. If the taut legs are installed in a vertical orientation, the footprint becomes
even smaller requiring even less mooring line, but high vertical anchors forces will
require more complex and costly anchors with limited anchoring options.

2.1.3 Single Point Moorings (SPM)

Single point moorings allow to floating structures to have weathervane behaviour.

A floating structure may be directly connected to the mooring system or to an
intermediate moored buoy. There is wide variety in the design of single point
moorings but they all essentially perform the same function. Examples of typical
single point mooring systems are described in the following points.

Catenary mooring

Multi catenary mooring

Vertical mooring

Taut-leg mooring

Fig. 1 Spread/taut mooring congurations

Mooring and Anchoring 93

Turret Mooring

A turret mooring system consists of lines that are attached as in a CALM or SALM
buoy system. The turret is attached to the floating structure through a bearing joint
or other linkage that allows relative yaw motion.

Catenary Anchor Leg Mooring (CALM)

A CALM system consists of a large buoy that supports a number of catenary

mooring lines. The floating structure is connected to the buoy by a single con-
nection point.

Single Anchor Leg Mooring (SALM)

A SALM system consists of a large buoy that supports a single taut vertical
mooring line. The buoy floatation induces tensions that tend to restore the buoy to
the vertical position. The floating structure is connected to the buoy by a single
connection point.

Articulated Loading Column (ALC)

In this case the structure is moored to a column that is xed to the bottom by a
hinge (Fig. 2).
The SWAY project [9] is the only concept that has used the single anchor leg
mooring. On the contrary, some wind turbines such as DeepCwind in USA [10] or
Blue H in Italy [11] have used TLP as the reference design for their mooring

Turret mooring

Catenary anchor leg mooring (CALM)

MOORING Single anchor leg mooring (SALM)

Articulated loading column ALC

Fixed tower mooring

Single point mooring

Fig. 2 Single point mooring congurations

94 R.R. Arias et al.

2.2 Anchor Types

The anchoring system consists mainly of an anchor and various ancillary elements
such us shackles, swivels, etc. used during the installation, maintenance and
decommissioning manoeuvres.
The anchors have been used for centuries, so there are a number of anchoring
solutions available in the market, depending on the mooring conguration, seabed
conditions and holding capacity required. Consequently, anchor selection will be
project and site specic, often dictated by the seabed conditions. A summary of the
main anchor types is detailed below, but there is great variety even within these
typologies. All are proven concepts which have been used extensively in the marine
and oil and gas industries and can be used in FOWT industry.

2.2.1 Gravity Anchors

It consist of elements of simple geometries such as parallelepipeds or truncated

pyramids made in concrete and/or steel (Fig. 3). They support horizontal and
vertical loads. Horizontal loads are generated due to the friction between the anchor
and the seabed, while vertical loads are generated with the apparent weight of the
anchor itself.
Over time, and usually in soft seabed, the anchor sinks into the seabed pro-
gressively increasing the forces that the anchor is able to withstand.
The advantages of this type of anchor are its low cost, also that, even if they
move from its original position to exceed its capacity (for example during a storm),
maintaining its capacity. Another advantage is its versatility, as it can be used in any
type of seabed.
One important disadvantage for the application to the floating offshore wind
market is the installation costs. The dimensions needed are large and their rate load
capacity/weight is usually less than unity.

Fig. 3 Gravity anchor [12]

Mooring and Anchoring 95

2.2.2 Drag Embedment Anchors

Drag embedment or fluke anchors were used initially in the offshore industry in no
permanent moorings. In recent times, the technology of this type of anchors has
suffered a considerable progress.
Drag anchors (Fig. 4) are elements with triangular geometry or similar, which
generate their holding capacity when they are buried into the seabed when a hor-
izontal load is applied. They have a large holding capacity/weight ratio but only for
horizontal loads due to they cannot support vertical loads.

2.2.3 Piles

Piles are long elements, with cylindrical tubular or double T geometry mainly
(Fig. 5). They are able to resist horizontal and vertical loads generated by friction
and lateral strength between the pile and the surrounding soil material. They are

Fig. 4 Drag embedment

anchor [12]

Fig. 5 Pile anchor [12]

96 R.R. Arias et al.

introduced into the seabed by a hammer or vibrators, being this task usually the
most limiting load condition in its design.
The principal dimensions are the length, the diameter and the thickness. The
holding capacity of the pile depends on the seabed friction between the surface of
the pile and the soil. It is interesting to highlight that this type of anchor is capable to
withstand both, vertical and lateral loads. The lateral load capacity can be increased
considerably by adding special elements such as ns on the top of the pile.
Its main advantage is the accuracy of the positioning and the large loads that they
withstand. Its main disadvantage is the impossibility of removal, limiting its
application in some countries.

2.2.4 Suction Piles

They are elements with elongated circular geometry, similar to a pipe and with an
opening at the bottom, where the soil penetrates (Fig. 6). They obtain their hori-
zontal and vertical capacity, as in the case of piles, by friction and lateral strength
between the surfaces of the pile and the surrounding soil. During installation, a
pump is used to create a vacuum inside the pile, penetrating the pile into the seabed
by the pressure difference.
The great advantage of this type of anchor is that, as the piles, their location can
be determined very accurately, which can be useful when several anchors are
needed per device.
One of the main problems of suction piles is that the optimum anchor point of
the line is commonly found below the midpoint of the height of the pile and is
necessary to have a special connector to get a vertical line. Another restriction in
their use is that they need a high water column in order to avoid cavitation problems
in the pumps.

Fig. 6 Suction pile [12]

Mooring and Anchoring 97

2.2.5 Plate Anchors

The plate anchors are elements of flat planar geometry usually with triangular or
rectangular shape and they can withstand horizontal and vertical loads (Fig. 7).
There are many variants of this type of anchors due to which each manufacturer has
his specic design.
The installation of the majority of them tends to be similar to drag anchors,
although in some designs the methodology is different. One example is the
driven-plate anchor, where a pile is used as a guide for the installation into the
seabed. After the pile is removed it is necessary to pre-tension the line.
Their main advantage is that they resist high forces in the vertical direction,
which makes them very useful for taut moorings. Depending on the specic design
the anchor can or cannot be removed from the soil, which can be a limiting factor in
their application.

2.2.6 Screw Anchors

This type of anchors consists of a steel bar on which a number of discs are xed
along a helix (Fig. 8). They literally work as a screw into the seabed. They
primarily resist loads applied in the axial direction. Screw anchors might be an
interesting alternative as an offshore foundation. The have the advantage that they
do not require highly specialized and costly equipment for their installation in
shallow water. The key point of the application of this technology is the devel-
opment of underwater machinery able to produce very high torque values [13].

2.2.7 Free-Fall Anchors

Free-fall or gravity installed anchors are installed as projectiles that penetrate the
sea floor under velocity (Fig. 9). The holding capacity of the anchor is generated by
the friction of the sediment and lateral sediment resistance. A capacity to resist both
out of plane lateral and vertical loads is possible depending on anchor design.
Generally, they apply to great depths, although some new designs are applicable to
shallower depths. In most of these anchors, the mooring system is tied at the top

Fig. 7 Plate anchor [12]

98 R.R. Arias et al.

Fig. 8 Screw anchor

anchor what simplies the installation manoeuvre. Its installation is more or less
simple, it is only necessary to place the anchor on the required coordinates and drop
it until it penetrates into the seabed. The main disadvantage of its use is that special
ships, such as AHV (Anchor Handling Vessel) or AHT (Anchor Handling Tug), are
needed for its installation.

2.3 Connectors

Connectors are ancillary elements used to connect different segments of mooring

lines or other objects such as anchors. There are several types or concepts but the
most used in the offshore industry are shown below.
Mooring and Anchoring 99

Fig. 9 Free-fall anchor [12]

2.3.1 Shackles

Shackle or D type connector is the most used connector in offshore industry in

general (Fig. 10). It consists of a body with two holes through which a bolt is
inserted. There are many types of shackles depending on the specic application.
Shackles can be used in permanent and temporary moorings systems.

2.3.2 Kenter

The Kenter connector is used for connecting two chain links of the same size
(Fig. 11). This type of connector is not used in permanent anchoring systems,
essentially because the fatigue life of the connector is less than the chain of
equivalent diameter.
100 R.R. Arias et al.

Fig. 10 Shackle D type

connector [14]

Fig. 11 Kenter [14]

2.3.3 Type H Connector

This type of connector is one of the most valued connectors by the offshore industry
because of its robustness and flexibility (Fig. 12). Not only can be used as an almost
universal connector, it can be used to adjust lengths during installation. It can also
be used as a Subsea Connector and can be designed with a strong lifting point for
installation and recovery.

2.3.4 Pear-Shaped Connector

This type of connector is similar to the Kenter connector type, but it is intended to
connect two links of different dimensions. Like the Kenter connector, these con-
nectors are not used in permanent anchoring systems.
Mooring and Anchoring 101

Fig. 12 H type connector


2.3.5 Type C Connector

This type of connector is similar to the Kenter, but while in the Kenter type the
assembly is performed in the axial direction, in this type of connector the assembly
is performed in the transverse direction to the direction of work. This connector is
not used in permanent anchoring systems.

2.3.6 Swivel

This type of connector is used to prevent twisting and bending of the lines by
enabling some degrees of freedom (Fig. 13). It is usually mounted near the
anchoring system (a few links) but it is also allowable to attach a chain with a bre
rope. This kind of connectors has an important drawback, a large friction occurs in
its internal mechanism due to high loads. There are some special designs that avoid
this problem by using special surfaces as bearings.

Fig. 13 Swivel [14]

102 R.R. Arias et al.

2.3.7 Ballgrah Connector

The Ballgrab connector generates a detachable connection capable of transmitting

axial and torsional stresses. It consists of two conical surfaces, one of which has
spheres. By means of friction it is possible maintaining the joint.

2.3.8 Delmar Connector

Delmar connector is usually used to link anchors to mooring systems. The par-
ticularity of this type of connector is its simplicity and versatility, and can be used
in permanent moorings.

2.4 Guidance of Mooring/Anchor Application

The selection of the mooring system conguration is the most important step before
designing a mooring system as it provides a rst lter of the loads.
As a general rule, in the case of catenary mooring, all kinds of anchors can be
applied. For taut moorings, inclined or vertical, the vertical forces should be con-
sidered with extremely careful, because it is usually the critical parameter.
Table 1 shows a guidance for the selection of the best anchor type in terms of the
type of mooring conguration.

Table 1 Guideline of mooring/anchor application

Type Drag anchors Piles Suction piles Plate anchors Screw piles
Applicable Applicable but Applicable but Applicable but Applicable
Catenary but not the not the best not the best not the best but not the
best choice choice choice choice best choice
Applicable No Applicable
Recommended Recommended Recommended
Taut-leg but not the recommended but not the
(3) (3) (3)
best choice (2) best choice
Applicable No Applicable but Applicable but Applicable
Vertical but not the recommended not the best not the best Recommended but not the
best choice (2) choice (4) choice (4) best choice

(1) Among all the options to moor a structure, and if a catenary system is selected, the best option
is a traditional anchor due to its low cost
(2) In the case of taut systems, it is necessary anchors which withstand vertical loads, this is the
reason why traditional anchors are not applied
(3) When a taut system is applied, the best choices are the piles, the suction anchors and plate
(4) In these cases the use of a ROV for penetrating and dismantling work is necessary, which
increases costs. For permanent moorings can be a cost effective solution
Mooring and Anchoring 103

3 Materials and Future Prospects

As commented previously, a mooring system consists of a mooring line, connectors

and an anchor. Mooring lines are usually composed by a combination of metallic
chains, metallic wires or synthetic ropes. Chains are typically the rst choice due to
their low cost and reliability. Synthetic lines, however, are used for special cases,
for example when low weights are required.
In this section, a classication of the main components of a mooring system
according to its construction materials is shown.

3.1 Mooring Lines

A classication of the mooring lines according to their construction material is as

follows (Fig. 14).

3.1.1 Metallic Lines

In this section, a description of the most widespread metallic mooring line types is
explained below.


Steel chains are widely used because of their proven durability in offshore opera-
tions. They have very good resistance to seabed wear and contribute signicantly to
the anchor holding capacity, but as the depth increases, the weight of the chain
required becomes prohibitive, and lighter mooring lines must be used.
Chains are particularly suitable for long-term mooring systems but it is necessary
periodic inspections and maintenance to keep them clean from biofouling and
marine corrosion.

Fig. 14 Line types according

Line Types
to construction materials

Metallic Non Metallic

Chains Synthetic Fibers

Wire ropes Elastomeric lines

104 R.R. Arias et al.

Fig. 15 Stud-less common link (left), Stud common link (right) [14]

Fig. 16 Mechanical properties [14]

At present, two types of chain are commonly employed (Figs. 15, 16):
Studless link: Maintenance, inspection costs and weight can be reduced; how-
ever, the fatigue life can be comparable to half of a studded chain.
Stud link: are preferable in terms of ease of handling and are considered to have
a higher reliability than studless link chains.
Depending on the nominal tensile strength of the steels used for manufacture,
chains are to be subdivided into ve grades: R3, R3S, R4, R4S and R5, although
other grades such as R6 or R7 are being developed at present with a tensile stress of
1400 MPa. In general, R3, R3S and R4 are provided for offshore mooring systems,
whereas grade R5 or higher are characterized by the highest proof tensile strength.


Ropes made of metallic wires are extensively used as mooring line components.
They are easy to handle, and their cost is relatively low. Wire ropes are charac-
terized by a high elasticity, so that they are usually used in tensioned mooring
applications. In general, wire ropes provide more restoring force in deep water than
chains and require lower pretensions. Also, they are signicantly more lightweight
than chains.
Mooring and Anchoring 105

Fig. 17 Typical wire ropes [15] a spiral strand b six strand c multi-strand

The most common used wire ropes are (Fig. 17):

Spiral strand is the simplest rope constructions. It consists of concentric helical
layers of wires. The outer layers of a spiral strand, which constitute the bulk of
the cross section, generally have wires of the same diameter, disposed in
opposite helical senses and with the same or similar helix angle. The outer
surface is cylindrical, so that it facilitates the cover by polymeric jacket to
provide long-term protection.
Six strand. Six or eight strand ropes can be twisted around a core to form the
strands (bre or wire rope cores are commonly used). Since the outer layer is not
cylindrical, it is not possible to cover with a polymeric jacket, so these ropes are
not generally selected for very long-term exposure. This rope has also good
resistance to torsion loads.
Multi-strand ropes have two or more layers of strand. These ropes are com-
monly used in offshore for applications requiring bending flexibility and high
tensional and torsion resistance. Multi-strand ropes are more expensive than
conventional six strand ropes of similar strength.

3.1.2 Non-Metallic Lines

These ropes do not corrode nor deteriorate appreciably in sea water. Their strength
to immersed weight ratio is excellent. They are easy to handle and terminate,
however, are relatively easy to cut, especially when they are under tension.

Synthetic Fibres

An alternative solution is synthetic bre ropes (Fig. 18). Contrariwise from chains,
where the resistant force is due to their weight, the synthetic lines offer a resistance
which depend on their elastic characteristics. The weight and elasticity properties of
the ropes make them more common for very deep water tether applications.
106 R.R. Arias et al.

Fig. 18 Polyester Rope (left) and Polypropylene Rope (right) [16]

The elastoplastic response of the synthetic line is a complex phenomenon. First, a

permanent deformation occurs at the rst loading, which stretches the bres for the
rst time; an important nonlinear elastic response is present at each successive loading,
mixed to a plastic deformation which is due to a microscopic re-aligning of the bres
and that dissipates energy. Such behaviour can be described by the axial stiffness
EA [N], which gives the proportionality constant in the stressstrain relation,
being E the elastic modulus and A the cross section of the line. It has to be pointed out
that the elastic modulus is not constant and depends on the past load history.
A synthetic line results from the assembly of different components in a hierar-
chic structure which follows the production line, starting from the bre that com-
poses the line. The four basic constructions of bre ropes are:
Stranded. Most common construction, rope generally made of three strands
running in the same direction (left or right).
Plaited. Made generally of an even number of strands, plaited or twisted, in pairs
of opposite direction.
Braided. Made of a great number of equally distributed left- and right-hand yarns.
Parallel yarns. Made of a large number of yarns held together by a minimum of
twist and enclosed in a protective plastic jacket or external braid.
Most common materials are:
Nylon. A large percentage of synthetic bre ropes are made of nylon. Nylon
ropes are highly elastic, have excellent endurance under repeated loadings, and
have excellent energy absorption capacity.
Polyester. Are less elastic than nylon ropes and absorbs less energy. Their
endurance is very good, and their resistance to abrasion and chemicals are
Polypropylene. It is the lightest material used in ropes. Polypropylene ropes
elongate less than nylon, but more than polyester ropes.
Polyethylene. Polyethylene ropes are not extensively used in mooring applica-
tions. The initial and elastic stretch characteristics of polyethylene ropes are
approximately the same as the characteristics of polypropylene, however,
polyethylene creeps at a much faster rate than any other synthetic.
High modulus organic bres (Kevlar). Ropes manufactured with this material
have excellent characteristics, are light, and their strength can nearly equal the
strength of steel cables of the same diameter.
High density polyethylene (Dyneema). This material has higher strength,
denser and stiffer than high modulus organic bres.
Mooring and Anchoring 107

Elastomeric Lines

The main characteristic of an elastomeric line is that its elongation under load is
signicantly higher than the elongation of the traditional mooring lines; chains, wire
ropes and even bre ropes.
The mooring line is selected by the appropriate length, diameter of cord and
hardness of the rubber used. The simplest approach is one or more natural rubber
cords. The properties of natural rubber such as tear and abrasion resistance as well
as its creep properties, high tear strength and excellent elongation properties make it
a suitable candidate for mooring. Other rubber mooring cords use composition
rubber including both, natural rubber and synthetic rubbers designed to optimize
performance and reduce costs.
The light weight compared to other mooring lines can also make handling during
deployment much easier; on the other hand they are more susceptible to cutting and
breaking than their mooring line counterparts.
Table 2 summarizes the mooring line characteristics of the main materials.

Table 2 Mooring line summary

Material Features Comments
Chain Broad use experience Unsuitable for water depths greater
Readily available than about 450 m
Susceptible to corrosion
Good abrasion resistance
Steel Wire Rope Broad use experience Unsuitable for water depths greater
Readily available than about 900 m
Susceptible to corrosion
Polyester High dry and wet strength Most durable of all bre line
Moderate stretch materials
Frequent use in deep water Moderate cost
taut moorings
Nylon High dry strength Wet strength about 80 % that of dry
High stretch Low fatigue life
Moderate cost
Polypropylene and Low weight Low strength
Polyethylene High stretch Low melting point
Susceptible to creep
Low cost
HMPE Low stretch Replacing wire for towing-increased
High strength to weight handling safety
ratio High cost
Aramid Very low stretch Minimum bending radius similar to
High strength to weight steel wire rope
ratio Low abrasion resistance
High cost
Elastomer Low weight Susceptible to cutting and breaking
High elongation capacity
High tear strength
108 R.R. Arias et al.

3.2 Anchoring Materials

Steel is the main material used in the manufacturing of anchors. Other materials are
also used, such as aluminium or concrete but in a lesser extent.

3.2.1 Steel

Steel is the most widespread material when manufacturing anchors. The selection of
the suitable steel is very important, this selection can be carried out taking into
account the manufacturing process, as each provides different properties and
Pile, suction pile, plate and screw anchors are mostly manufactured with lami-
nated steel. On the other hand, drag embedment anchors uses laminated and casted
steel. Gravity anchors can be made of steel although is not the typical material.

3.2.2 Other Materials

There are other materials that could be used for the design of anchoring systems.
Among this group of materials highlights the concrete, due to its low cost is used in
gravity anchors and sometimes even in suction caissons. Aluminium is used almost
exclusively in connectors.

3.3 Future Prospects

The trending materials in the mooring systems for FOWT should be stronger and
lighter than common steels. Synthetic materials such as polyamide and polyester are
identied as the best candidates for the application in near future. This is supported
because of taut mooring concepts can reduce the footprint area and length of
mooring lines which could reduce the weight and costs. On the contrary, elastomers
can be a revulsive to reduce peak loads.

4 Requirements and Design Considerations

A mooring system is used to conne the floating structure to a specic location and
the anchoring system secures the mooring line to the seafloor. In previous sections,
the different mooring congurations have been presented. For the different FOWT
concepts, the more relevant mooring systems are: catenary mooring systems, ver-
tical mooring systems and taut-leg mooring systems [17].
Mooring and Anchoring 109

The major difference between these systems is that catenary mooring lines arrive
to the seabed horizontally, vertical mooring lines arrive to the seabed vertically and
taut-leg mooring lines arrive at an angle. This difference is the primary design
consideration: the taut leg mooring system is capable of resisting both horizontal
and vertical forces, and restoring forces are generated by elasticity of the mooring
lines. The vertical mooring system is also capable of withstanding both, horizontal
and vertical loads, although horizontal loads are not withstood to the same extent.
For catenary mooring, the anchor point is only subjected to horizontal forces, and
most of the restoring forces are generated by the weight of the mooring line. In
terms of the footprint, the catenary mooring has a clear disadvantage compared to
the other systems as its mooring radius is bigger with a similar application, affecting
total wind farm seabed area [18].
The load capacity of any anchor system depends mainly on the bottom soil
conditions (shear strength and weight) and on the direction of the applied force. The
shear strength of the soil is the primary mechanism for resisting the forces applied.
The weight of the soil is also a major factor. The deeper the anchor can be
embedded, the greater the quantity of affected soil, and hence, the greater the
holding capacity. Finally, if the applied force is parallel to the bottom, an anchor
can be very effective without deep embedment because as resisting forces are
applied, it digs deeper into the seabed [19].

4.1 Functional Requirements

The choice of mooring line depends on a number of technical parameters such as the
type of mooring systems, water depth, seabed characteristics, desirable footprint, and
excited loads and required motion characteristics of the floating structure (allowable
horizontal excursion, natural periods, maximum accelerations and displacements).
As mentioned previously, there are several types of anchors foundations of
relevance for anchoring FOWT units, namely: gravity anchors, fluke anchors, piles,
suction anchors, plate anchors, screw anchors, plate anchors or free-fall anchors.
The selection and applicability of each anchor type has been studied in the literature
in terms of the mooring conguration, the direction and variation of the loading, the
type of soil, cost, installation and removal and environmental impact [1923].
For a FOWT the anchoring and mooring system shall be designed such that a
sudden failure of any single mooring line will not cause progressive failure of the
remaining lines. Other key design consideration for the mooring and anchor system
is to maintain the position of the FWOT under extreme loading conditions while
allowing efciency in normal operational conditions. Special load cases are
investigated in detail and the results for the loads are obtained from frequency or
time domain simulations. The design of the station-keeping system should consider
all relevant conditions to pre-service operations during installation and commis-
sioning, in-service conditions including operations, maintenance and repair
110 R.R. Arias et al.

The tensions experienced by a mooring system at any time are driven by the
following parameters:
Static component of the force on the FOWT from wind, mean wave drift and
Wave frequency component, caused by rst-order wave frequency motions and
drag/inertia effects on the mooring line,
Low frequency component, due to second-order low frequency waves and wind

4.2 General Design Considerations

In general, a FOWT mooring system should be designed to:

Maintain the position of the FOWT within a specied limit from its reference
Control the directional heading of the FOWT if the orientation is important for
safety or operational considerations.
Assist in maintaining the acceleration and the tilting angle at the tower top
within a specied limit.
When considering the design load acting on the anchor and arising from the line
tension in a mooring line, which is hooked up to the anchor, this shall be taken as
equal to the design line tension in the mooring at the interface between the mooring
line and the anchor, as resulting from calculations according to the specications
provided in the relevant design standards. When more than one mooring line is
hooked up to the anchor, the design force shall be calculated with due considera-
tions of the design force contributions from all mooring lines. In selecting the
appropriate anchor type, consideration must be given to the mooring system con-
guration and design characteristics (load, out of plane lateral, and uplift),
site-specic seabed conditions, direct loading from current and wave action, and
installation constraints. The above considerations as well as the site specic risk
prole will dictate anchor selection.
It is not the intention of this section to detail current offshore position mooring
and anchoring guidelines. Instead a summary of the more relevant general design
considerations is provided. The following list shows a range of requirements that
should be considered for FOWT mooring and anchoring systems:
The mooring must maintain the FOWT on station during normal operating
conditions and survive extreme storm conditions and must be designed to
accommodate tides.
Mooring anchors must be designed to accommodate hazardous conditions such
as seabed sediment movement, earthquake, cables and pipelines.
Mooring and Anchoring 111

Unless otherwise stated, the characteristic anchor resistance is dened as the

mean anchor resistance as set up by the supporting soils or rock. For all anchors,
the design load and holding capacity shall be clearly dened for all limit state
conditions. Holding capacity is the maximum force that can be resisted by the
anchor and can be determined from anchor manufacturer or standard design
tables and semi-empirical models.
All components must have adequate strength, fatigue life and durability for the
operational lifetime. The design life should be dened according to the appli-
cable standards [2426].
Corrosion must be considered and controlled in the design, and abrasion due to
bottom contact or contact with other lines must be minimized or avoided if
Marine growth is to be considered in the design of mooring components.
Particular attention is to be paid to increases in hydrodynamic loading due to
increased diameters and surface roughness of members caused by biofouling as
well as to the added weight and increased inertial mass. The types of marine
growth likely to occur and their possible effects on corrosion protection coatings
are also to be considered.
The mooring and anchoring must be designed to minimize environmental
impacts on the seabed or native flora and fauna.
The mooring systems should be designed to keep the platform and wind turbine
devices at optimum orientation relative to the environmental conditions (wind,
waves and currents).
The mooring system must not allow twisting or over tensioning of the electrical
transmission cable.
The mooring system design should not adversely affect the performance of the
FOWT device (i.e. natural periods).
The mooring system must be designed with sufcient compliance to achieve a
reasonable balance between anchor and mooring line loads and station-keeping
Special considerations have to be given to the degradation mechanism in the
mooring lines [27]. Many of the mooring issues mentioned in this reference refer to
chain. This is because chain is normally selected at the two most challenging
locations, namely the platform interface and the seabed touchdown; since the
loading regime is severe degradation may sometimes occur. However, experience
over the years has shown that using wire in these areas does not give a true
long-term solution. The same would almost certainly apply to the use of bre ropes.
Various sources of potential failure were identied, namely: overload/overstress,
fatigue in a catenary (at a sheave or connection), brittle fracture, corrosion, wear and
abrasion and mechanical failure of the mooring line handling/lock off system.
112 R.R. Arias et al.

5 Analysis Methods and Associated Reference Standards

There are several standards all over the world in the eld of floating structures, but
not too many dedicated to FOWT. There are no standards at international level
although at national level DNV-GL [28] and ABS [29] have started the process of
publishing standards that cover FOWT, and specically for moorings.
The international standard for offshore applications is the ISO 19900 series,
addressing offshore platforms for the oil and gas industry. These standards were
based on previous API standards for xed and floating steel structures. Detailed
information about applicable standards is in Appendix 1 of this chapter.
At present new standards are being developed in order to dene the best
methodologies to specify the particular requirements of FOWT. Some research
projects in the eld of marine energies have dened the particular requirements and
guidelines for the design of marine energies [30, 31].
Mainly three different approaches can be followed when the structural design of
a floating turbine is carried out: (1) design by partial safety factor, (2) design
assisted by testing and (3) probability-based design. Most used and recommended
procedures are 1 and 3, due to testing is very costly and uncertainties of different
scaling laws always exist. These two methods will be explained below.
In all cases, standards dene different design load conditions. DNV uses the
name of limit state and includes the following:
Ultimate Limit State: Used to ensure that components have adequate strength to
withstand loads resulting from extreme environmental conditions.
Accident Limit State: Used to ensure that the mooring system is capable of
withstanding the failure of one mooring line, assuming that the system has
built-in redundancy.
Fatigue Limit State: Used to ensure that individual mooring lines can withstand
cyclic loading.

5.1 Design by Partial Safety Factor

The partial safety factor method is a design method by which the target safety level
is obtained as closely as possible by applying load and resistance factors to char-
acteristic values of the governing variables, and subsequently fullling a specied
design criterion expressed in terms of these factors and these characteristic values.
The governing variables consist of:
Loads acting on the structure or load effects in the structure.
Resistance of the structure or strength of the materials in the structure.
The safety level of a structure or a structural component is considered to be
satisfactory when the design load effect Sd does not exceed the design resistance Rd.
Mooring and Anchoring 113

Sd  Rd

In turn, these values are multiplied by different coefcients to increase the loads
and reduce the resistance of materials. These coefcients vary with the load con-
dition, the consequence of the failure and other factors such as the possibility of
inspection the critical areas. Typical values for the magnication of the loads are
1.31.35 and the reduction of resistance is 0.90.95. If the failure triggers a big
impact for persons or other objects, the coefcients will be higher.
Another important element in the design of a mooring system is the numerical
model used. In terms of modelling mooring lines in a numerical analysis two
alternatives exist, commonly denoted quasi-static or dynamic modelling. In brief
one may say that the mooring system in a quasi-static analysis is modelled as
nonlinear springs, and hence does not include mass and drag forces acting along the
length of the mooring lines. In a dynamic analysis the mooring lines are modelled as
slender elements so that mass and drag forces acting along the length of the line are
included. The need to use a dynamic mooring line model in the analysis increases
with the water depth and with nonlinearity behaviour of the floating platform.

5.2 Probability-Based Design

In all kind of structures there is always an inherent risk that the structure may fail
that is an unexpectedly loss of the properties provided during the design and
manufacture stages. To estimate this risk, it is necessary to know two fundamental
variables. The rst is the probability of occurrence of a failure and the second is
based on the consequence associated with that failure.
The risk category for each failure mode should be identied. The failure modes
with high or medium risk must be investigated thoroughly, and should be dened as
failure modes of interest. The low risks should not be ignored, since these can be
addressed in later stages of design.
There are different methodologies to assess the risk, being the ALARP (As Low
As Reasonably Practicable) one of the most interesting. The main objective of this
methodology is to reduce risks to reasonable levels. This level of reasonableness
is dened on the basis of economic losses, human lives and environmental damage,
making a balance between the cost required to reduce the risk and the benet
There are three different thresholds, but the most restrictive is the unacceptable
risk. In this case there are four ways to proceed:
1. Risk mitigation: Risk mitigation is implemented with the modication of the
system, since the element generator of risk is eliminated.
2. Risk reduction: Risk reduction can be implemented by reducing the conse-
quences and/or probability of occurrence. In practice, which is usually done is a
physical modication of the system considered.
114 R.R. Arias et al.

3. Risk transfer: The risk transfer is based on the assumption of risk by a third
party. Usually, through insurance or nancial agreements.
4. Acceptance of risk: If the risk is not in accordance with the relevant criteria, this
can be accepted with their possible consequences.

5.3 Reliability Analysis

The deterministic approach does not take into account probabilistic information,
and uncertainties in the system models. Engineering structures under operation are
influenced by signicant uncertainty from various sources, namely the loading, and
the material properties. These uncertainties and probabilistic information are
important in many design scenarios including designs requiring satisfactory per-
formance for a certain lifespan.
An implication of the probabilistic approach is that some likelihood remains that
the structure will fail under operation. It is generally accepted that for efciently
designed structures, this likelihood is neither too large nor too small. The quan-
tication of this likelihood, i.e. of the failure probability, is the objective of relia-
bility analysis.
The probability of failure is computed by Pf = N() where is the reliability
index being a useful indicator to compute the failure probability. There is a rela-
tionship between probability of failure and the reliability index (Fig. 19).
Typical values for the target annual probability of failure: DNV [28] standard
denes a value of 1 104 whilst IEC [26] selects 2 104.
Generally, the main steps in a reliability analysis are:
1. Selection of a target reliability level.
2. Identify the signicant failure modes of the structure.

Fig. 19 Representation of the reliability index and the relation with probability of failure
Mooring and Anchoring 115

3. Decomposition of the failure modes in series systems of parallel systems

(only needed if the failure modes consist of more than one component).
4. Denition of failure functions (Limit State Functions) corresponding to each
component in the failure modes.
5. Denition of stochastic variables and the deterministic parameters (distribution
types and statistical parameters for the stochastic variables).
6. Estimate the reliability of each failure mode. In a reliability analysis, the reli-
ability is compared with the target reliability.
7. Evaluate the reliability result by performing sensitivity analyses.
To proceed with these methods, stochastic variables and processes must be
introduced and failure and non-failure behaviour of the structure must be decided.
Codes and standards provide guidelines for target reliability levels based on the
purpose of the structure, the nature of failure and the consequences of failure.

5.4 Connectors

There are no specic standards dedicated for connectors while they must designed
to take the full breaking strength of the mooring line. The number of connectors
should be minimized for safety, fatigue and operational and maintenance consid-
erations. It should be noted that some of these items, such as Kenters or swivel
shackles, may not be suitable for long-term mooring in excess of 12 years due to
poor fatigue performance.

6 Discussion

In this chapter, the importance of the mooring and anchoring systems when
designing a FOWT has been emphasized. Each section could be expanded into a
single chapter, but give the scope of this book, the main objective is to provide a
starting point for those who are not familiar with these systems in the floating
offshore wind eld.
In Sect. 2 a description of the main mooring systems is done since the point of
view of geometric conguration, single point mooring and spread mooring, as well
as, the different mooring congurations within each. At the same time, a brief
description of the most usual anchoring elements is described.
Section 3 aims to give an overview of the different materials that can be found in
the eld of mooring lines and anchoring materials.
Section 4 gives insight to the main design requirements and design considera-
tions from the functional point of view. It has been pointed out that the design of the
station-keeping system should consider all conditions relevant to pre-service
operations, during installation and commissioning, in-service conditions including
116 R.R. Arias et al.

operations, maintenance and repair operations. The mooring system has an

influence in the dynamic response of the FOWT, so its design should not adversely
affect the performance of the device. Special consideration is required to the dif-
ferent degradation mechanism which should affect the structural integrity of the
components of the mooring lines and anchors.
Finally, in Sect. 5, which is directly linked to Sect. 4, design methods and
reference standards have been explained.

Appendix 1: Mooring and Anchoring Guidelines

Standard Year Issuing organisation

Guidance notes on the application of bre rope for 2014 American Bureau of Standards
offshore mooring (August 2011 and updated (ABS)
February 2014)
Floating offshore wind turbine installations (January 2014 American Bureau of Standards
2013 and updated July 2014) (ABS)
Design guideline for station-keeping systems of 2013 American Bureau of Standards
floating offshore wind turbines (June 2013) (ABS)
Design and analysis of station-keeping systems for 2008 American Petroleum Institute
floating structures (API RP 2SK) (API)
Recommended practice for design, manufacture, 2014 American Petroleum Institute
installation and maintenance of synthetic bre ropes (API)
for offshore mooring (API RP 2SM)
Mooring chain (API Spec 2F) 2010 American Petroleum Institute
Classication of mooring systems for permanent 2012 Bureau Veritas
offshore units (NR 493 DT R02 E)
Certication of bre ropes for deepwater offshore 2007 Bureau Veritas
services (NI 432 DTO R01E)
Rules for the classication of offshore loading and 2006 Bureau Veritas
offloading buoys (NR 494 DT R02 E)
Classication and certication of floating offshore 2010 Bureau Veritas
wind turbines (NI 572 DT R00 E) November 2010
Position mooring (DNVGL-OS-E301) 2015 DNV GL
Offshore mooring chain (DNV-OS-E302) 2013 DNV GL
Offshore bre ropes (DNV-OS-E303) 2013 DNV GL
Offshore mooring steel wire ropes (DNV-OS-E304) 2009 DNV GL
Design and installation of fluke anchors 2012 DNV GL
Design and installation of plate anchors in clay 2002 DNV GL
Geotechnical design and installation of suction 2005 DNV GL
anchors in clay (DNV-RP-E303)
Mooring and Anchoring 117

Standard Year Issuing organisation
Environmental conditions and environmental loads 2010 DNV GL
Design of floating wind turbine structures 2013 DNV GL
Certication of tidal and wave energy converters 2012 DNV GL
Guidelines on design and operation of wave energy 2005 DNV and EMEC
Guidelines for marine energy converter certication 2009 European Marine Energy Centre
schemes (EMEC)
Guidelines for design basis of marine energy 2009 European Marine Energy Centre
conversion systems (EMEC)
Guidelines for reliability, maintainability and 2009 European Marine Energy Centre
survivability of marine energy conversion systems (EMEC)
Technical policy boardguidelines for moorings 2010 GL Noble Denton
Floating production system JIP FPS mooring 2006 GL Noble Denton
Guidelines for the survey of offshore mooring chains 2010 International Association of
cable in use Classication Societies (IACS)
International standard. Wind turbines. Part 3. Design 2009 International Electrotechnical
requirements (IEC-61400-3) Commission (IEC)
Marine energyWave, tidal and other water current 2015 International Electrotechnical
convertersPart 10: Assessment of mooring system Commission (IEC)
for Marine Energy Converters (MECs). (IEC TS
Petroleum and natural gas industries: Specic 2013 International Standards
requirements for offshore structures. Part 7: Organisation (ISO)
Station-keeping systems for floating offshore
structures and mobile offshore units
Shipbuilding and marine structures: Mooring 2012 International Standards
winches (ISO3730:2012) Organisation (ISO)
Fibre ropes for offshore station-keeping: Polyester 2007 International Standards
(ISO18692:2007) Organisation (ISO)
Fibre ropes for offshore station-keeping: High 2012 International Standards
modulus polyethylene (HMPE) Organisation (ISO)
Ships and marine technology: Stud link anchor 2008 International Standards
chains (ISO1704:2008) Organisation (ISO)
N 1066: The design of floating aid to navigation 2010 International Association of
moorings Lighthouse Authorities (IALA)
N 1040: The maintenance of buoys and small aids 2008 International Association of
to navigation structures Lighthouse Authorities (IALA)
Rules and regulations for the classication of a 2013 Lloyds Register
floating offshore installation at a xed location
Guidance on offshore wind farm certication: 2012 Lloyds Register
Design, build and operational requirements
118 R.R. Arias et al.


1. Bratland S (2009) Powering the futuremarine energy opportunities. Hywindthe world rst
full-scale floating wind turbine. Seminar and B2B meetings (presentation)
2. Thygeson JS (2010) Hywind demo construction and installation. Offshore craneslifting
conference (presentation)
3. Andrus C (2011) WindFloat presentation. European wind energy conference and exhibition
4. Aubault A, Cermelli C, Lahijanian A, Lum J, Peiffer A, Roddier D (2012) WindFloat
contraption: From conception to reproduction. Paper presented at the 31st international
conference on offshore mechanics and arctic engineering, Rio de Janeiro, Brazil, Paper
No. OMAE 201-84036
5. Cermelli CA, Roddier DG, Weinstein A (2012) Implementation of a 2 MW floating wind
turbine prototype offshore Portugal. Offshore technology conference (OTC), Houston, Paper
No. OTC 23492
6. Fukushima Offshore Wind Consortium (2011) Fukushima floating offshore wind farm
demonstration project (Fukushima FORWARD)
pamphlet3.pdf. Accessed 15 May 2015
7. Musial W, Buttereld S, Boone A (2004) Feasibility of floating platform systems for wind
turbines. In: Proceedings of 23th ASME wind energy symposium, Reno, Nevada, USA
8. Harris RE, Johanning L, Wolfram J (2004) Mooring systems for wave energy converters: a
review of design issues and choices. 3rd international conference on marine renewable energy,
Blyth, UK
9. SWAY wind turbine (2015) Accessed 13 May 2015
10. DeepCwind Consortium (2013)
deepcwind-consortium/. Accessed 13 May 2015
11. Blue H Engineering (2012) Accessed 13 May 2015
12. IEC (2015) IEC 62600-10:2015. Marine energyWave, tidal and other water current
convertersPart 10: Assessment of mooring system for marine energy converters (MECs)
13. Spagnoli G (2013) Some considerations regarding the use of helical piles as foundation for
offshore structures. Soil mechanics and foundation engineering, vol 50, No.3
14. Vicinay Cadenas S.A. (2015) VCSA Catalogue
15. Chaplin CR (1998) Torsional failure of a wire rope mooring line during installation in deep
water. Eng Fail Anal 6:6782
16. WireCo World Group (2015) Lankhorst Euronete Rope Catalogue
17. Bjerkseter C, gotnes A. (2013) Levelised cost of energy for offshore floating wind turbine
concepts. Department of Mathematical Sciences and Technology, University of Life Sciences,
p 206
18. Vryhof Anchors BV (2010). Anchor manual 2010the guide to anchoring
19. Rodriguez R, Gorrochategui I, Vidal C, Guanche R, Caizal J, Fraguela JA, Diaz V (2011)
Anchoring systems for marine renewable energies offshore platforms. OCEANS 2011
IEEE-Spain. Issue Date: 69 June 2011. Print ISBN: 978-1-4577-0086-6. doi:10.1109/
20. Weller S, Hardwick J, Johanning L, Karimirad M, Teillant B, Raventos A, Baneld S,
Delaney M, Yeats B, Sheng W, Crawford T, Roberts J, Bull D, Grifth T (2014)
A comprehensive assessment of the applicability of available and proposed offshore
mooring and foundation technologies and design tools for array applications. Deliverable
4.1 DTOceanoptimal design tools for ocean energy arrays
21. Myhr A, Bjerkseter C, gotnes A, Nygaard T (2014) Levelised cost of energy for offshore
floating wind turbine concepts. Renew Energy 66:714728
22. Zanuttigh B, Martinelli L, Castagnetti M (2012) Screening of suitable mooring systems.
Deliverable 2.1 SDWEDstructural design of wave energy devices
23. Carbon Trust (2015) Floating offshore wind: market and technology review
Mooring and Anchoring 119

24. DNV-0S-J103 (2013) Design of floating wind turbine structures

25. ABS (2013) Design guideline for stationkeeping systems of floating offshore wind turbines
26. IEC (2009) IEC 61400-3:2009 Wind turbinespart 3: design requirements for offshore wind
27. Noble Denton Europe Limited for the Health and Safety Executive (2006) Floating production
system JIP FPS mooring integrity
28. DNVGL-OS-E301 (2015) Position mooring
29. ABS (2014) Floating offshore wind turbine installations
30. Heath J (2014) Specic requirements for MRE foundation analysis. Deliverable 4.2
DTOceanOptimal design tools for ocean energy arrays
31. Weller S, Johanning L, Davies P (2013) D3.5.3 Best practice report mooring of floating marine
renewable energy devices. MERIFICmarine energy in far peripheral and island
Resource Assessment Methods
in the Offshore Wind Energy Sector

N. Salvao and C. Guedes Soares

Abstract Offshore wind industry has experienced a large development over the
past decades. Some key factors in the development, sitting and operation of an
offshore wind farm include the accurate estimation and forecast of the wind
resources and the quantication of the inherent variability in wind power genera-
tion. Wind resource estimates are characterized by various degrees of uncertainties
that could lead to highly misleading results. Most often, risk-based nancial models
on which wind project investments are based, are strongly dependent upon these
uncertainties, which constitute a barrier for wind energy penetration into the elec-
tricity grid. On that account, for a correct and reliable wind power assessment it is
of utmost importance to increase the quality and quantity of wind data available. An
accurate estimation of wind eld is a challenging task that requires reliable sources
of data. Offshore meteorological observations from meteorological masts or marine
buoys constitute the most commonly used source of data to build energy density
maps due to its large reliability. Nevertheless, in situ measures are often sparsely
located, not available where the observations are most needed, and have a poor
temporal coverage. To overcome these limitations, the wind eld may be obtained
from a variety of alternative methods available for energy assessment, from satellite
remote sensing observations to numerical weather prediction models. This paper
gives an overview of the available methods for addressing the wind resource and
overall development potential of a candidate site together with a review of the
statistical methods to deal with variability and long-term extrapolation of wind
speed time series. Finally, the present and future challenges and perspectives are
addressed and identied, highlighting the reforms that may be crucial in the
forthcoming period.

Keywords Offshore wind power Wind resource assessment  Offshore wind

turbines Wind resource assessment methods

N. Salvao  C. Guedes Soares (&)

Centre for Marine Technology and Ocean Engineering (CENTEC),
Instituto Superior Tcnico, Universidade de Lisboa,
Av. Rovisco Pais, 1049-001 Lisbon, Portugal

Springer International Publishing Switzerland 2016 121

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_7
122 N. Salvao and C. Guedes Soares

1 Introduction

The development and increased production of the global energy needs from
alternative energy sources have been undergoing a signicant increase. The
declining reliance on fossil fuels such as coal and oil, in conjunction with concerns
about global warming and the growing energy requirements are some of the factors
that have been pushing the use of renewable energies forward. Fossil fuels are
non-renewable and release atmospheric pollutants when burnt, including green-
house gases. In order to meet the Kyoto Protocol target, countries over the globe are
working towards cutting the emissions to levels that would prevent dangerous
anthropogenic interference with the climate system. This will imply reducing the
greenhouse emissions by up to 90 % between 2040 and 2070, according to the
Intergovernmental Panel on Climate Change [1].
For these reasons, in recent years there has been an upsurge in the utilization of
renewable sources of energy that meet the targets for reducing and eliminating
emissions and increasing demand for energy. Notwithstanding, responding to the
new energy demands in terms of efciency and productivity, yet responding to the
environmental problems, is a major challenge. Solar energy, hydropower,
geothermal, wind, wave and tidal power are some of the potential sources to obtain
electricity, reducing the dependency on fossil fuels.
Among the new sources of energy, wind power is one of the most promising and
rapidly expanding industries, mainly due to its availability and price competitive-
ness. Furthermore, oceans offer good opportunities for sustainable economic
development and for this reason the offshore wind industry has experienced
remarkable development, towards optimizing the exploitation of the resources.
Offshore winds are generally more abundant, stronger and steadier than land-based
winds, increasing with the distance to the coast. Therefore, offshore wind facilities
can produce a signicantly large amount of electricity, which motivates the search
and improvement of stateof-the-art technologies. For instance, innovative foun-
dations like floating wind turbines allow harvesting wind power in harsher condi-
tions associated with deeper waters. In fact, in 2014, the average water depth of
offshore wind farms was 22.4 m, and the average distance to shore was 32.9 km [2].
These numbers increased in relation to the previous year leading to extra produc-
tivity benets that are often achieved from distance to shore. Against the stated
advantages, the costs of building offshore are much higher, therefore an accurate
assessment of the wind resources at potential wind turbine construction sites
becomes even more imperative.
On this basis, wind resource maps are commonly the primary method used to
present the amount of energy available for extraction. Usually, in the initial stages
of a wind project, the average wind speeds are calculated at the hub height of
modern wind turbines (70100 m) and presented as wind power density maps. Over
the past few years, the latest computer mapping techniques optimized for offshore
modelling have allowed improving the resource assessment. Moreover, a variety of
Resource Assessment Methods 123

methods for obtaining long-term wind speed data is now available for the devel-
opment of offshore wind resource databases.
Typically, wind masts or offshore buoys constitute the primary method for
obtaining reliable time series of long-term reference wind data. Regardless of its
advantages, these sources of data are very expensive and often sparsely located,
thus not sufcient for detailed wind sensing. The spatial and temporal character-
istics of remote sensing-based wind measurements turns them into alternative
potential tools to get more accurate estimations of the wind resource even though
this type of data can be noisy near coastal areas due to land contamination effects.
For this reason, one of the most mainstream methods to address wind assessment
offshore is the use of numerical weather prediction models. Particularly, regional
numerical models provide high temporal coverage and sampling along with high
spatial resolution on a reasonable computational cost. Still, even though these tools
have been used over the past few years showing reasonably accurate results, as
offshore wind parks move into deeper waters, the cost of production and mainte-
nance increases dramatically. For this reason, the need for reduction of the asso-
ciated measurements errors demands state-of-the-art techniques that often include
the combination of multiple wind resource assessment tools.
For a reliable assessment, long-term evaluation of the wind frequency distri-
bution for a target site, describing its variation with respect to time, becomes
necessary. For instance, probability distributions are a close approximation to the
wind variations providing a representation of how often the wind blows at a certain
speed. Low-frequency variability of above and below average wind speed, allow
determining the wind power variations and, more particularly, whether a high
average wind speed is caused by a prevailing occurrence of high wind episodes or a
lower frequency of occasional extreme events such as violent storms. To account
for these possible effects, the mean seasonal and diurnal wind variability assessment
is of extreme importance in order to increase the wind farms efciency. It is
important to identify the signicant sources of production variation on an
intra-annual basis, and calculate productivity scenarios to evaluate revenue risk.
Comparing these estimates with actual production measures from operating wind
farms will allow determining the feasibility and uncertainty associated to these
projections, and determine the impact of power fluctuations.
Despite the unquestionable advantages the offshore energy offers, a few draw-
backs cannot be disregarded. On one hand the capital costs are higher than onshore
yet partially compensated by the larger energy yields. Challenges in terms of
operation, maintenance and the need to connect them to the main electricity system,
highly dependent upon the water depth and the distance to the shore, are other key
challenges for the sector. However, with the development of new techniques and
the increased interest in the offshore wind market, these constraints are becoming
increasingly less binding over time, thus lowering the market prices.
Within this scope, this paper provides a broad review on the different methods
and strategies used in wind power assessment, together with the present and future
124 N. Salvao and C. Guedes Soares

2 Global Offshore Wind Resources and European

Offshore Wind Market

The global energy system is adapting to a changing world with renewable energies
meeting 5 % of global power output, and accounting for nearly 3 % of primary
energy consumption [3]. This section summarizes the major energy trends around
the world highlighting how well these goals and objectives are being met by
different countries, providing a global perspective on offshore wind energy.
Globally, offshore wind is a major trend in the market that has been increasing in
popularity since the worlds rst offshore wind farm was established at Vindeby,
Denmark in 1991, and is expected to continue to increase in the next decades.
According to the Global wind energy council, the projections by 2020 indicate that
offshore wind will be about 10 % of global installed capacity. Moreover, wind
power costs are expected to reduce about 25 % on land and 45 % offshore by 2050.
Currently, around 7 GW of offshore wind capacity has been installed worldwide,
with Northern Europe leading the world in offshore wind market, retaining 90 % of
offshore wind farms. In China, due to the increasing focus on renewable energies,
the government has established a target for offshore wind development: 30 GW by
2020, which would turn them into the world leaders in this sector. On the other
hand, Japan and the United States are two other possible world-leaders in the
offshore industry that have huge objective targets for installations off their coasts by
2020. Additionally, some governments in Korea, Canada and even India have
recently shown interest in developing the offshore wind sector.
Similarly, the potential of offshore wind energy in Europe is immense. Like most
of the countries worldwide, Europe has strong policies in place to cut CO2 emis-
sions, increase renewable energy, and improve energy efciency. Currently, 2488
turbines are installed and grid connected, corresponding to a cumulative total of
8045.3 MW in 74 wind farms along Europe. In the rst six months of 2014, Europe
fully grid connected 224 offshore wind turbines in 16 commercial wind farms and
one offshore demonstration site yielding a combined capacity of 781 MW.
Among the European countries, the UK has the largest amount of installed
offshore wind capacity in Europe (4494.4 MW) corresponding to 55.9 % of all
installations, followed by Denmark that holds a market share of 15.8 % with
1271 MW installed capacity. Germany comes in third place with 1048.9 MW (13 %
of total European installations), followed by Belgium (712 MW: 8.8 %), the
Netherlands (247 MW: 3.1 %), Sweden (212 MW: 2.6 %), Finland (26 MW:
0.3 %), Ireland (25 MW), Spain (5 MW), Norway (2 MW) and Portugal (2 MW).
According to the European Energy association (EEA) the expected growth in the
offshore wind energy industry includes a total of 40 GW installed capacity by 2020
and 150 GW by 2030. It is expected that 12 more offshore projects will be com-
pleted by the end of 2015/2016, increasing the installed capacity by a further
2.9 GW thus bringing the cumulative capacity in Europe to 10.9 GW.
Due to the rapid expansion, the sector is beginning to benet from strong
funding opportunities. The investment in offshore wind projects ranged from 4.2
Resource Assessment Methods 125

Fig. 1 Annual offshore installations (19932014). European energy association report, 2015

billion to 5.9 billion from January to December of 2014, connecting 408 new
offshore wind turbines. Also, one of the largest ever project nance transaction in
the offshore wind sector, the EUR 2.8 billion Gemini project, reached nancial
close in May 2014 by a consortium of four sponsors and 16 creditors. The 600 MW
Gemini offshore wind farm is to be located 55 km north of the island of
Schiermonnikoog in the Dutch North Sea, an area with excellent wind conditions,
approximately 85 km off the coast. Designed to meet the power needs of 1.5 million
people, while cutting carbon dioxide by 1,250,000 ton, the farm is expected to be
one of the worlds biggest offshore wind farm and will produce approximately
2.6 TWh renewable electricity. The facilities will have a total surface area of
68 km2 in water depths expected to range between 28 and 36 m and the turbines
rotors, at 150 m above sea level, will the tallest in Europe. Figure 1 shows the
cumulative offshore wind power and annual addition installed in Europe, evi-
dencing the fast growth of the offshore sector.

3 Methodologies for Wind Resource Estimation

The global efforts to reduce greenhouse emissions resulted in a large investment in

the offshore wind sector. Given that the modern wind turbine now lasts up to
25 years before needing an upgrade, wind energy is now considered a long-term
investment. The inherent nature of the wind turns wind power into a fluctuating
source of electrical energy. Accurate energy production estimates are difcult to
acquire and rely on accurate long-term wind data. Therefore, the accurate assess-
ment of the long-term wind speed variations and the sea states at a site, together
126 N. Salvao and C. Guedes Soares

with the associated uncertainty in such predictions is of most importance for the
design of a wind turbine model and the selection of a wind farm construction site.
Classical approaches include the use of meteorological wind masts, numerical
weather prediction models and remote sensing data. In present days, some of these
long-term gridded wind data sources are freely available and easily accessible from
public Internet databases. This stimulates data analysis and further research and
development in wind power applications. Although this is one of the many
methodologies adopted, Fig. 2 depicts a graphic representation of the steps involved
in the creation of an offshore wind chart. Typically, the main steps involve
obtaining accurate wind information from a variety of sources. Using statistical
methods, it is possible to estimate long-term wind speed time-series and combine it
with the wind turbine characteristic power curve. This allows for an estimation of
the gross and net energy, which can be obtained using different wind turbine
models. Finally, the uncertainty associated to the calculations and economic risk
assessments are performed. Different procedures are followed by the authors
depending on, among other factors, the amount of data sources used to collect wind

Obtain and analyze wind data (meteorological masts, off-

shore buoys, remote sensing instruments, numerical weather
prediction models)

Estimation of long-term frequency distributions using long

term time-series of wind measurements or statistical methods
(ex. Measure correlate predict)

Estimation of the amount of wind energy

available for extraction

Estimation of the gross energy Estimation of the net energy pro-

production by combining the wind duction accounting for energy losses
turbine model characteristic power like electrical losses wind turbine
curve with the wind speeds frequen- performance, environmental losses,
cy distribution wake effects, among others

Economical risk assessment

Fig. 2 Flowchart of the main steps involved on a typical wind power assessment
Resource Assessment Methods 127

information, the height were the measures are taken, and the criteria for uncertainty
and risk evaluation.
Over the past few years, the combinations of two or more methods have raised
prominence in energy assessment strategies. For instance, deeper water foundations
often represent larger investments mainly due to higher construction costs and
difcult access as a consequence from the larger distance from the coast. Grid
connection alone can account for up to 30 % of the project cost. Also, higher wave
heights and stronger winds can cause damage to the equipment. For these reasons,
the use of combined technologies increases the reliability of the results, mitigating
investment costs. Soukissian and Papadopoulos [4] studied the effect of different
data sources in offshore wind power assessment. Using wind data obtained from
numerical weather prediction models and Seawinds blended satellite data, the
authors rst veried that the use of this type of data is subject to large uncertainties
when compared with in situ buoy measurements. After applying a calibration
procedure, major improvements were veried. This suggests that the use of reliable
calibration relations can help reducing the uncertainty associated to data sources
other than in situ measurements. However, this study shows the importance of a
correct use of the type of data source selected.
Hereafter, a review of the different methods applied to different approaches is
presented, highlighting the advantages and disadvantages that arise from each
method, and the major improvements accomplished over the past few years through
optimization procedures.

3.1 Wind Masts and Offshore Buoys

Multiple technologies and methods are available to assess the offshore wind
resource at potential sites. As already stated, direct methods are often preferable, but
the increased measurement accuracy is largely compensated by the substantial costs
to collect such data. Moreover, the scarcity of the data and the fact that they are
often sparsely located, suggests that the utilization of alternative datasets could
represent an added value over the use of direct measurements. For this reason, in
present days, atmospheric modelling coupled with remote sensing techniques is
state-of-the-art in offshore wind industry. Nevertheless, high-quality wind data,
measured at different locations and altitudes, are critical for an accurate assessment
of the resource potential at a proposed site. Also, they constitute a powerful tool to
validate and calibrate numerical models or other wind estimation techniques.
Meteorological masts have the advantage of collecting data at different heights
while offshore buoys often collect data within the 38 m ranges. In these cases, in
order to calculate the wind resources at the turbine hub height, the wind has to be
subject to extrapolation methods increasing the degree of uncertainty, thus making
it less representative of the marine environment.
Due to its numerous advantages, wind meteorological masts have been installed
worldwide to collect wind data for offshore wind energy estimation purposes. For
128 N. Salvao and C. Guedes Soares

instance, the Korea Electric Power Corporation installed the HEMOSU-1 (Herald of
Meteorological and Oceanographic Special Unit-1) meteorological mast about
100 m above mean sea level. The equipment consists of an offshore tower that
measures wind speeds in the Younggwang region. Furthermore, wind records from
meteorological towers are the most effective method for detecting the occurrence of
long periods of low wind speed. Previous studies have demonstrated that even
individual events with approximately 10 year return period can have strong impli-
cations for wind power, leading to drops in the annual energy yield up to 5 % [6].
The authors emphasize the importance of detecting such events using long-term
wind records obtained from meteorological towers, and propose two techniques for
the examination and quality control of such datasets. The methodology focuses on
the prolonged low wind speed events, over timescales that range from hours to days.
Innovative technologies are currently in development aiming to reduce the cost
of wind measurement campaigns, particularly offshore and at larger distances from
the coast, where the cost of obtaining wind information is higher. For instance, the
FLOATMAST project [7] demonstrates that a tension leg platform for combined
cup anemometer and Lidar wind measurements is suitable as means to provide
accurate measurements for offshore wind applications. Also, it can be redeployed at
multiple offshore locations, and distances from the coast, which signicantly
reduces the costs of wind measurement campaigns. Figure 3 displays an example of
a typical offshore wind measuring tower.

Fig. 3 Example of a wind

measuring tower, the Hornsea
meteorological mast. (Source
Smart wind: http://www. [5]
Resource Assessment Methods 129

3.2 Remotely Sensed Data

Remotely sensed data is globally available and has been widely used to describe the
spatio-temporal distribution of wind resources at potential locations. In addition,
such data has been used in conjunction with numerical modelling tools through data
assimilation systems, aiming to improve the latter. A variety of instruments are
available to study the offshore wind climatology which include LiDAR, sodar,
altimeters, scaterometers and synthetic aperture radars. These instruments have
different spatial resolutions and temporal coverage, therefore serving different
In coastal regions, because of the complex topography resulting from the sea
land surface transition, high spatial resolution images are required. Therefore, wind
vectors retrieved from satellite-borne synthetic aperture radar (SAR) can provide a
better spatial coverage of the coastal areas of interest with resolutions about few
hundred meters, providing the most detailed wind maps [8]. SAR emits microwaves
into the sea surface and uses the measurement of backscattering signals reflected by
capillary waves, enabling to estimate the ocean surface wind speeds. Its major
drawback is the narrower swath that yields a lower temporal coverage. Special
attention has to be paid to near shore areas to avoid contamination by the coast in
the backscatter. SAR-derived images wind elds can have a higher spatial reso-
lution than the ones from wind scaterometers, thus a higher precision, and a better
spatial coverage than in situ data. C-band SAR can be used to extract wind speed
and direction over the ocean at high spatial resolution, and for this reason has been
used in a large number of wind resource assessments.
Based on the work of Hasager et al. [9] the influence of increasing the number of
observation samples through the combined use of Envisat ASAR, ASCAT and
QuikSCAT is investigated. The analysis is based on SAR data from the Envisat
10-year archive from 2002 to 2012, processed by CMOD-IFR2 geophysical model
function. Although there is no evidence to support whether the uncertainty in wind
assessment is reduced, the results demonstrated that more spatial detail is added
from SAR. The use of SAR is relatively recent and has been going several
improvements. Wind retrieval algorithms have been tested in order to take the best
advantage of this type of data and increase the performance of the results.
In addition, weather radars immerge as novel tools for tackling the lack of
predictability of offshore power. Recent studies have shown the effectiveness of
these tools as means for anticipating changes in the wind pattern, allowing for more
advanced control strategies, planning and grid operation for offshore wind farms.
They are also useful tools for observation of turbine wake effects at large wind
farms, responsible for signicant power losses. These systems are the perfect tool
for detecting meteorological phenomena such as weather fronts, often responsible
for wind fluctuations, for the reason why they are receiving considerable attention if
the wind sector in present days [10].
Likewise, scaterometer instruments can provide estimations of surface wind
vectors with high spatial and temporal resolution over oceans. These types of
130 N. Salvao and C. Guedes Soares

sensors include for example SEASAT-SASS, QuikSCAT-Sea Winds,

ADEOS-NSCAT and XERS-AMI-Wind of the European Space Agency
(ESA) among others. For instance, NASAs SeaWinds scaterometer on board
QuikSCAT has provided surface wind estimates since 1999, remaining fully
operational until it stopped working in 2009 due to technical failure. The wind
speeds derived from the scaterometers have an accuracy of approximately 2 m/s
and a spatial resolution in the range of 2550 km.
Although having a poorer resolution than SAR data, scaterometers have proved
to be reliable tools to provide site specic data, and determine long-term choices for
energy investors and for this reason have been used in various wind resource
assessments [11, 12]. Figure 4 illustrates a typical wind power map obtained using
QuikSCAT data. The wind power density is represented for the entire globe in the
Northern Hemisphere winter and summer, providing an initial estimate of the
energy available for extraction.
Despite the many advantages, satellite data can be noisy near coastal areas due to
land contamination. Moreover, satellite sampling can be insufcient to accurately
represent the wind eld variability. The combination of satellite information with
other wind resource assessment tools is therfore of most importance. In fact,
combining various sources of data leads to less statistical uncertainty and a better
coverage of the diurnal wind speed variation. For this reason, wind resource esti-
mation based on wind samples obtained from different remote sensing instruments,
has been a subject of scientic interest in recent years.

Fig. 4 Wind power density in winter and summer (top and lower panels respectively) obtained
using QuikSCAT data. Image credit [13]
Resource Assessment Methods 131

3.2.1 LiDAR Wind Measurement

LiDAR (Light Detection and Ranging) is a remote sensing technology capable of

measuring three-dimensional relative wind velocity, using the light in the form of a
pulsed laser and detecting the scattered light resulting from the interaction. The use
of LiDAR has become popular for its technical capabilities of capturing accurate
and reliable wind data, both inshore and offshore. It has been recently tested for the
use in wind resource assessment campaigns, the study of propagation of wind
turbine wakes and their implications for wind turbulence, and the study of
preview-based wind control systems to reduce turbine structural loads. Moreover,
floating-LiDAR systems have been going through considerable progress in the past
few years. Composed by a LiDAR device placed on top of a floating platform, the
system has been tested against meteorological masts with wind sensors in both
onshore and offshore environments. The floating platforms may include different
sea-based structures, like barges ships and buoys. Furthermore, LiDAR can be
mounted on a wind turbine hub and measure the incoming wind flow for wind
turbine control optimization.
Comparisons of floating LiDAR devices with land-based LiDAR in both
ten-minute averaged wind speed and wind direction have showed promising results
[14]. This is an indicator that this system is a cost effective option to perform high
accuracy wind measurements for offshore wind power development. For instance,
the Fraunhofer IWES Wind LiDAR Buoy is a floating-LiDAR system that inte-
grates a Windcube-v2 LiDAR device on an adapted marine buoy with a height of
7.2 m, and has been undergoing some tests and validations. As the motions from a
floating system can cause systematic measurement errors of the wind speed, a
motion-correction algorithm was implemented in order to remove the effects of sea
conditions. The results were validated against the FINO1 meteorological mast
FINO 1 showing a very good correlation. Since the measurement uncertainties are
similar to those for offshore mast measurements, it can be concluded that the use of
the LiDAR can help to lower the cost of offshore wind, turning into a competitive
alternative to traditional offshore met masts [15].
These tools have been receiving much attention and have been undergoing a
series of tests across the globe. Recent surveys have suggested that dual-Doppler
LIDAR can be used for obtaining microscale winds and examine ne-scale vari-
ations in the wind flow, at the wind turbine scale [16]. They can be coupled with
traditional in situ devices to obtain space and time resolved measurements for wind
energy research.

3.3 Numerical Weather Prediction Models

Numerical models are often used to calculate the wind resources at a specic site,
particularly offshore, due to the challenges that arise from the lack of in situ
measurements. Numerical models have the key advantage of simulating
132 N. Salvao and C. Guedes Soares

high-quality wind data for any region of the globe, particularly where surface
measurements are scarce or non-existent, at a relatively low computational cost
[17]. Moreover, mesoscale models can use results from global models or reanalysis
datasets as input, many of which were made public in the past few years. For this
reason, in recent years, wind power assessment studies have been conducted using
data generated by mesoscale and microscale numerical weather prediction models,
becoming the stateof-the-art method for obtaining energy estimations.
In complex terrain sites where micro-scale flow models are needed, computa-
tional fluid dynamics (CFD) models such as WAsP, WindSim and 3DWind account
for a better representation of the wind climate, as well as the better estimation of
possible high altitude extreme wind events [18]. In fact, when the complex
topography or localized roughness changes require a model that can resolve
micro-scale processes, mesoscalemicroscale coupling is an alternative effective
modelling method. In this case, mesoscale models provide the initial and boundary
conditions to the microscale flow model.
The standard model for wind resource assessment has been WAsP (Wind Atlas
Analysis), a linear computational model developed by the Wind Energy and
Atmospheric Physics Department at Ris National Laboratory (now the Department
of Wind Energy at the Technical University of Denmark). In addition,
WindSim CFD analysis computational model is also frequently used. The WAsP
combines the advanced Computational Fluid Dynamics (CFD) with the effective
Wind Atlas Method. It is often used when a spatial resolution to account for the
atmospheric micro-scales or higher is required, in order to obtain a more detailed
analysis of the wind pattern. The input data consists of measured or simulated wind
information though running WAsP with input data obtained from a numerical
mesoscale model has become a common practice. In fact, numerical flow models
are often coupled with NWP models to predict variations in the wind flow, par-
ticularly at complex terrain where a NWP model is not enough to resolve the
complex flow phenomena.
The WRF has been widely used for wind resource assessment studies. Salvao
et al. [19] demonstrated that the model constitutes a powerful tool for identifying
the wind resources in potential areas for the development of offshore wind parks.
The authors also concluded that the use of a numerical model improves the quality
of the predicted wind elds, when compared with coarser resolution datasets such
as reanalysis wind data. Carvalho et al. [20] also used the mesoscale model WRF
and the WAsP microscale analysis tool to carry out wind resource assessment of
two complex sites in Portugal. Comparisons with measured wind data show that the
insertion of mesoscale output into the microscale model leads to deviations mostly
linked to an incorrect representation of the terrain features, a pattern that can be
corrected leading to signicant improvements. The authors also discuss the
importance of a correct representation of local terrain complexity, on the grounds
that its misrepresentation can cause distorted mesoscale induced terrain effects,
leading to strong underestimation of the true wind speed. Although this study was
performed onshore, the methodology has been used to study complex coastal areas
at offshore locations.
Resource Assessment Methods 133

Comparisons between numerical wind flow models show that coupled mesoscale
models perform better than standalone linear JacksonHunt type and nonlinear
steady state CFD/RANS models, that do not have the physical equations to fully
describe the atmospheric phenomena. The only disadvantage is the computational
power required to perform such calculations. In numerical models, computational
costs typically increase with increasing spatial and temporal resolutions. Although
several improvements that aim turning numerical models into computational ef-
cient tools have been undertaken, these tools can be time-consuming on a wind
farm project. Furthermore, offshore or in relatively smooth terrain, the variability in
wind speeds is generally smaller, thus the horizontal resolution of mesoscale
models is high enough to resolve the atmospheric processes, reducing computa-
tional costs. For these reasons, different model solutions with the appropriate res-
olutions are crucial tools to obtain a rst micro-siting and energy estimates of a
wind farm, minimizing the risks for the project. For instance, mesoscale models are
particularly useful to explore complex mesoscale flow phenomena such as coastal
low-level jets or landsea breezes [21]. In particular, low-level jets can have a
damaging impact on wind turbines due to the increased load and fatigue. A detailed
characterization of these effects becomes necessary particularly at sea, where the
lack of wind measurements is large.
Aiming to compare different sources of data, a comparative study has been
carried out using winds retrieved from WRF, satellites (Cross-calibrated
Multi-platform ocean wind vectors, QuikSCAT scaterometer, NCDC Blended
Sea Winds and IFREMER Blended Wind Fields), reanalyzes (NCEP-CFSR,
ERA-Interim, NASA-MERRA and NCEP-RII) and analyses (NCEP-FNL and
NCEP-GFS). The results reveal that high-resolution winds derived from numerical
weather prediction model WRF account for the best results, indicating that it is the
best alternative to perform offshore wind resource characterization when on-site
measurements are not available. Compared to the remaining data sources, the
modelled data presents the lowest errors in terms of offshore wind power flux
estimations [22].
Combined approaches or merging all available data is a possible solution to get a
more accurate estimation of wind resources. Chang et al. [23] combined multiple
satellite winds and numerical results from the WRF model in order to reconstruct
offshore winds and veried the applicability of the satellite-data-based method for
wind resource mapping in part of the South China Sea. Wind speeds obtained from
SAR and ASCAT retrieval data were validated using wind measurements from
meteorological masts, and posteriorly introduced in the WRF model using a data
assimilation technique. The results showed that sources of data that do not originate
from on-site measures should be dealt with carefully. The comparisons of the wind
speed with respect to buoy measurements showed an overestimation by satellite
data, while the model underestimates the true wind speed. Similarly, methodologies
for combining satellite winds and model simulations have been developed. Studies
have shown that satellite data can be assimilated into numerical models in order to
obtain accurate reconstructed offshore wind information. So far, the results have
indicated that this method is valuable for some offshore regions characterized by the
134 N. Salvao and C. Guedes Soares

scarcity of local in situ measurements. Fernandes et al. [24] investigated the effects
of ingesting surface wind speed data from QuikSCAT and sea surface temperature
(SST) from GHRSST Level 4 analysis into the WRF model, using a Newtonian
relaxation assimilation technique. The model was congured to produce 20 km
winds on a domain centred in the North Sea, and the output was compared against
observational data from the FINO 1 meteorological mast. The results showed
signicant improvements up to 5 % over the entire domain, demonstrating the
potential benets of assimilating real-time observations into the numerical models
for wind energy purposes.
Numerical models also serve the short-term wind power forecasting needs. Wind
power is stochastic, especially in the short-term time horizon, fluctuating due to the
variable and intermittent nature of the wind. With respect to that, it is necessary to
plan the maintenance of the wind farms in order to, among other tasks, schedule
operations, power system planning for unit commitment and dispatch and energy
storage operations, maximizing power production, improving integration and
reducing imbalance charge. Many of these operations may be costly so optimal
planning from accurate estimates of the expected production are essential, and have
been undergoing signicant attention, particularly in a very short term [25].

4 Wind Resource Assessment

4.1 Wind Power Density and Energy Production: Statistical


The development of strategies for reducing uncertainty inherent to wind speed

long-term variability is a challenge. Due to the stochastic nature of the wind,
changes in its magnitude and direction occur frequently. Therefore, mathematical
distribution methods and statistical tools help describing the wind prole. When
exploring wind energy resources, the average wind speed is not enough to char-
acterize the climate patterns of a specic site. Complete knowledge of the wind and
wind power variability, efciency, stability and security are of uppermost impor-
tance for a correct utilization of high-resolution spatial information, and guarantee
the reliability of offshore wind farm generation. For instance, intermittent nature of
wind is an important problem for wind power producers. For the wind power
industry, an effective quantication of risk or probability that an offshore wind
turbine suffers from severe extreme wind speeds, is indispensable.
One of the most commonly used methods as initial guidance for selecting the
most energetic areas is the wind power density (WPD). As it is not dependent upon
turbines specications, the WPD is a useful way to evaluate the available wind
resource in order for planning suitable places for wind farm sitting. The wind power
density is proportional to the cube of the velocity and for that fact, accurate mea-
surements of the wind speed are crucial for a correct resource assessment. The mean
Resource Assessment Methods 135

wind speed itself already provides an initial rst guess of the wind potential of a
specic region, however, the nonlinear relationship between wind speed and the
available power may cause misleading estimations of the actual amount of available
energy. To account for this issue, the Wind Energy Pattern Factor (WEPF) or cube
factor can be introduced in the WPD formula to improve the accuracy of the wind
resource estimates, by introducing a correction factor [26]. Otherwise the data can
be separated into wind speed intervals or bins in which they occur, so that the wind
power density can be calculated taking into account the wind distribution within a
certain range. Combining this information with the wind turbine characteristics,
namely the rotor area or the wind turbine power curve, the annual wind power
production can be estimated for different wind turbine types.
Zhang et al. [27] developed a new method for wind resource assessment using wind
power potential (WPP) estimations, taking into consideration the joint distribution of
wind speed and direction. The authors further investigated the effect of wind distri-
bution on the optimal net power generation of a wind farm. The method has proved to
be efcient on the early stages of wind farm design, although special care should be
taken in the latter stages of a wind farm planning, in which the use of computationally
more complex and computational expensive models is recommended.
Similarly, Jesus et al. [28] proposed an analytical method that combines statis-
tical techniques using MATLAB code and the creation of a wind chart, to be used
in the rst decision steps towards choosing a wind farm construction site. The chart
includes important information regarding the long-term distribution adjusted by a
Weibull probability distribution function, the analysis of extreme wind events the
vertical prole and climate variations such as the intra-annual and seasonal
In terms of extreme winds estimations, little work has been carried out so far.
A signicant factor that has to be taken into consideration is the probability that
offshore turbines suffer from extreme wind events. Phenomena like typhoons
although not common in every areas, should not be disregarded since they can give
rise to a series of events, from sudden changes in the wind direction, strong tur-
bulence, thunderstorms and storm surges, representing a great threat to offshore
wind turbines. The peaks over threshold analysis (POT) and maxima over a specic
time period analysis (MOSTP) are the two most common approaches to investigate
these events. Wang et al. [29] provided a review of the current status of extreme
wind speeds and wind energy assessment, and proposed a GH statistical calibration
technique based on articial intelligence optimization algorithms to re-analyze
future samples of extreme wind speeds.
Monte Carlo simulation and Measure-Correlate-Predict (MCP) are two of the
methods that can be used to provide a detailed description of extreme wind prob-
ability distribution. The two methods have been compared for extreme wind con-
ditions caused by both tropical and extratropical cyclones. In the case of
extratropical cyclones, the MC method performs well, however, extreme winds
induced by extratropical cyclones have large uncertainty associated. On the other
hand, the MCS method accurately estimates the probability distribution of extreme
winds for mixed climate regions [30].
136 N. Salvao and C. Guedes Soares

Another linked issue is the lack of long-term records of wind speed measure-
ments. Standard approaches to address this problem, like the MCP, consists in
correlating on-site wind data with data recorded at a long-term reference station, in
order to estimate the wind resources that represent the long-term conditions at a
target site. Although it is one of the most used techniques, other methods have
shown to be effective in reconstructing missing on-site wind data. The analog
ensemble technique is an alternative method that uses long-term samples of wind
data, such as reanalyzes datasets or observations that match the mast measurements,
and identies analogs in order to reconstruct the missing data. It only requires one
years worth of site data measurements, and when compared with observation, this
technique showed its high quality and computational efciency, having consistently
higher correlations as well as smaller biases and centred RMSEs versus the com-
monly used MCP techniques [31].
When the wind speed probability distribution is known, the wind energy dis-
tribution can be obtained. This information is of key importance in assessing wind
energy potential, particularly when year-long time series of meteorological obser-
vations are unavailable. Frequently, Weibull distribution tting is adopted to
describe the wind speed distribution. The Lognormal and Rayleigh functions follow
as the most suitable distributions when the Weibull cannot properly represent the
wind distribution on a specic geographical area. Still, a variety of probability and
cumulative distribution functions are available to describe the behaviour of the
wind eld at a particular location. The suitability of each probability function in
describing a wind speed time series can be accessed by evaluating the goodness of
t for each method, through a series of tests including, for instance, the Pearsons
chi-squared test or KolmogorovSmirnov t test.
Being the most suitable probability function for wind energy assessments,
comparative studies have been focusing on the efciency of different methods in
determining Weibull parameters for wind energy potential. The maximum likeli-
hood method is the most commonly used to estimate the Weibull parameters,
although other traditional methods including the Moment method or the L-moment
method can be employed. The work of Chang et al. [32] provides a review of six
numerical methods for estimating Weibull parameters: method of moments,
empirical, graphical, maximum likelihood, modied maximum likelihood and wind
energy pattern factor method. Similarly, Don et al. [33] proposed three intelligent
optimization algorithms to estimate Weibulls parameters, including particle swarm
optimization (PSO), differential evolution (DE) and genetic algorithm (GA).

4.2 Wind Power Error EstimationUncertainty Analysis

So far, it has been shown that there is a variety of methods to study the amount of
available wind energy to be extracted by offshore wind devices exists, and that wind
energy resource assessment applications require a complete knowledge of the wind
regime and variability. Due to the natural intermittency of the wind, the energy
Resource Assessment Methods 137

quality and stability are additional factors that cannot be disregarded, as well as
establishing wind speed trends with a high degree of certainty. Reliable wind power
estimates are therefore crucial for the industry, giving it more condence to invest
in offshore wind projects. The uncertainty in the estimation of the long-term wind
speed is a major contributor to the total uncertainty of a wind farm project. For this
reason, there must be a quantication of the uncertainty associated to the average
annual wind power predictions, in order to minimize the risk of building turbines in
areas where the wind resource is actually highly uncertain.
Throughout this work, it has been demonstrated that several methods are being
developed and improved, in order to increase the reliability of wind resource
assessments. An accurate wind power error estimation, nancial risk assessments
and production losses are some of the concerns that need to be addressed and
quantied, in order to maximize the yield of future offshore wind developments. For
instance, mesoscale modelling has shown to be an effective framework to accu-
rately estimate losses due to wake effects, which represent up to 20 % of a wind
park annual energy production in large wind farms (Jimnez et al. 34).
Notwithstanding, the reliability of a wind resource assessment does not solely
depend on the quality of the wind measuring equipment or modelling technique.
Depending upon the methodology, the number of sources of uncertainty can be
extremely large. In fact, a typical Measure-Correlate-Predict (MCP) resource
assessment can reach out to 14 sources of uncertainty [35]. On the other hand, the
range of uncertainties induced by wind modelling, depend on a variety of other
factor such as the initial and boundary conditions, domain resolution, dynamical
options and nudging techniques. In fact, on a typical wind resource assessment
from either mesoscale model simulations or statistical approaches, inherent and
unavoidable uncertainties have two main sources: those related to wind speed
assessment and those related to the energy assessment. Both are intrinsically related
since the error in wind speed estimation propagates to the power production
The spatio-temporal analysis of propagation of errors is equally important for
wind power management and optimization procedures. Statistical methods can be
used to determine, for example, the goodness of t of the probability distribution, so
that the effect of wind uncertainty propagation on energy production estimates can
posteriorly be calculated.

5 Present and Future Challenges

The offshore industry has experienced a vast growth in the past few years and is
expected to continue to increase in the forthcoming period. Despite the many
advances in wind resource assessment techniques, upcoming challenges motivate
the search for new or improved methods of wind resource estimation. In present
days, the main differences between the various assessment methods lie funda-
mentally on the type of equipment or numerical model used to obtain accurate
138 N. Salvao and C. Guedes Soares

information on long-term wind speed data. The main challenges involve the choice
of an accurate source of wind speed data, and the statistical method to deal with
variability and long-term extrapolation.
Among the various potential sources of uncertainty, particularly in long-term
projections, understanding climate change and its impact on wind resources is one of
the main challenges, and a topic of vital importance. Due to the risks and threats to
wind power potential it presents, this matter cannot be disregarded. The spatial dis-
tribution and the intensity of the wind may be affected if there are substantial changes
in the near-surface atmospheric flow. Recent studies have investigated the impact of
climate change scenarios in the availability of the wind resources. Results have shown
reductions in the wind energy density up to 20 % in some regions, based on projections
of mean wind speed changes for 20312050 [36]. Additionally, previous studies over
Europe suggest a potential increase in the energy density during the wintertime in
northern areas, and in parts of the Mediterranean in summer, while decreases may
occur in the south-eastern regions by the end of the twenty-rst century [37].
Another issue related to the use of the above mentioned approaches is that, in
general, the standards used in offshore wind industry are the same as for onshore.
This implies simplications regarding the marine boundary layer which could have
large implications in the wind energy calculations. Due to the different character-
istics of the boundary layer over land and sea, an increased knowledge of the
complex atmospheric flow phenomena is crucial for a better design of offshore wind
harnessing devices. Also, the thermal effects that influence the wind prole have to
be known to avoid misleading calculations [38]. Finally, a recent research topic
involves the combined wind and wave energy harnessing. The ocean is a large
source of natural resources wind and wave therefore optimizing their exploitation is
of commercial interest, due to the many advantages it presents. Alternatives like
energy islands or hybrid wavewind energy converters are solutions to develop a
more sustainable energy. Prez-Collazo et al. [39] provides a review of the existing
possibilities for combined wave and wind offshore energy, highlighting the current
limitations of these systems.

6 Concluding Remarks

Throughout the world, wind energy has the largest share in renewable energy. In
particular, the development of offshore wind power has been increasing rapidly
over the past few years, due to the large advantages it presents. Globally, most
countries are motivated in meeting medium/long-term targets for renewable energy,
which plays a crucial role in mitigating climate change, and tackle the growing
energy demands. To accomplish that, wind energy resource assessment applications
require a complete knowledge of the wind regime and variability. Due to the natural
intermittency of the wind, the energy quality and stability are other factors that
cannot be disregarded, as well as establishing wind speed trends with a high degree
of certainty.
Resource Assessment Methods 139

Currently, a large variety of methods are available for a correct wind resource
assessment. From meteorological masts, numerical models and remote sensing
tools, a large variety of wind data sources can be easily accessible and be used to
produce wind energy estimates. The number of methodologies adopted worldwide
in the scientic community is enormous therefore a brief overview of the most
commonly used methods is presented in this paper, highlighting the main advan-
tages and disadvantages as well as possible future developments. It can be con-
cluded that wind maps provide a good starting point for obtaining an initial
assessment of a regions wind resources, during a projects pre-feasibility phase,
through the evaluation of the wind power potential. Additionally, the wind turbines
specic characteristics can be combined with the wind probability function, most
often the Weibull distribution, in order to determine the expected annual energy
production of one or more wind turbines.
Apart from understanding how the wind resource is distributed along a proposed
site, one of the suggestions arising from the early studies is the need for improved
estimates of loss factors, and the development of more sophisticated approaches for
uncertainty analysis. Uncertainties play an important role in the market decision,
accounting for loss effects such as wakes or wind-turbine availability, together with
possible changes in the climate including climate change scenarios or the natural
annual/seasonal variability.
The development and improvement of wind power forecasting and the explo-
ration of innovative ways for assessing offshore wind resource will help optimizing
wind turbine sitting and maintenance. Developing a robust wind resource assess-
ment campaign is the most effective method for reducing the uncertainty on the
energy estimates; however, the use of rened flow modelling techniques and the
optimization of remote sensing post processing tools will allow a better represen-
tation of long-term site wind resources. The development of better tools and
techniques and investigating new statistical techniques will allow investors to make
clear and informed decisions on their projects that, together with a complete
nancial analysis, allows minimizing the risks for renewable energy projects.

Acknowledgements This work was performed within the Strategic Research Plan of the Centre
for Marine Technology and Ocean Engineering, which is nanced by Portuguese Foundation for
Science and Technology (Fundao para a Cincia e Tecnologia-FCT).


1. IPCC (2014) Climate Change 2014: Mitigation of climate change. In: Edenhofer O,
Pichs-Madruga R, Sokona Y, Farahani E, Kadner S, Seyboth K, Adler A, Baum I, Brunner S,
Eickemeier P, Kriemann B, Savolainen J, Schlmer S, von Stechow C, Zwickel T, Minx JC
(eds) Contribution of working group III to the fth assessment report of the intergovernmental
panel on climate change. Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA
2. The European Wind Energy Association, February 2015 Report
140 N. Salvao and C. Guedes Soares

3. BP Statistical Review of World Energy June 2014 (

4. Soukissian T, Papadopoulos A (2015) Effects of different wind data sources in offshore wind
power assessment. Renewable Energy 77:101114. ISSN 0960-1481
5. SMart Wind Ltd.
6. Leahy PG, McKeogh EJ (2013) Persistence of low wind speed conditions and implications for
wind power variability. Wind Energ 16:575586
7. Peppas A, Papatheodoru T, Tsakalomatis D (2015) An innovative TLP platform for combined
reliable and bankable offshore cup Anemometer-Lidar wind measurements. Presented at
EWEA Offshore 2013, Copenhagen
8. Dagestad KF, Horstmann J, Mouche A, Perrie W, Shen H, Zhang B, Li X, Monaldo F,
Pichel W, Lehner S, Badge M, Hasager CB, Furevik B, Foster RC, Falchetti S, Caruso MJ,
Vachon P (2012) Wind retrieval from synthetic aperture radaran overview. In: Proceedings
of the SEASAR 2012 advances in SAR oceanography, ESA SP-709, Tromso, Norway
9. Hasager C, Mouche A, Badger M, Bingl F, Karagali I, Driesenaar T, Stoffelen A, Pea A,
Longp N (2015) Offshore wind climatology based on synergetic use of Envisat ASAR
ASCAT and QuikSCAT. Remote Sens Environ 156:247263
10. Trombe P-J, Pinson P, Vincent C, Bvith T, Cutululis NA, Draxl C, Giebel G, Hahmann AN,
Jensen NE, Jensen BP, Le NF, Madsen H, Pedersen LB, Sommer A (2014) Weather radars
the new eyes for offshore wind farms? Wind Energ 17:17671787
11. Bentamy A, Croize-Fillon D (2014) Spatial and temporal characteristics of wind and wind
power off the coasts of brittany. Renewable Energy 66:670679
12. Salvao N, Guedes Soares C, Bentamy A (2015) Offshore wind energy assessment for the
Iberian coasts using remotely sensed data. In: Guedes Soares C (ed) Renewable energies
offshore. Taylor & Francis Group, UK, pp 237244
13. National Aeronautics and Space Administration.
14. Hsuan C, Tasi Y, Ke J, Prahmana R, Chen K, Lin T (2014) Validation and measurements of
floating LiDAR for nearshore wind resource assessment application. Energy Procedia
15. Gottschall J, Wolken-Mhlmann G, Viergutz T, Lange B (2014) Results and conclusions of a
floating-lidar offshore test. Energy Procedia 53:156161
16. Newsom RK, Berg LK, Shaw WJ, Fischer ML (2015) Turbine-scale wind eld measurements
using dual-Doppler lidar. Wind Energ 18:219235
17. Salvao N, Bernardino M, Guedes Soares C (2014) Assessing mesoscale wind simulations in
different environments. Comput Geosci 71:2836
18. Chang PC, Yang RY, Lai CM (2015) Potential of offshore wind energy and extreme wind
speed forecasting on the west coast of Taiwan. Energies 8(3):16851700
19. Salvao N, Bernardino M, Guedes Soares C (2013) Validation of an atmospheric model for
assessing the offshore wind resources along the Portuguese coast. In: Proceedings of the 32nd
international conference on ocean, offshore and arctic engineering (OMAE2013), June 914,
Nantes, France, Paper OMAE2013-11631
20. Carvalho D, Rocha A, Silva Santos C, Pereira R (2013) Wind resource modeling in complex
terrain using different mesoscalemicroscale coupling techniques. Appl Energy 108:493504
21. Nunalee CG, Basu S (2014) Mesoscale modeling of coastal low-level jets: implications for
offshore wind resource estimation. Wind Energ 17:11991216
22. Carvalho D, Rocha A, Gmez-Gesteira M, Alvarez I, Silva Santos C (2013) Comparison of
reanalyzed, analyzed, satellite-retrieved and NWP modelled winds with buoy data along the
Iberian Peninsula coast. Remote Sens Environ 152:480492
23. Chang R, Ron Z, Badge M, Hasager CB, Xing X, Jiang Y (2015) Offshore wind resources
assessment from multiple satellite data and WRF modeling over South China Sea. Remote
Sens 7(1):467487
24. Fernandes M, Costa P, Estanqueiro A (2011) Improving offshore wind resource assessments
using a data assimilation technique. In: European offshore wind conference & exhibition,
Amsterdam, 8p
Resource Assessment Methods 141

25. Rasheed A, Sld J, Kvamsdal T (2014) A multiscale wind and power forecast system for wind
farms. Energy Procedia 53:290299
26. Chandel SS, Ramasamy P, Murthy KSR (2014) Wind power potential assessment of 12
locations in western Himalayan region of India. Renew Sustain Energy Rev 39:530545
27. Zhang J, Chowdhury S, Messac A (2014) A comprehensive measure of the energy resource:
wind power potential (WPP). Energy Convers Manag 86:388398
28. Jesus F, Menndez M, Guanche R, Losada IJ (2014) A wind chart to characterize potential
offshore wind energy sites. Comput Geosci 71:6272
29. Wang J, Qin S, Jin S, Wu J (2015) Estimation methods review and analysis of offshore
extreme wind speeds and wind energy resources. Renew Sustain Energy Rev 42:2642
30. Ishihara T, Yamaguchi A (2015) Prediction of the extreme wind speed in the mixed climate
region by using Monte Carlo simulation and measure-correlate-predict method. Wind Energ.
31. Vanvyve E, Delle Monache L, Monaghan A, Pinto J (2015) Wind resource estimates with an
analog ensemble approach. Renewable Energy 74:761773
32. Chang TP (2011) Performance comparison of six numerical methods in estimating Weibull
parameters for wind energy application. Appl Energy 88:272282
33. Don Y, Wang J, Jiang H, Shi X (2013) Intelligent optimized wind resource assessment and
wind turbines selection in Huitengxile of Inner Mongolia, China. Appl Energy 109
34. Jimnez PA., Navarro J, Palomares AM and Dudhia J (2015) Mesoscale modeling of offshore
wind turbine wakes at the wind farm resolving scale: a composite-based analysis with the
Weather Research and Forecasting model over Horns Rev, Wind Energy 18: 559566.
35. Lackner MA, Rogers AL, Manwell JF (2008) Uncertainty analysis in MCP-based wind
resource assessment and energy production estimation. J Solar Energy Eng Trans ASME:
AIAA 45th aerospace sciences meeting and exhibit
36. Gonalves-Ageitosa M, Barrera-Escoda A, Baldasano J, Cunillera J (2015) Modelling wind
resources in climate change scenarios in complex terrains. Renewable Energy 76:670678
37. Kjellstrm E, Nikulin G, Hansson U, Strandberg G, Ullerstig A (2011) 21st century changes in
the European climate: uncertainties derived from an ensemble of regional climate model
simulations. Tellus A 63:2440
38. Kalvig S, Gudmestad OT, Winther N (2014) Exploring the gap between best knowledge and
best practice in boundary layer meteorology for offshore wind energy. Wind Energy 17:161
39. Prez-Collazo C, Greaves D, Iglesias G (2015) A review of combined wave and offshore wind
energy. Renew Sustain Energy Rev 42:141153
A Spatiotemporal Methodology for Deep
Offshore Resource Assessment

Ana Estanqueiro, Antnio Couto and Luis Rodrigues Jr.

Abstract Reliable wind measurement campaigns needed to address deep offshore

wind energy deployment are constrained by their prohibitive installation and
maintenance costs. Floating LIDAR systems are a viable alternative to bottom xed
met masts, although have survivability problems during storm seasons. This chapter
presents a methodology, based on already well-established standards, able to reduce
the offshore measurement campaigns duration by relating them with reliable and
low-cost coastal measurements. A two-step calibration procedure, spatiotemporal,
is presented to obtain an accurate characterization of the wind resource in deep
offshore regions taking into account the time shift between the two measurement
points associated with the different atmospheric phenomena propagation. The
methodology is applied in two experimental case studies: the rst one deals with
measurements taken by a LIDAR installed on an islet (10 km away from the coast),
while in the second case study a LIDAR system is mounted on a buoy in a deep
offshore region. Results show that the added temporal calibration step is gradually
more important as the distance between the measurements points increases.
Precision enhancements on the order of 45 % were observed in the expected
annual energy production for a given offshore site. The proposed calibration pro-
cedures presented can be applied in many phases of the offshore development:
resource assessment, power performance evaluation and even for completion of
missing data in a measurement campaign.

Keywords Deep offshore Floating wind turbines  Wind resource assessment 

IEC61400-12-1 Spatiotemporal calibration

A. Estanqueiro (&)  A. Couto  L. Rodrigues Jr.

LNEGNational Laboratory for Energy and Geology,
Estrada do Pao do Lumiar 22, Lisbon, Portugal

Springer International Publishing Switzerland 2016 143

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_8
144 A. Estanqueiro et al.

1 Introduction

Offshore wind energy is a key contributor towards the EUs achievement of its
target of 20 % of the total energy consumption provided by renewable sources by
2020 [1]. The current standard for offshore wind installations in Europe is the
bottom xed wind turbines, a technology that is still limited by high installation
costs and technical limitations that unable the exploitation of the wind resource in
zones with high bathymetry [2]. New solutions, especially for deep offshore
regions, are inevitable and the floating wind turbines (e.g. WindFloat or Hywind)
can overcome the previous limitations and become a cost-effective technology in
the near future [3].
As a side effect with this transition to floating wind turbines further benets are
expected: (1) avoidance of land use conflicts; (2) access to higher wind resources
further offshore thus allowing to minimize the visual impacts; (3) large deep off-
shore wind power plants can be strategically electrically connected to serve major
primary load centres hence reducing costs, losses and the loading of the existing
power transmission system; and (4) quayside assembly and commissioning (sim-
plied offshore installation proceduresno specialized vessels are required) [37].
The installation and maintenance of floating turbines are currently being demon-
strated and current data enables to conclude it is far simpler than for xed to the
sea-bottom technologies [8]. As a consequence, the availability of the few existing
floating wind turbines tends to be noticeably higher than the average availability of
xed to the bottom offshore technologies [9].
On the other hand, the floating wind turbine industry is still in its infancy both in
terms of technology and the associated methodologies. As an example, studies to
demonstrate that floating wind turbines do not show energy generation decit (due to
the wave-induced effects) with respect to onshore or xed to the bottom technologies
are still being conducted [2, 10]. Evaluation procedures for wind turbines (e.g.
resource assessment, site calibration and power curve at their site of operation) are
usually mandatory for conventional wind turbine technologies and are expected to be
replicated for floating wind systems installed and operating in deep offshore regions.
The main challenge in assessing the wind resource and wind turbine perfor-
mance at deep offshore locations consists on installing wind measurement equip-
ment. Met mast for bathymetries higher than 30 m may not be cost effective [11].
The initial investment for a relatively shallow water installation (e.g. Cape Wind,
Massachusetts) gures around $2.5 million, while in bathymetries greater than 30 m
(e.g. FINO 1, Germany) it has exceeded the $10 million mark [11, 12]. Thus, the
actual approach for deep offshore wind resource evaluation is based on the use of
long-term simulations using high-resolution numerical weather prediction models
[7], floating met ocean buoys [6] or the combination of both [13]. However, those
approaches have serious limitations in what concerns their reduced accuracy. The
recent deployment and commercial availability of floating LIDARs and their
installation in ocean buoys brought positive expectations for a cost-effective deep
offshore wind resource assessment.
A Spatiotemporal Methodology for Deep Offshore 145

The floating LiDAR technology is also in its infancy and some drawbacks as the
oscillations induced by the wave motion on the platforms (buoy or vessel) can put
at risk the LIDAR accuracy by introducing deviations on the data gathered. To
overcome the previous limitation, algorithms for motion compensation have been
developed [14]. Hence, the floating LIDAR in latest years has been considered as in
its pre-commercial stage since the baseline stage was already overcome with its
acceptance within the wind energy sector [15]. According to the Offshore Wind
Accelerator program, this technology can represent a capital expenditure (CapEx)
savings up to 90 % when compared with the deployment of a conventional met
mast at high bathymetries [15]. It should be noted that the floating LIDAR becomes
a promising technology not only due to a much lowest investment cost but also due
to lower requirements for corresponding marine license application and ability to
measurement the wind resource at higher heights [12, 15]. Nevertheless, the recent
experience of using floating LIDARs in maritime environments [16] does not
enable them to overcome the operational ocean conditions during the periods of the
year when severe sea storms are expected to occur (e.g. winter in the European
coasts of the Atlantic and hurricane season in the American coast).
Due to the technical challenges and the high costs to obtain wind observations at
sea, the wind resource assessment and power performance procedures for deep
offshore regions needs to be revised and improved. In that sense, this chapter
presents a new methodology for wind resource assessment in deep offshore areas.
The method is inspired by the IEC 61400-12-1 Ed.1 site calibration procedures [17]
with the use of a meteorological mast installed on the onshore coast and the
installation and operation of floating remote devices (e.g., LIDARs) during periods
when sea storms that may affect their integrity are not expected to occur. The
experimental setup to be used and the spatiotemporal calibration methodology
developed may be rst used during the wind resource assessment as described.
Later on, the spatiotemporal calibration achieved can be extended for the (simpler)
application of IEC 61400-12-1 standard, eventually with the removal of the floating
LIDAR during stormy periods occurring between the two different phases of the
deep offshore floating power plant deployment.
In Sect. 2 further details regarding the spatial calibration procedure are provided
and the proposed methodology is presented. In Sect. 3 the results for two case
studies are presented using measurement campaigns in the Portuguese coast.
Finally, in Sect. 4 some conclusions are drawn.

2 Wind Resource Assessment Using a Calibration


Obtaining the in situ data necessary for an accurate offshore wind resource eval-
uation is an expensive process that assumes insurmountable technical difculties for
deep offshore regions. Most of the initial offshore wind parks were deployed
146 A. Estanqueiro et al.

without such local wind resource assessment, hence large deviations between
observed data and estimates of wind power production were identied during the
operational phase of those parks [9]. Reliable wind measures in deep offshore
applications are extremely important considering the total amount of investment
and the high technical and nancial risks associated with the challenging ocean
environment [5].
In the following sections, a new methodology for wind resource assessment in
deep offshore areas is presented. The method is based on the use of the well tested
and widely accepted site calibration procedure of meteorological masts [17] and it is
capable of making the best use of the homogeneous flow characteristics over the
ocean. The measurement setup uses a temporary floating LIDAR measurement
system, thus reducing the experimental campaign costs and the risk of loss/damage
associated with the severe conditions in open ocean environment. In this sense, the
campaign period can cover all wind regimes in the region to be characterized
avoiding exposing the expensive equipment to harsh storm conditions. Moreover,
the experience so far on floating wind turbines shows that for these stormy con-
ditions the wind turbine is expected not to be operating (for safety reasons) hence
the method proposed covers adequately the operation range of floating wind
A two-step method is followed for a spatial calibration on a deep offshore region
using a floating LIDAR as a temporary met mast and one permanent anemometric
mast onshore. The temporary LIDAR is installed at a deep offshore region
characterized by bathymetries that prevent the use of meteorological masts with
xed foundationsand the permanent mast is placed at a coastal onshore location.
In the rst step, a spatial calibration (normally referred as a site calibration in the
IEC power performance standards) based on a linear regression is conducted. The
second step introduces a time shift (lead/lag) between the two measurement points.
This second step presents an improvement in the correct characterization of the
wind resource on deep offshore regions by taking into account the wind direction
associated with different atmospheric phenomena.
Through the application of this methodology, it is possible to assess the spatial
variability of the atmospheric flow in the deep offshore region as well as to
determine the frequency distributions of wind directions and speeds correction

2.1 Spatial CalibrationStep A

The spatial calibration takes place for each direction sector, with the wind speed
correction factors applied as in [17], (Fig. 1).
A Spatiotemporal Methodology for Deep Offshore 147

Fig. 1 Experimental N
campaign setup for the step A N

Floating LIDAR
(temporary met mast)
Onshore permanent

Ucalibrated tk ; hi ai  Umeasured tk \hi 1

where, Umeasured is the 10 min average wind speed at the coastal onshore permanent
mast [m/s], Ucalibrated is the 10 min average wind speed at floating remote sensing
(temporary system location) [m/s], tk is the time variable [s], i is the direction bin
referenced to the North as 0 with an angular dimension of 10 (clockwise) [], i is
the linear correction factor for the i-th direction bin [adim.].

2.2 Spatiotemporal CalibrationStep B

On deep offshore regions where floating wind turbines are to be installed, the large
distance between the measurements points will result in a noticeable time shift
between the events registered. For instance, in a deep offshore location, the distance
D between the point of interest to be measured and the permanent mast located
onshore is typically greater than 5 km.
The spatial calibration as described in [17] is only suitable to adjust two mea-
surements made relatively close to each other (between 2D to 4D). That method-
ology does not take into account the time that a registered phenomenon will take to
propagate from one measurement point to another. A calibration equation more
suitable to extrapolate registered phenomena to another location is obtained by
adding the time shift variable :

Ucalibrated tk ; hi ai  Umeasured tk s\hi 2


s hi ; Uj  cos/ h: 3

where, represents the angle formed by a line segment connecting the two mea-
surement points and the north [] (Fig. 2), is the time advance/delay variable [s]
148 A. Estanqueiro et al.

Fig. 2 Experimental N
campaign setup for the step B N

Floating LIDAR
(temporary met mast)

Onshore permanent mast

Fig. 3 Time association between measurements in step A of the methodology (left) and step B

and Uj is the mean wind speed in the jth bin, within an interval of 4 m/s, according
to the Umeasured [m/s].
It is implied that can assume either positive or negative values depending on
the relative position of the measurement points in relation to the direction of the
wind. The most notable consequence in the comparison to the procedure established
by step A is that average measures taken synchronously are no longer directly
related. Instead, depending on the direction and propagation speed of a phe-
nomenon, the relatively large distance between the measurement points will require
to relate a measurement to another measurement taken at a different instant (before
or after) to be determined in each case (Fig. 3). For a 10 min average wind time
series, the values are rounded to the nearest interval, e.g., for a time shift of 26 min
the value will be 30 min.
Figure 4 shows the results of the methodology application for the time shift
calculation (3) for the two case studies under analysis. Further information
regarding the case studies conguration and data is provided in the following
section. The results depicted enable to understand the temporal shift correction
needed to adjust the data measured in the onshore mast, according to the different
A Spatiotemporal Methodology for Deep Offshore 149

(a) (b)

Fig. 4 Time shift in minutes obtained for each wind direction sector (i), for the two case studies
under analysis, with: a equal to 270 and D equal to 15 kmrst case study; b equal to 320
and D equal to 35.5 kmsecond case study

wind direction and wind speed associated with each meteorological phenomenon.
As expected, the minimums values of time shift (advance/delay) are observed when
the phenomena propagated perpendicular to the two measurement points, and on
the other hand, the maximum values are found when it propagated parallel to the
monitoring systems.
It should be highlighted that the methodology described above intends, not only
to provide wind data for wind resource assessment, but also to create the basis for
the power performance procedure of a floating wind system, and therefore, the time
resolution of 10 min average will be used as required by the actual standards.

2.3 Methodology Evaluation

After the methodology application, the quality of the wind data series obtained for
the two case studies will be assessed through the following statistical parameters
Bias error (BE), mean square error (MSE), root mean square error (RMSE), and
Pearson correlation (r), which are mathematically dened as:

1X N
BE Utemporary tk  Uestimate tk 4
N i1

1X N
MSE Utemporary tk  Uestimate tk 2 5
N i1
150 A. Estanqueiro et al.

u N
u1 X
RMSE t Utemporary tk  Uestimate tk 2 6
N i1
cov Utemporary ; Uestimate
r q
var Utemporary  varUestimate

where N is the total number of wind observations.

3 Application of the Methodology to the Wind

Resource Assessment

As part of the development of this methodology, two case studies were designed
using two analogous experimental setups, i.e. an onshore coastal permanent mast
acting as a permanent measurement system and a floating LIDAR acting as a
temporary measurement system. The main difference found between the case
studies is related to the linear distance (D) among the measuring systems that is:
15 km in the rst case; and 35.5 km in the second case.

3.1 Case Study 1Berlengas: Test of Concept

This section illustrates the application of the methodology previously described

using 1 year of data from an experimental setup composed by an onshore coastal
permanent mast (PS #1) and a temporary offshore LIDAR (TS #1) sited 15 km
away from Portuguese coast (Table 1 and Fig. 5). The offshore data (for 80 m above
mean sea levela.m.s.l) used for this case study was obtained from a LIDAR
system installed in the Berlenga Islet after applying orography correction factors
obtained in wind tunnel scale model [18] in order to clean the influence of the
Islet thus allowing to obtain the undisturbed characteristics of the wind flow over
the sea. Figure 5b also depicts the wind rose at PS #1 location, and the prevailing
winds are from the North/Northwest sectors.
Following the methodology presented in Sect. 2 and according to Eqs. (8.1) and
(8.2), wind flow correction factors () were obtained using the wind data from the
LIDAR installed at the Berlenga Islet acting as a (temporary) floating LIDAR
system and (permanent) coastal met mast during a short synchronized period
between 01 June 2011 and 30 September 2011 (summer period). The remaining
period was used for methodology validation. In Fig. 6, two different periods of the
wind speed data from the temporary system (TS #1), permanent mast (PS #1) and
both wind speed estimates (step AWSE-A; step BWSE-B) with the correction
factors application are presented.
A Spatiotemporal Methodology for Deep Offshore 151

Table 1 Experimental setup data for the rst case study

Monitoring system WGS84 polar Campaign period
Latitude Longitude
Onshore coastal permanent mast (PS #1) 39.36N 9.41W 01.06.201131.05.2012
Temporary offshore LIDAR (TS #1) 39.36N 9.58W

Fig. 5 Location of measurement points at the Portuguese coast for the rst case study. Wind rose
measured during the experimental campaign in the PS #1 is presented on the top left corner of the
detailed map

Table 2 presents the statistical results obtained for the evaluation period (01
October 201131 May 2012) of each method (WSE-A and WSE-B) based on the
data from the floating system for the wind speed estimates and permanent mast
Results depicted in Table 2 indicate that the application of the proposed method
can effectively reduce the wind assessment errors and improve the wind speed
estimations for a deep offshore area. The characterization and implementation of the
time shift step in the methodology enabled to decrease the estimate wind speed
error in, approximately, 40 %.
The evaluation of the economic viability of a wind power plant is performed
through parameters such as the annual energy production (AEP) and the equivalent
number of hours at full capacity (NEPs). In this sense, based on the previous wind
152 A. Estanqueiro et al.



TS #1 WSE-A WSE-B PS #1 Dir (PS #1)

Fig. 6 The methodology application for two different periods using the data from the rst case
study. The black solid line represents the data from the TS #1; the dashed green line represents the
WSE using the step A; the dashed red line represents the WSE using the step B; and the blue
dashed line represents the data from the PS #1. The grey dashed represents the wind direction
observed in the PS #1 (Color gure online)

Table 2 Statistics results Data BE (m/s) MSE (m2/s2) RMSE (m/s) r (%)
steps A and B for the
evaluation period (01 October TS #1 0.1681 1.555 1.247 93.6
201131 May 2012)rst WSEA 0.1649 1.472 1.213 94.1
case study WSEB 0.0688 1.059 1.029 94.8

speed data, the AEP and NEPs values for a generic offshore wind turbine (with
6 MW nominal power) were calculated to determine the deviation of the wind
resource assessment and the wind production estimates over the deep offshore area
under analysis (Table 3. AEP and deviation results for the evaluation period). The
wind turbine availability is assumed to be 100 % and no losses (e.g. electrical) were
taken into account. The deviation results presented in Table 3 were obtained by
calculating the difference from the temporary (in situ) system data and the estimate
wind speed results.
A Spatiotemporal Methodology for Deep Offshore 153

Table 3 AEP and deviation results for the evaluation period (01 October 201131 May 2012)
rst case study
Data AEP (GWh) NEPs (h/year) AEP deviation (GWh) AEP deviation (%)
TS #1 21.318 3,553
WSEA 21.586 3,598 268 1.26
WSEB 21.552 3,592 234 1.10

As expected, for deep offshore locations far from the coast, the results show that
using step B for calculating AEP generates more accurate results than using just the
step A method, that is only suitable for very short distances.
The small difference found between the two steps can be partly explained by:
(a) the intrinsic conditions of the case study since the distance for deep offshore
regions is usually higher than the 15 km considered in the present case study; and
(b) the main characteristics of atmospheric circulation in the area under study. The
atmospheric circulation is influenced by seasonal migration of the mid-latitudes
weather circulation systems [19] with prevailing winds from the North/Northwest
sectors, Fig. 5b. Therefore, the high frequency of occurrences (40 % from the North
sector during the validation period) observed from the wind direction sectors that do
not require the time shift correctionsee Fig. 4acan statistically mitigate the
benet of the step B methodology.

3.2 Case Study 2Viana Do Castelo: Real Offshore


The experimental setup available for the case study was installed within the
DemoWFloat project ( The onshore met mast (PS #2)
was installed in Aguadoura, in Pvoa do Varzim (Portugal). The met mast is
equipped with an Ammonit METEO M40 data logger, cup anemometers (model
Thies First Class Advanced), and wind vanes (model Thies First Class TMR). The
data acquisition system has a sampling rate of 1 Hz, and the data were gathered in
10 min average. Only data measured at 61 m above ground level were used in this
The met mast is located over sand dunes and the roughness in its vicinity is
spatiality distinct, i.e. the west sector is characterized with low roughness values
(identical to values observed in the offshore regions) and for the east side the
roughness values are quite higher due to the existence of small buildings, agri-
culture lands, and trees (Fig. 7).
A buoy from Axys Technologies with a Vindicator LIDAR (www. was placed in an offshore region near Viana do Castelo
acting as a temporary system (hereafter designated as TS #2). To prove and improve
this device for wind resource assessment many trials were successfully performed
154 A. Estanqueiro et al.

Fig. 7 Onshore met mast (PS

#2) location in Aguadoura
view from the Northwest

fullling all the acceptance criteria of the roadmap best practice key performance
indicators [15, 20]. This system is a NOMAD (Navy Oceanographic
Meteorological Automatic Device) buoy equipped with a Vindicator LiDAR,
meteorological sensors, synchronized GPS systems and the power supply system,
Fig. 8. The system allows to infer the wind speed and direction at six different
heights. Only data inferred at 81 m were analysed in this study.
Table 4 presents the location and the wind measurement available period for the
case study, and the location of each system is presented in Fig. 9.
In order to obtain correction factors with statistical meaning, it was not possible
to consider only the summer period for this case study due to the late start of the
experimental campaign (end of July 2014). Fortunately, the floating LIDAR did not
suffer any major incident during its operation in the non-sheltered open Atlantic
waters. The correction factors () for each methodology were acquired with the
wind data from the TS #2 (average wind speed at 81 a.m.s.l.) acting as a temporary
system and from the PS #2 (average wind speed and wind direction) acting as
permanent mast, during a short synchronized period between 29 July 2014 and 28
November 2014. The methodology was evaluated using the data collected in the
period from 29 November 2014 to 30 April 2015.
In Fig. 10 two different periods the wind speed data are presented for the
temporary mast (TS #2), permanent mast (PS #2) together with wind speed esti-
mates (step AWSE-A; step BWSE-B) after the application of correction
Table 5 presents the statistical parameters of wind speed deviations obtained for
the evaluation period (29 November 201430 April 2015). For this error assessment
was used the data from the LIDAR floating system and the wind speed after the
corrections applied to the measures in the permanent meteorological mast.
The statistics results depicted in Table 5 show that also for this case study the
application of the proposed method improves the wind speed estimations at a deep
A Spatiotemporal Methodology for Deep Offshore 155

Fig. 8 Axys Technologies with a Vindicator LIDAR and the schematic conguration of the
equipment on board. Figure extracted from [20]

Table 4 Experimental setup datasecond case study

Monitoring system WGS84 polar Campaign period
Latitude Longitude
Onshore coastal permanent mast (PS #2) 41.44N 8.78W 29.07.201431.04.2015
Temporary offshore LIDAR (TS #2) 41.69N 9.05W
156 A. Estanqueiro et al.

Fig. 9 Location of measurement points at the Portuguese coast for the second case study. Wind
rose measured during the experimental campaign in the PS #2 is presented on the bottom left
corner of the detailed map

offshore area. In this case, the implementation of the time shift (step B) decreases
approximately 30 % the estimate wind speed error when compared with WSE-A.
Using the wind speed data corrected with both steps of the methodology the
AEP and NEPs values for a generic offshore wind turbine (with 6 MW nominal
power) were calculated to assess the deviation in the wind energy estimates for the
area under analysis (Table 6). The deviations presented in Table 6 were obtained by
calculating the difference of energy generation using as input the temporary system
(TS #2) wind data and the estimate wind speed results obtained with the data from
the permanent system (PS #2) duly corrected. In this case study, due to conden-
tiality constraints only the percentage AEP deviation is presented.
In case study 2, as previously observed for case study 1, the results show that
using step B, i.e. the spatiotemporal correction approach, for calculating the AEP
produces more accurate results than using just the step A of the method. In this
case, the difference observed between the two approaches is higher with respect to
the test of the concept; this result suggests that as the distance between the tem-
porary and permanent mast increases the added value of the spatiotemporal cor-
rection introduced by the step B of the methodology also increases.
A Spatiotemporal Methodology for Deep Offshore 157



TS #2 WSE-A WSE-B PS #2 Dir (PS #2)

Fig. 10 Example of the methodology application for two different periods using the dataset from
the second case study. The black solid line represents the data from the TS #2; the dashed green
line represents the WSE using the step A; the dashed red line represents the WSE using the step B;
and the blue dashed line represents the data from the PS #2. The grey dashed represents the wind
direction observed in the PS #2 (Color gure online)

Table 5 Statistics results steps A and B for the evaluation period (29 November 201430 April
2015)second case study
Data BE (m/s) MSE (m2/s2) RMSE (m/s) r (%)
PS #2 1.725 3.539 1.881 68.5
WSEA 0.318 2.969 1.723 77.1
WSEB 0.082 2.833 1.683 72.8

Table 6 AEP and deviation results for the evaluation period (29 November 201430 April 2015)
second case study
Data AEP deviation (%)
WSEA 9.46
WSEB 5.21
158 A. Estanqueiro et al.

0.4 80

BE (m/s)

r (%)
3 months 4 months 5 months 6 months
-0.1 65

-0.3 60
Calibration period of the wind flow correction factors
BE - WSE-A BE - WSE-B r - WSE-A r - WSE-B

Fig. 11 Relation between BE and correlation (r) values with the calibration period, since the
initial results (3 months) until 6 months

3.3 The Impact of the Calibration Period on the Wind

Assessment Accuracy

The main objective of the proposed methodology is to produce the best deep
offshore wind assessment while minimizing the measurement campaign costs and
duration by ensuring to achieve representative correction factors of the wind flow
for offshore regions where floating wind turbines are expected to be installed.
Nevertheless, to understand the impact of the calibration periods duration in the
accuracy of the methodology, sensitivity tests were performed by increasing the
data available for this procedure (Fig. 11).
From Fig. 11 it is possible to infer that for both steps of the methodology, the BE
and correlation values are dependent from the calibration period and as such the
floating Lidar (temporary system) operation should be as long as possible.
Nevertheless, the BE in the step B presents a lower dependency of the calibration
period duration when compared with step A. The absolute improvement when
increasing from 3 months to 6 months the calibration period is 66 % in the case of
step A, while in the step B this value is only 38 %. Based on the results presented in
Fig. 11 one may conclude that an offshore experimental campaign with a duration
above 5 months enables to obtain correlation factors above 75 % with BE below
0.2, what suggests the suitability of the installation of floating LIDARs in the
spring/summer period in the Atlantic coasts.

4 Conclusions

New methodologies, capable of assessing the offshore wind resource in a more cost
efcient manner and having added accuracy are strongly needed to full the
European targets for this technology. This chapter presents a new methodology for
A Spatiotemporal Methodology for Deep Offshore 159

wind resource assessment that enablesbased on data gathered at an onshore mast

(permanent mast) and a temporary floating LIDARto obtain correction factors for
the onshore met mast to make it representative of the wind flow in deep offshore
regions where floating wind turbines are foreseen to be installed. The method builds
on and develops the site calibration concept of the IEC 61400-12-1 standard, widely
used for power performance assessment of wind turbines.
To interpret and characterize the quality of the new methodology presented, a
statistical evaluation was performed using the common statistical parameters.
Results show that for an offshore wind resource analysis the time shift introduction
on the correction factor calculation enables a remarkable reduction on the errors of
the wind speed estimates against the simple spatial calibration method. The
implementation of the time shift allows to obtain AEP deviations of 1.1 % for the
test of concept dataset and +5.21 % for the deep offshore real operation case
study analysed. The experimental setup measurement system installed during the
EC FP7 DemoWFloat project was a key step to validate the present methodology.
Through the data gathered during the experimental campaign offshore Northern
Portugal, it was possible to demonstrate the effectiveness of the methodology,
especially at large distances from the coast where the deployment of floating wind
turbines is likely to occur.
Although further research is desirable to enable its full validation (e.g. for dif-
ferent weather regimes and coastal characteristics), the proposed methodology
presents a promising improvement in the accuracy of actual deep offshore wind
energy resource assessment and allows to increase the precision of the resource
assessment and the validation of AEP estimates during the due diligence procedures
of deep offshore floating wind power plants, thus reducing the risk of the invest-
ment and promoting the deployment of the deep offshore wind sector. Moreover,
the correction factors obtained during this wind assessment experimental campaign
may be used for many different purposes as recovery of periods of missing data and
power performance evaluation, thus further enabling the reduction of costs asso-
ciated with the deployment of floating wind power plants.

Acknowledgement This work was partially funded by the European Commission FP7 project
DEMOWFLOATDemonstration of the WindFloat Technology, Grant Agreement number:
ENER/FP7/296050/DEMOWFLOAT. The authors gratefully acknowledge EDP-Inovao, Repsol
and Principle Power for granting access to the experimental wind data that enabled the validation
of the methodology in real offshore conditions and LNEG for co-nancing and providing the
conditions to conduct this research. Luis Rodrigues Jr. acknowledges support from the Portuguese
Foundation for Science and Technology (FCT) through the MIT Portugal Program.


1. EC, Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on
the promotion of the use of energy from renewable sources and amending and subsequently
repealing Directives 2001/77/EC and 2003/30/EC. Ofcial J Eur Union, 2009
160 A. Estanqueiro et al.

2. Roddier D, Cermelli C, Aubault A, Weinstein A (2010) WindFloat: a floating foundation for

offshore wind turbines. J Renew Sustain Energy 2:134
3. Castro-Santos L, Diaz-Casas V (2015) Sensitivity analysis of floating offshore wind farms.
Energy Convers Manag 101:271277
4. Zhang P, Ding H, Le C, Huang X (2013) Motion analysis on integrated transportation
technique for offshore wind turbines. J Renew Sustain Energy 053117 5(5):115
5. Carvalho D, Rocha A, Gmez-Gesteira M, Alvarez I, Silva Santos C (2013) Comparison
between CCMP, QuikSCAT and buoy winds along the Iberian Peninsula coast. Remote Sens
Environ 137:173183
6. AWS Truewind L (2010) New Yorks offshore wind energy development potential in the
Great Lakes : feasibility study, p 169
7. Costa P, Simes T, Estanqueiro A (2006) Assessment of the sustainable offshore wind
potential in Portugal. In: European wind energy conference (EWEC), pp 18
8. Silva N, Estanqueiro A (2013) Impact of weather conditions on the windows of opportunity
for operation of offshore wind farms in Portugal. Wind Eng 37(3):257268
9. IEA (2005) Offshore wind experiences. International Energy Agency, 2005
10. Sant T, Bonnici D, Farrugia R, Micallef D (2015) Measurements and modelling of the power
performance of a model floating wind turbine under controlled conditions. Wind Energy
11. Standridge C, Zeitler D, Nieves Y, Turnage TJ, Nordman E (2012) Validation of a
buoy-mounted laser wind sensor and deployment in Lake Michigan. Michigan Alternative and
Renewable Energy Center, 2012
12. Hsuan C, Tasi YS, Ke JH, Prahmana R, Chen KJ, Lin TH (2014) Validation and
measurements of floating LiDAR for nearshore wind resource assessment application. In:
International conference on applied energy, ICAE2014 61:16991702
13. Sheridan B, Baker SD, Pearre NS, Firestone J, Kempton W (2012) Calculating the offshore
wind power resource: robust assessment methods applied to the U.S. Atlantic Coast. Renew
Energy 43:224233
14. Gottschall J, Wolken-Mhlmann G, Viergutz T, Lange B (2014) Results and conclusions of a
floating-lidar offshore test. Energy Procedia 53:156161
15. Carbon Trust (2013) Carbon trust offshore wind accelerator roadmap for the commercial
acceptance of floating LIDAR technology, 2013
16. Fernandes M, Marujo R, Costa P, Estanqueiro A (2011) Validation report on local tests site
(Berlenga) : deliverable 4.8, 2011
17. IEC (2005) International standard IEC 61400-12-1 Ed.1: power performance measurements of
electricity producing wind turbines. International Electrotechnical Commission, 2005
18. Silva J, Marques da Silva F, Couto A, Estanqueiro A (2015) A method to correct the flow
distortion of offshore wind data using CFD simulation and experimental wind tunnel tests.
J Wind Eng Ind Aerodyn 140:8794
19. Couto A, Costa P, Rodrigues L, Lopes VV, Estanqueiro A (2015) Impact of weather regimes
on the wind power ramp forecast in Portugal. IEEE Trans Sustain Energy 6(3):934942
20. Howe G (2014) Developing a buoy-based offshore wind resource assessment system. Sea
Technol Megazine 55(2):4146
Tools for Ocean Energy Maritime Spatial

Paulo Costa, Teresa Simes and Ana Estanqueiro

Abstract Offshore wind power has been in recent years a highly debated topic and
subject to investigation by the several players in the renewable energy sector. The
advantages of this form of renewable energy are well known being some of the
most important aspects the large wind resource and available area for wind farms
installation. Nevertheless, the planning of an offshore wind energy project needs to
be addressed carefully so as not to interfere with other economic activities or
sensitive environmentally protected areas, among other constraints. Similarly,
although the offshore wind resource is naturally higher than the resource onshore,
the investment costs (including operation and maintenance) are also considerably
higher, hence the need to establish straightforward and transparent methodologies
for offshore wind planning. Most of the methodologies used for planning purposes
are based on geographic information systems (GIS) due to its powerful capabilities
in managing, formatting and modelling large amounts of georeferenced data usually
involved in planning procedures. This chapter presents a methodology for offshore
wind energy planning taking into consideration the several aspects of sea use and
the main restrictions which can be relevant for wind farm installation. The
methodology consists on the use of a GIS to manage the required information
(resource and restrictions maps) and to develop a set of tools that enable the user to
add, modify and model the information according to his objectives.

Keywords Ocean maritime spatial planning  Offshore wind  Geographic infor-

mation system

P. Costa  T. Simes  A. Estanqueiro (&)

LNEGNational Laboratory for Energy and Geology, Estrada do Pao do Lumiar 22,
Lisbon, Portugal
P. Costa
T. Simes

Springer International Publishing Switzerland 2016 161

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_9
162 P. Costa et al.

1 Introduction

In the latest years several research groups have dedicated their studies to offshore
wind energy in its different areas with very stimulating results which gave rise to a
new era in funding entities. Examples are the European funded projects TradeWind
[1], WindSec [2], UpWind [3], OffshoreGrid [4], Orecca [5], SEANERGY 2020 [6]
and WindFloat [7]. In the same way, the development and testing of new wind
resource assessment methodologies has also caught the attention of researchers in
the area of offshore wind energy since there is the need to nd new techniques that
can overcome the extremely high costs associated with conducting offshore wind
experimental campaigns. In the period 20082012 a European funded project FP7
NORSEWInD [8] took place and had as main objective the search for alterna-
tives for the offshore wind measurements using satellite data, LiDAR sensors and
coastal anemometric stations as data sources for wind resource assessment. The
regions under study within this research project were the North Sea and the Baltic
Sea, being the validation of the methodologies performed with a case study in a
small Portuguese islandBerlengas [8].
Side by side with all the advances in technology and wind resource assessment is
the planning of offshore wind resources. Researchers all over the world have
gathered in groups to develop planning methodologies that can be applied to off-
shore wind energy assessment taking into account the different constraints to the
deployment of offshore wind power plants especially to be integrated in the marine
spatial planning of the different countries.
Offshore wind farms are most of the times planned where the wind conditions
are favourable and the visual impact is reduced. Nevertheless, the installation of
offshore wind turbines faces a large set of constraints that are related to environ-
mental issues, besides the visual impact, for example (i) seabed protection and
geology, noise emissions and effects on ecosystems, among others; (ii) physical
conditionsavailable area, deployment logistics (existence of ports and shipyards
with the right conditions, among others); and (iii) economic aspects as high
equipment, installation and operation and maintenance costs. The next sections
present an overview of the work developed in the area of offshore wind planning
that resulted in a systematic methodology for the identication of suitable areas for
offshore wind energy system deployment as well as its application to a case study in

2 Offshore Wind PlanningOverview

The use of geographic information systems (GIS) for the study of renewable energy
has been frequently used, either for planning or for handling and processing large
amounts of information. The planning of energy uses from renewable sources
requires a high number of conditions for its study and development, which involves
Tools for Ocean Energy Maritime Spatial Planning 163

the handling of a wide range of maps and data tables which often appear in varied
formats and with different coordinate systems. The possibility to process and model
different types of information on a wide range of formats and georeferencing sys-
tems makes the GIS platforms valuable and powerful tools for planning renewable
energy exploitations in a macroscale perspective.
Offshore wind farms are in a rst approach located in areas with good wind
conditions and low environmental impacts. Also, there is the need to avoid conflicts
with other economical uses such as sheries, mineral exploitation or other.
Therefore, a careful identication of the most suitable locations is needed.
Several authors have dedicated their studies to the planning of wind and tech-
nologies installation using GIS platforms and related techniques. Hong and Mller
[9] have identied the offshore wind energy potential in China using technical,
spatial and economic constraints. These authors have developed a methodology for
the estimation of the wind potential with the help of WindPro software [10] and GIS
to dene and apply a set of criteria for offshore wind energy development.
Some authors gave special attention to economic aspects such as the establish-
ment of economic indicatorsLCOE, NPV and IRR. Punt et al. [11] have devel-
oped a methodology for the spatial planning of offshore wind farms although GIS
was not an option for this work. These authors have developed a probabilistic
model using economic and ecological constraints in order to nd the most suitable
locations for offshore wind farms installation. In the work of Punt et al. [11] the
ecological impacts of the wind turbines are evaluated and the available wind
potential for offshore wind installations is estimated for the Dutch exclusive eco-
nomic zone (EEZ).
Gimpel et al. [12] have developed a methodology for the co-existence of off-
shore wind farms and aquaculture in the German EEZ. The authors have combined
GIS and multi-criteria techniques to rank suitable co-sites in the North Sea.
Different spatial scenarios were evaluated in order to support efcient and sus-
tainable marine spatial management strategies. Gimpel et al. [12] have exploited the
idea that the sites for the installation of wind turbines can also be used for other
economic activities such as the aquaculture (IMTAIntegrated Multi-Trophic
Aquaculture) of sh species, bivalves and seaweeds.
The offshore planning methodologies can also be applied in the development of
interactive decision-making tools. As an example, Mekonnen and Gorsevski [13]
have developed a web tool for offshore wind farm suitability in an America region
lake Erie in Ohio. The authors have developed a toolPGISintegrating GIS and
decision-making tools that have the main objective of involving different stake-
holders and the public in general for solving planning problems and achieve con-
sensus among the different planning actors. This integration of decision-making
tools and GIS capabilities is known as SDSSSpatial decision support system [13].
In Portugal, several studies have been conducted in recent years for the devel-
opment of planning tools with the main objective of planning the integration of
renewable energies. That work included the development of a set of interactive GIS
tools that enable the user to identify suitable locations for the development of wind
energy projects having into account the restrictions for the installation of wind
164 P. Costa et al.

turbines, onshore and offshore. All the information is mapped, including wind
resource, environment, electric grid and other, and a set of criteria is dened for the
location of the wind turbines [14]. The above-mentioned methodology is presented
in the next sections and applied to the case study of the Portuguese coast. The main
characteristics of the developed methodology are its ability to be replicated to other
geographies and the simplicity of its application to large arease.g. district,
country or sea region.

3 Spatial Planning Methodology

A spatial planning methodology consists on a set of spatial information and layer

operation constraints that are managed by a GIS platform capable to connect and
relate the offshore spatial information usually designated as layers. The outputs from
a GIS spatial planning methodology are the georeferenced areas and the most suitable
regions for the deployment of offshore wind parks, and the information on the
available sustainable offshore wind potential for a region of interesta country or a
specic area. Also, a spatial planning methodology can be applied to offshore [14],
or be adapted to operate under onshore situations [15].
This section describes a planning methodology developed under a GIS envi-
ronment which can be applied to most offshore regions. The spatial layer infor-
mation needed for the GIS planning methodology can be distinguished into two
categories. The rst category is related to the wind resource. In fact an offshore
wind resource map, usually referred as the offshore wind atlas, is needed to identify
the wind resource sites across the coastal areas that are the most suitable to wind
park deployment and which are used to compute the sustainable and available
offshore wind potential. The wind resource map is developed with the help of wind
data (measured or generated) and numerical modelling (e.g. mesoscale) and should
also include the energy produced by a certain wind turbine model. The wind energy
nal map can be subjected to a criteria threshold condition that pre-establishes, for
example, the minimum value for the economic sustainability that guarantees the
protability of the project. Generally, the wind resource map can be presented in
terms of the wind parameters (mean wind speed, power density, Weibull parame-
ters) or in energy terms in power production or in annual equivalent number of
hours at full capacity (NEPs).
The second category takes into account the information regarding sea con-
straints. The existence of a database with the most signicant and important sea
constraints that are present in the area under study is crucial so that the GIS spatial
planning methodology will operate only over the free areas, i.e. the areas that have
no constraints to the deployment of offshore wind turbines. As an example of sea
constraints databases are the sea environmental protected areas, navigation chan-
nels, submarine cables and anchorage sites, among others.
Distance constraints can also be set and considered for the renement of the GIS
planning tool under operation. For example, establishing a threshold distance from
Tools for Ocean Energy Maritime Spatial Planning 165

the coast, existing navigation ports or electric grid connection point can be used to
dene the visibility distance of the wind farm from a certain point on the coast and
establish the criterion for visual impact.
Tables 1 and 2 present all the information that should satisfy the spatial selection
criteria for the GIS planning methodology according the two presented categories.
A simplied diagram of the GIS planning methodology is presented in Fig. 1.
The diagram shows the main steps on the GIS planning under operation and takes
into account the criteria mentioned above.
According to Fig. 1, spatial raster datasets are needed for computations, such as
the bathymetry and the georeferenced offshore wind energy resource atlas. As
expected, raster datasets must have the same spatial dimensions and also the same
spatial resolution. These requirements are guaranteed by the GIS and the datasets
are re-projected into a common standard geographic coordinate system.
In the onshore case, the terrains slope is an important restriction and plays an
important twofold role in wind turbines performance concerning the energy pro-
duction and the structural safety of the wind turbine according IEC 61400-12-1
[16]. Based on eld experience for onshore projects, slope values higher than 15 %
should be excluded from analysis. For the offshore case the bathymetry is handled
in the same way (as terrain) and the same value may be considered as a limit for the
offshore wind turbines installation in near shore.
The restrictions and constraints datasets are used in shape le format or similar
(e.g. polygon layers) in order to exclude non-free areas through blanking operations
and cross area classication. In the case of economic activities constraints, sea
occupation factors are applied in the areas already under other uses, e.g. sheries. In
this case, a numerical constant parameter is established with a value between 0 and
1 according to the degree of usability of the area under analysis, where 0 means
totally exclusion and 1 means totally available area.
The constant factor assigned to each economic area is usually established in a
case-by-case analysis, but the choice of this factor should be performed according
to the sea legislation, international directives and different sectors regulations
directly involved in the areas economic activities.
The calculation of the sustainable and available offshore wind potential is usu-
ally better rened if a georeferenced database of all the operating and projected
offshore wind farms with the correspondent capacity (in MW) is available. This
way, the GIS planning methodology is capable to evaluate and provide the available
sustainable offshore wind energy potential still available for future offshore

Table 1 Wind energy resource requirements

Offshore wind atlas A spatial wind energy map from a wind potential atlas
Reference nominal power of the wind turbine model
Minimum NEPS (h/year)economic protability purposes
Equivalent energy losses (MWh)
166 P. Costa et al.

Table 2 Sea constraints database

Bathymetry Bathymetry slope
Environmental Protected sea areas
Restrictions Military areas, seismic faults, seabed types, navigation channels,
maritime buoys, anchorage locations, electrical submarine cables, others
Economic sea Fisheries and other economic activities
Grid connection Available capacity, voltage level
Accesses Maritime ports, coastal, visibility area, others

Fig. 1 GIS planning methodology scheme under operation to estimate suitable offshore areas and
the sustainable offshore wind energy

The sustainable wind potential capacity is obtained with the following


Pi abci gi ei  wi 1

where Pi represents the sustainable wind power (in MW) per identied area (or
polygon) and

a EL 2
Tools for Ocean Energy Maritime Spatial Planning 167

b 3
ci SOi 4

gi 5

ei NEPsji 6

wi MWji 7

The parameters in Eqs. (27) are:

EL coefcient for equivalent potential energy losses;

Pot nominal power of the reference wind turbine (MW) used to compute the
offshore wind energy resource map (hours per year);
8760 the number of hours in a year;
SOi economic occupation factor for polygon i;
dx dy pixel area in the wind energy resource map;
dx dy D2 sea area required for the installation of each offshore wind turbine,
expressed as a multiple of rotor diameter D, where x, y represent the
minimum distances for crosswind and along wind directions,
NEPsji NEPs map raster value after applying all GIS exclusions in grid point
j inside polygon i;
Ni number of grid points inside polygon i;
MWji nominal power (MW) from each wind farm operating inside polygon i;
k total number of wind farms inside polygon i.

4 Application of the MethodologyThe Portuguese Case


To test the GIS planning methodology a database of different spatial layers rep-
resenting the restrictions and constraints for the Portuguese coast was considered.
A high-resolution offshore wind resource assessment atlas and a high-resolution
bathymetry database were also used in the study. According to the two main
categories described in the previous section, a detailed explanation of the infor-
mation used for the GIS planning methodology with the data collected for the
Portuguese case study is presented in the next section.
168 P. Costa et al.

4.1 Offshore Wind Resource Assessment

The offshore wind resource assessment for the Portuguese case study was derived
from previous research studies [1621] where a high-resolution (3 3 km) long-term
simulation with a sophisticated atmospheric mesoscale model MM5 [22] was used.
This model is suitable to simulate atmospheric coastal effects caused by local
thermal or shearing phenomena circulations (e.g. sea-breeze circulations) among
others on a staggered sigma [23] coordinate grid. The MM5 model was coupled with
a complete year of reanalysis [24] data from NCAR/NCEPs mass storage systems
with 6-h intervals. Only one offshore wind turbine model was simulated in order to
obtain results for wind power production. The simulated offshore wind turbine
model was VESTAS V80 with 2000 kW of rated power and a hub height of 80 m.
The wind power results were converted into the well-known wind power parameter
of NEPsyearly number of hours at full capacity (h/year) processed for the same
height as the turbines hub. The tuning parameters to set the MM5 model were
chosen according to Costa et al. [20] that also present the validation study for this
long-term simulation. An inter-annual variability factor was also computed and it
was considered in the models results. The determination of the inter-annual vari-
ability factor is also detailed in [20]. In Fig. 2 the offshore wind resource assessment
obtained with the test turbine model for the hub height at 80 m is presented. The
wind resource map was clipped between the shoreline and the 200 m bathimetric
countour, and the resource map is plotted for NEPs greater or equal to 2900 h/year.
The obtained offshore wind resource for the Portuguese coast is favourable for
the development of offshore wind parks. Values above 2700 h/year indicate eco-
nomic feasibility, if the Portuguese Government adopts a feed-in Tariff (FIT) plan
similar to the ones being used in other European countries.

4.2 Sea Constraints and Restrictions in Continental


There are several sea constraints around the shoreline of Continental Portugal. They
can be categorized by; the wave energy pilot zone, seismic faults, seabed types,
navigation channels, maritime buoys, military zones, anchorage locations, eco-
nomic and environmental protection zones and submarine electrical cables. The
information for each constraint was transformed into a shape le and imported into
the GIS as layer. All the layers were then merged into a single restrictions layer in
order to facilitate the analysis.
Figures 3 and 4 depict the constraints in four separate maps and Fig. 5 depicts
the overall constraints merged into a single layer used in the evaluation of the
sustainable offshore wind power.
Tools for Ocean Energy Maritime Spatial Planning 169

Fig. 2 Offshore wind

resource for Continental
Portugal (h/year) computed
for 80 m a.g.l. with
inter-annual variability

In terms of the Portuguese legislation, the sea economic areas are free to be
crossed and no conflicts were appointed by the sea economic users, therefore a
constant sea occupation factor equal to 1 was applied to all the available sea space.
The seabed type is a restriction applied for bottom-xed wind turbine types
whose technology can support turbine deployments up to 40 m depth. Generally, a
seabed constituted by almost 100 % sand is ideal for xing the turbine structures.
For other technology types, namely floating devices, this restriction is not applied
since the mooring technology of is independent of the type of seabed.

4.3 Bathymetry of the Portuguese Continental Platform

The bathymetry for the Portuguese Continental platform was processed from the
International database GEBCO [25]. The bathymetry database used corresponded to
a 500 500 m high resolution pre-processed from the original 30 database
170 P. Costa et al.

Fig. 3 Illustration of the sea constraints. Left image illustrates the economical/environmental
areas, the pilot zone and the military restricted zones. The image on the right depicts the navigation
channels, buoys, anchorage and platform areas

Fig. 4 Illustration of the sea constraints. Left image depicts the seabed type and seismic faults.
The image on the right depicts areas with electrical cables
Tools for Ocean Energy Maritime Spatial Planning 171

Fig. 5 Illustration of the sea

constraints merged into a
single layer around the
shoreline of Continental
Portugal used into the
evaluation of the sustainable
offshore wind power

available at GEBGO website ( Figure 6 illustrates the

bathymetry map around Continental Portugal.
Figure 6 presents the bathymetry of the Portuguese Continental platform until
300 m. Most of the territorys bathymetry can be classied as smooth and
therefore it is favourable for the deployment of offshore wind parks. From the GIS
planning methodology presented in the previous section, it was possible to compute
the slope values (in percentage) and the resultant raster dataset is illustrated in
Fig. 7. The depicted slope values in Fig. 7 were truncated at 200 m depth which
corresponds to the deepest level capable to support the most modern floating off-
shore wind turbines.
From Fig. 7 the slopes up to 200 m depth at 500 500 m spatial resolution in
almost all available space are lower than 8 % and this constitutes another favourable
condition for the development of offshore wind parks in Portugal. This value is
below the maximum recommended for the deployment of wind turbines, which is
15 % due to structural safety precautions.
172 P. Costa et al.

Fig. 6 Bathymetry depths over Continental Portugal. Spatial resolution of 500 500 m. Orange
contour line represents the 300 m level depth. (Source (Color gure online)

4.4 Scenario for Fixed and Floating Technology Devices

Nowadays the state-of-the-art wind turbine technology devices can be split into two
categories: the bottom-xed and floating wind turbines. Currently, the bottom-xed
technology can go to 40 m depth, while floating devices can reach 200 m depth
depending on the mooring system technology used. Anyway the floating wind
turbines are still considered a technology niche that produces prototypes under
operation in real ocean conditions, with the rst MW-sized floating wind turbine
installed on the coast of Norway in the summer of 2009 [26] and the WindFloat [7]
installed and operating since December 2011 in Aguadoura, a Northern coastal
region of Continental Portugal. Most recently, a project to build a pilot offshore
wind farm in Scotland [27] is undergoing.
Tools for Ocean Energy Maritime Spatial Planning 173

Fig. 7 Bathymetry slope up

to 200 m depth. Spatial
resolution of 500 500 m

4.4.1 The Near-Shore Technology

The bottom-xed offshore wind technology is now available in most of the main
turbine manufacturers. The current structure for those types of wind converters
already enables their installation on shallow waters with depths up to 40 m and
slopes below 15 %.
Using the GIS planning methodology it was possible to produce the availability
of the wind power resource up 40 m depth as depicted in Fig. 8. Once the 40-m
depth contour is very close to the shoreline, a query xing the NEPs values greater
or equal to 2700 h/year was established. This limit was considered by assuming for
Portugal a FIT for offshore in the same order than the one followed by other
European countries with offshore wind projects (e.g. UK, Spain or Italy), which
ensures economic feasibility. From Fig. 8 it is clear that the most suitable areas for
wind power production are mainly located in front of the main cities and con-
sumption centres, therefore covered by electric grid connections to the Portuguese
transmission grid.
174 P. Costa et al.

Fig. 8 Existing near-shore

wind potential in Continental
Portugal for equivalent
number of hours at full
capacity above 2700 h/year
(depth <40 m, turbine

Table 3 Near-shore wind power sustainable potential for Continental Portugal (2 MW test
Locals Availability of the sustainable offshore wind power capacity
in Portugal
NEPs >= 2700 h/year (MW) NEPs >= 2900 h/year (MW)
Viana Castelo and Porto 1200 550
Figueira da Foz 1300 100
and neighbourhood
Peniche and Lisboa 950 650
South regions 50 10
(very small regions, not visible
in the map)
Total 3500 1400
Tools for Ocean Energy Maritime Spatial Planning 175

Table 3 reports the sustainable offshore wind capacity obtained by the GIS
planning methodology according to the mathematical formulation (17) at each area
and to the macro-regions pre-identied in Fig. 8. In Table 3 results for other
NEPs value condition (NEPs greater or equal to 2900 h/year) are also considered.
This value corresponds to a more conservative scenario in order to ensure the
economic viability of the project. Observing the values obtained for the
bottom-xed turbine technology, it is expected that Portugal has a sustainable
offshore wind capacity of 3500 MW ensured by an offshore wind production of
2700 h/year (considering the current European offshore tariffs) and for a more
restrictive scenario, with NEPS higher than 2900 h/year the country has a sus-
tainable offshore capacity of 1400 MW.

4.4.2 Deep-Offshore Floating Technology

The floating offshore wind turbines are still in a phase of research and denition of
the best conguration to tackle the interference between wind and wave effects
while maintaining (or even reducing) the installation costs, when compared with the
already existing solutions. It is expected thatwith some optimismby 2020, a
floating offshore wind turbine is already available for the offshore wind market.
Considering this new technology, the GIS planning methodology was used to query
the wind resource between 40 and 200 m depth, this time without considering the
seabed type constraints. The results obtained for the wind power resource, at deep
open sea, is illustrated in Fig. 9.
In this case, the estimated wind power resource is almost unlimited. The GIS
planning methodology analysis shows a total capacity over 40 GW of wind capacity
available for all the area conned between depths from 40 to 200 m.

4.5 GIS Interactive Toolbar

In order to make this a more effective methodology and fully enjoy the GIS
potential, it was developed an interactive tool which enables to upgrade, modify and
analyse the information used and the obtained results. As usual in this kind of
analysis, the amount of mapped and georeferenced information is huge and it is
very important that the user may change and/or update the input criteria if the
objective of the work changes, whether in what concerns the initial conditions and
input data, or the geographic location of the project.
Also, the processing of the information towards the results includes the appli-
cation of a large set of tools included in the GIS which forces the user to repeat the
whole process every time he needs to perform a new analysis. This way, a toolbar
was developed in order to enable grouping the whole sequence of actions that will
lead to the results. Also, some additional options can be introduced in this kind of
analysis, such as the existence of ports and shipyards with suitable conditions for
176 P. Costa et al.

Fig. 9 Offshore wind power resource assessment availability in Continental Portugal for open sea
areas between 40 and 200 m depth (turbine VESTAS V802 MW)

the deployment of the wind turbines and the quantication of the number of wind
turbines that can be installed in a certain region.
The toolbar was developed using an option in the GIS platform customize
modewhere the command actions for the whole sequence were added in a
structured and sequential way and that work independently of each other in a way
that the user can perform some actions in the middle of the process without being
obliged to execute the rst part of the analysis if he does not need to. Figure 10
presents an example of the developed toolbar for a 2020 scenario where ve
commands were introduced.
This toolbar enables the user to run the process for the whole country or only for
a small coastal area, introducing (or not) the information on ports and shipyards in
its analysis, and also to quantify the number of units one can install in a certain
region. Figure 11 presents an example of the results that can be achieved with this
Tools for Ocean Energy Maritime Spatial Planning 177

The organization of the required information and models in such a toolbar as

exemplied in Fig. 10 provides ease of handling of the whole process and a more
pleasant interface for the user.

Fig. 10 Toolbar with ve command buttons enabling the process to run in several steps [28]

Fig. 11 Number of units that can be installed in a certain region using the presented
methodology [28]
178 P. Costa et al.

5 Conclusions

In this chapter an overview of the work developed in the area of offshore wind
energy systems planning was presented. A methodology using a GIS platform and a
set of criteria for the selection of suitable areas for the deployment of offshore wind
energy systems was presented together with an application to the Portuguese coast
in order to illustrate the methods and information that are required in this type of
The methodology enables the identication of suitable areas for the installation
of offshore wind turbines as well as an estimation of the available offshore wind
potential within those areas. The application of this methodology presents several
advantages: (i) it enables to process large areas and a large set of georeferenced
information; (ii) it is very user friendly as the user may easily change the input data;
and (iii) it can be applied to any other geographic region as long as the needed input
information is available.
The application of the methodology consists of a large number of actions that
can be programmed and organized into a GIS toolbar so the user executes the base
models of the whole analysis in a structured sequence to achieve the targeted
The application of the methodology to the Portuguese case study enables to
conclude on the adequability of such a tool to the original purposes, i.e. the
selection of areas for offshore wind deployment taking into consideration the
existing technical, environmental and economic constraints.

Acknowledgements The authors want to acknowledge LNEG for co-nancing the research that
supported the work presented and providing the means and equipment that enabled its develop-
ment; the European Commission (e.g. FP7 Norsewind and IEE Seanergy 2020) and FCT (project
Roadmap WW) for co-nancing part of this work.
The authors also want to express their appreciation to Drio Gambo for the valuable contri-
bution in the development the presented toolbar.


1. The TradeWind project. Available at Accessed in 21 Aug 2015

2. The WindSec project. Available at Accessed in 21 Aug 2015
3. The UpWind project. Available at Accessed in 21 Aug 2015
4. The OffshoreGrid project. Available at Accessed in 21 Aug 2015
5. The Orecca project. Available at Accessed in 21 Aug 2015
6. The SEANERGY 2020 project. Available at Accessed in 21
Aug 2015
7. The WindFloat project. Available at Accessed in 21 Aug 2015
8. The NORSEWInD project. Available at Accessed in 21 Aug 2015
9. Hong L, Moller B (2015) Offshore wind energy potential in China: under technical, spatial and
economic constraints. Energy 20:44824491. doi:10.1016/
10. Available at Accessed in 21 Aug 2015
Tools for Ocean Energy Maritime Spatial Planning 179

11. Punt MJ, Groeneveld RA, van Ierland EC, Stel JH (2009) Spatial planning of offshore wind
farms: a windfall to marine environmental protection? Ecol Econ 69:93103. doi:10.1016/j.
12. Gimpel A, Stelzenmuller V, Grote B, Buck BH, Floeter J, Nunez-Riboni I, Pogoda B,
Temming A (2015) A GIS modelling framework to evaluate marine spatial planning scenarios:
co-location of offshore wind farms and aquaculture in the German EEZ. Mar Policy 55:102
115. doi:10.1016/j.marpol.2015.01.012
13. Mekonnen AD, Gorsevski PV (2015) A web-based participatory GIS (PGIS) for offshore wind
farm suitability within Lake Erie, Ohio. Renew Sustain Energy Rev 41:16277. doi:10.1016/j.
14. Costa P, Simes T, Estanqueiro A (2010) Sustainable offshore wind potential in continental
Portugal. In: Proceedings workshop oceans as a source of energy. Academia de Engenharia,
Berlin-Brandenburgische der Wissenschaften. Lisbon, Portugal, pp 4043
15. Simes T, Costa P, Estanqueiro A (2009) A methodology for the identication of the
sustainable wind potential. The Portuguese case study. published with reviewers. In: IEEE
power system conference and exposition meeting. Seattle, Washington, EUA
16. IEC 61400-12-1 (1998) Wind turbine generator systemspart 1: Safety requirements. Int
Electro Tech Comm
17. Costa P (2004) Atlas do Potencial Elico para Portugal Continental. 06/2004, Degree: Master,
Supervisor: A. I. Estanqueiro, Pedro Miranda, lvaro Rodrigues, Faculty of Sciences,
University of Lisbon, Portugal
18. Costa P, Estanqueiro A (2004) Atlas do Potencial Elico para Portugal Continental. CD-ROM
publication, INETI/DER. ISBN: 972-676-196-4
19. Costa P, Estanqueiro A (2006) Assessment of the sustainable offshore wind potential in
Portugal. In: Publication in the proceedings of the European wind energy conference (EWEC),
20. Costa P, Miranda P, Estanqueiro A (2006) Development and validation of the Portuguese wind
atlas. In: Publication in the proceedings of the European wind energy conference (EWEC),
21. Costa P, Miranda P, Estanqueiro A (2006) Validation of the Portuguese wind atlas. Conf in
Mtodos Numricos em Mecnica dos Fluidos e Termodinmica, Fac Sci and Technol, Monte
da Caparica, Portugal
22. Grell GA, Dudhia J, Stauffer DR (1995) A description of the fth-generation Penn
State/NCAR mesoscale model (MM5) NCAR Technical Note, NCAR/TN-398+STR
23. Haltiner GJ, Williams RT (1980) Numerical prediction and dynamic meteorology, 2nd edn.
Wiley, New York
24. Kalnay E et al (1996) The NCEP/NCAR 40-year reanalysis project. Bull Am Meteorol Soc
25. Becker JJ, Sandwell DT, Smith WHF, Braud J, Binder B, Depner J, Fabre D, Factor J,
Ingalls S, Kim S-H, Ladner R, Marks K, Nelson S, Pharaoh A, Trimmer R,Von Rosenberg J,
Wallace G, Weatherall P(2009) Global bathymetry and elevation data at 30 arc seconds
resolution: SRTM30_PLUS, Mar Geodesy 32(4):355371. doi:10.1080/01490410903297766
26. The Hywind project. Available at: Last accessed 18 Aug 2015
27. The Hywind Scotland pilot park. Available at: Last accessed 18 Aug
28. Gambo D (2013) Planning and hierarchy usage spaceoffshore wind and waves energy.
Contribution to the FCT projectmarine energy roadmap. In: Portuguese. Project discipline
work in Renew Energy developed at the department of geographic energy, geophysics and
energy, Faculty of Sciences, University of Lisbon
Operation and Maintenance of Floating
Offshore Wind Turbines

Fernando P. Santos, ngelo P. Teixeira and Carlos Guedes Soares

Abstract This chapter starts by shortly addressing the statistics of accidents and
component failures of wind turbine structures based on a comprehensive dataset
publicly available. The distribution of the types of offshore wind turbine structures
installed in European waters is given. The operation and maintenance of xed
structures foundations is discussed. Then, the failure data of main subassemblies of
wind turbines are presented and discussed, followed by a description of available
and important condition monitoring systems, techniques and methods for operation
and maintenance of wind turbines. Finally, the knowledge on modelling, simulation
and optimization of operation and maintenance actions of xed offshore wind
turbines is discussed as a basis for the application in the operation and maintenance
of floating offshore wind turbines.

Keywords Operation  Maintenance  Floating offshore wind turbine

1 Introduction

The operation and maintenance (O&M) of offshore wind farms is more challenging
and costly than in onshore installations given that offshore conditions limit the
accessibility to the turbines. Even in face of favourable weather conditions offshore
O&M costs are higher due, for example, to the offshore distances, to being exposed
to the corrosive marine environment, and to the need for large vessels with special
lifting equipment which may not be always available at short notice to replace
major turbine components. The cost of energy in some European offshore wind
farms is approximately 33 % more than in onshore installations [1] and the O&M
costs can reach up to 30 % of such cost [2]. Most of European offshore wind

F.P. Santos  .P. Teixeira  C. Guedes Soares (&)

Centre for Marine Technology and Ocean Engineering (CENTEC),
Instituto Superior Tcnico, Universidade de Lisboa, Av. Rovisco Pais,
1049-001 Lisbon, Portugal

Springer International Publishing Switzerland 2016 181

L. Castro-Santos and V. Diaz-Casas (eds.), Floating Offshore Wind Farms,
Green Energy and Technology, DOI 10.1007/978-3-319-27972-5_10
182 F.P. Santos et al.

turbines are installed in the near coasts in xed substructures. However, new wind
turbine concepts using floating platform technologies have been proposed and are
being testedthere were only two experimental and two full-scale floating wind
turbines in European waters at the end of 2013 [3]but the O&M costs are dif-
ferent from the xed-bottom offshore wind turbines. Thus, challenges may lie ahead
and planning and optimizing the offshore wind O&M activities is of paramount
importance to reduce the O&M costs.
The lack of experience with floating wind turbines means that it is necessary to
resort to the experience gained with xed offshore wind turbines.
This chapter begins with a short introduction on the O&M of xed offshore wind
turbine structures. The failure data of xed offshore wind turbines is addressed
along with the condition monitoring systems and techniques available. Finally, it
briefly discusses the modelling of O&M activities of floating wind turbines.

2 Operation and Maintenance

2.1 Fixed Offshore Wind Turbine Structures

A recent study [4] has been performed on statistics of accidents and component
failures based on the worldwide wind turbine accident and incident data collected
by the Caithness Windfarm Information Forum [5] from 1980 to 30th September
2014. It is worth noting that there are almost no reports on accidents on offshore
wind turbines. Although the dataset does not represent the real number of acci-
dents occurred, it is believed to be the most comprehensive publicly available.
The study has revealed that within the structural failure category, tower failure is
by far the most frequent, representing 67 % of the events followed far behind by the
structural failure of blades, nacelle-housing and rotor. The main causes for tower
failure are storm (28 %), strong wind (21 %), typhoon (10.5 %) and braking system
failure (7 %), among others. The major consequence of tower failure is its collapse
(88 %). Almost all failures occurred onshore with the exception of the following
offshore wind turbines: that were certied despite having a construction error on
their foundations; whose foundations were in need of repair; with grout braking up,
potentially affecting 600 turbines at 13 wind farms; with failed grout.
At the end of 2013 there were 2474 offshore wind turbine structures fully
installed in European waters. The most common structures are the monopiles (1866
units), followed by the gravity-based ones (303 units), jackets (130 units), tripods
(115 units) and tripiles (55 units) [3].
Unlike with onshore wind turbines, the major sources of uncertainty for the
support structure of xed offshore wind turbines are the aerodynamic loading,
hydrodynamic loading, and soil properties given that they are subjected to random
wind and wave loads and are located in sites with variable soil conditions,
respectively [6].
Operation and Maintenance of Floating Offshore Wind Turbines 183

For example, two structural damage hot spots in offshore wind turbine structures
are the welded/bolted joins of tripods and the splash zone of the tower. This last is
prone to damage given the exposure to the corrosive marine environment [7].
Possible failures of the tower are crack formation, fatigue, vibration and foundation
weakness. Possible condition monitoring techniques (CMTs) for structural health
monitoring (SHM) of towers are vibration analysis (VA), shock pulse method
(SPM), strain measurement (SM), ultrasonic testing techniques (UTT) and visual
inspection (VI) [8].
Generally, and unlike turbines, foundations are not covered by any type of
warranty, thus foundation risks are somewhat insurable and mitigated based on
certication. Consequently, the maintenance is of a different kind of nature, mainly
based on visual inspections and survey work with corrective or remedial work when
necessary. Inspections are performed on structural strength, corrosion and scour
protection, where needed [9].
To guarantee the structural integrity of the turbine foundations it is required a
variety of specialist surveys. Routine surveys are expected to be performed in the
rst two years. From the moment the site has been characterised the surveys to
follow are quite rare, with several works required on a ve or ten year cycle. The
scour protection, which is the protection installed to prevent sediment erosion
where the turbine foundation meets the seabed can be performed by side-scan sonar
from a survey vessel. Surface inspections and surveys comprise monopile internal
inspections of the grouted connections and splash zone inspections [9].
The maintenance of the turbine foundation structure and transition piece can
involve several activities. Repairing the paintwork, particularly on the boat landing,
and cleaning marine growth is standard maintenance. Major works can include
repairs to grouted joints, rock placement to improve scour protection and irregular
repairs to wave-damaged secondary steelworks like ladders, gates, grills and plat-
forms [9].
In the case of the floating structures, much of the maintenance knowledge comes
from the long experience of the offshore oil and gas industry.

2.2 Fixed Offshore Wind Turbines

There are few surveys on wind turbine failure data available in the public domain.
The LWK and WMEP surveys with approximately 5800 and 15400 turbine years,
respectively, are represented graphically in Figs. 1 and 2 [1]. It is shown that
electric and electronic subassemblies fail more frequently than mechanical ones.
However, these last experience longer downtimes (e.g. drive train, gearbox, gen-
erator, blades). Also, the gearbox subassembly has the highest downtime and for
this reason the wind turbine industry efforts focus on gearbox [10].
A study by Faulstich et al. [11] categorised the WMEP subassembly onshore
failure data into minor and major failures as shown in Figs. 3 and 4 for the annual
failure rate and downtime per failure, respectively. The annual failure rates of the
184 F.P. Santos et al.

Fig. 1 Subassembly failure

rate from LWK and WMEP
surveys, 19932006 [1]

Fig. 2 Subassembly
downtime per failure from
LWK and WMEP surveys,
19932006 [1]

Fig. 3 WMEP subassemblies

annual failure rates divided
into minor and major failures
Operation and Maintenance of Floating Offshore Wind Turbines 185

Fig. 4 WMEP subassemblies

downtime per failure divided
into minor and major failures

electronic control and electronic system are considerably the highests. Combining
the annual failure rates and downtimes per failure, the electronic and electrical
minor failures are twice the blades downtime and eight times the generator
downtime, respectively. Therefore, is also important to focus on these minor fail-
ures in order to improve wind turbine availability.
The study also shows that minor failures (75 % of failures) are responsible for
only 5 % of the downtime, whereas major failures (25 % of failures) cause 95 % of
the downtime. Consequently, maintenance should focus on the 25 % of failures that
cause almost the total downtime. Moreover, the rotor, gearbox, generator and yaw
system failures lead to long downtimes and higher costs. Maintenance should also
focus on these subassemblies to improve availability. There is a need to improve the
reliability of electrical and electronic subassemblies given that they are responsible
for a considerable portion of the total downtime. The downtime due to minor
failures is expected to increase under offshore conditions like limited accessibility
to the turbines.
Reliability-centred maintenance (RCM) is the state-of-the-art approach to
determine the maintenance strategy in the wind turbine industry. It comprises
preventive maintenance based on performance and/or parameter monitoring and
consequent actions. Condition monitoring is used in RCM to nd the optimum
point between corrective and scheduled maintenance strategies [8].
The reduction of O&M costs and the improvement of reliability have been major
priorities in wind turbine maintenance strategies. This can be achieved by using
reliable and cost-effective CMTs. The implementation of adequate condition
monitoring systems (CMSs) and fault detection systems can considerably improve
the reliability and availability of systems. Sensors and measurements systems are
used to collect data at regular time intervals. The condition of important wind
turbine components can be known by data processing and analysis. Moreover, by
processing the data history faults can be detected (diagnosis) or predicted (prog-
nosis) and the adequate maintenance strategy can be selected. A CMS has CMTs
applied on wind turbine subsystems and/or CMTs applied on the overall (global)
wind turbine system. There are several CMTs applied on wind turbine subsystems
such as the following [8]:
186 F.P. Santos et al.

Vibration analysis (VA): is considered as the most efcient technique for an

early prediction and detection of failures in mechanical components. The VA is
applied to wind turbine components, such as shafts, bearings, gearboxes and
blades. Fast Fourier transformation is generally the signal processing technique
used in VA for transforming a time-domain signal to frequency-domain signal.
VA methods have a high level of interpretation facilitating the location of the
faulty component. However, it needs additional hardware and software,
increasing the production costs. Moreover, the use of sensors to detect
low-frequency faults is difcult;
Oil analysis (OA): it has been shown that the monitoring of oil debris is a viable
CMT for the early detection and tracking of the damage in wind turbines gear-
boxes components, such as bearings and other gear elements. The quantity and
type of metal debris caught by a lter from cracked gearbox wheels or bearings
can indicate the health of the component. The OA has the objectives of monitoring
the lubricant condition, guarantee the quality of the oil and safeguard the com-
ponents involved. The OA process typically involves the following main tests:
Viscosity analysis;
Oxidation analysis;
Water content or acid content analysis;
Particle count analysis;
Machine wear analysis;
The OA techniques can be performed through real-time continuous monitoring
and offline oil sample analysis. Although such processes are generally carried
out offline by taking samples, online real-time monitoring may be needed
in situations in which failure modes develop rapidly or when accessibility is
limited. Depending on the sensing techniques used, online detection technolo-
gies can be the following: electromagnetic sensing, flow or pressure drop, and
optical debris sensing. Internal gearbox failures can only be detected by OA.
However, this has limitations, i.e. failures outside the gearbox cannot be
detected and the equipment used for online monitoring is very expensive. Thus,
offline monitoring of oil samples is frequently used;
Temperature measurement (TM): is used to detect the existence of potential
failures through the changes in temperature in a component. It is a reliable CMT
since every equipment has limited operational temperature. TM is applied on
components, such as bearings, oil, generator windings, among others. The TM is
rarely used alone but generally as secondary source of information given that the
temperature develops slowly and is not sufcient for early and precise fault
detection. Moreover, the temperature that is being monitored can be influenced
by the surroundings;
Strain measurement (SM): is a technique used for SHM of blades and towers,
for example. The measurements are acquired with strain gauges and the data is
generally processed by the nite element method. However, in the long term
Operation and Maintenance of Floating Offshore Wind Turbines 187

these gauges are not robust. Nowadays, there are wind turbine manufacturers
that integrate bre-optic sensors in the blades so to decrease the connections
with the data logger and allow little to no weakening of the signal over a
signicant distance;
Optical bre monitoring (OFM): is a technique applicable in the SHM of wind
turbines and the optical bres must be placed on the surface or embedded into
the monitored components, such as blades, to monitor critical parameters. Thus,
when considering real-world applications OFM is expensive when comparing it
with other condition monitoring and fault detection methods. Nonetheless, the
cost is expected to decrease in the future due to the development of technology;
Visual inspections (VI): it can be used to monitor wind turbine components,
such as the rotor blades, nacelles, slip rings, yaw drives, bearings, generators
and transformers. Also, VI is applicable with other CMTs to identify problems
that such techniques could not identify. A limitation is that VI only identies
damages which are visible on the surface of a structure. Additionally, VI is
labour intensive and subjective given that the results depend on the experience
and judgment of the inspector. Nowadays, remote VI technologies are being
implemented by the wind turbine industry to inspect critical components, such
as gearboxes, blades, among others;
Acoustic emission (AE): is a costly technique usually applied for fault detection
in gearboxes, bearings, shafts and blades. The advantages of AE include a large
frequency range and a quite high signal-to-noise ratio. However, only some type
of faults occurs in the high-frequency range. An additional limitation is the
attenuation of the signal during propagation. Consequently, the AE sensor must
be as close as possible to the source, which may be a practical limitation in
applying AE to some wind turbines. It has been shown that AE is more sensitive
in detecting and monitoring faults than the vibration or spectrometric OA;
Ultrasonic testing technique (UTT): is a method based on elastic wave propa-
gation and reflection within the material. There are three different techniques for
this analysis: pulse-echo, through transmission, and pitch-catch. The wind
energy industry widely uses UTTs for the structural assessment of wind turbine
towers and blades. The estimation of the location and nature of a failure can be
performed with ultrasonic testing using wave propagation characteristics. The
laminate for dry glass bres and delamination can be checked below the surface
with ultrasound scanning;
Thermography analysis (TA): infrared thermography is a CMT used in the wind
turbine industry to control and diagnose of electric and mechanic equipment.
However, TA is not adequate for early fault detection given that temperature
develops slowly. Traditionally TA in wind turbine condition monitoring, has
been performed offline. Nonetheless, there are nowadays cameras and diagnostic
software available in the market which are adequate for online monitoring;
Radiographic inspection: is a technique that gives valuable information on the
structural condition of an inspected wind turbine component. However, radio-
graphic imaging using X-rays is seldom used in the wind energy industry.
188 F.P. Santos et al.

The technique is considerably efcient in detecting crack and delamination in

the blade, rotor and tower structures.
The global systems CMTs can be used online, increasing the wind turbine
reliability and decreasing the downtime and O&M costs. These techniques are the
following [8]:
Performance monitoring: wind turbine parameters readings (e.g. wind velocity,
rotor speed) are compared with values in operator manuals or manufacturer
performance specications to verify if the operation efciency of the system is
optimal. The relationships among different parameter readings can be used for
safety purposes;
Power signal analysis: power quality is an area of interest in wind turbine
condition monitoring given that quality can degrade due to wind speed turbu-
lence and switching events. Considering the mechanical power signal of the
wind turbine drive shaft and the total three-phase electrical power signal, both
energy flows are disturbed by malfunctions from mechanical or electrical faults.
Major variations in the wind turbine drive train torque are usually indications of
malfunctions. Torsional oscillations are caused by faults in the drive train. For
example, torque oscillations can be identied in a blade or rotor imbalance
Signature analysis (SA): it can be used to predict or detect electrical or
mechanical faults in wind turbines. Signals like voltages, power, currents,
among others can be detected by SA as also several faults, such as broken rotor
bars, bearing failures, air gap eccentricity and unbalanced rotors and blades;
SCADA data analysis: when used for wind turbine condition monitoring is
considered cost effective, since the sensors networks are already in place, and
reliable given that is based on the interpretation of SCADA data. The opera-
tional data collected from sensors can reflect the condition of wind turbine
systems and the analysis of SCADA data the relationship among different sig-
nals so to determine the health of the wind turbine. Given that the SCADA
system was not initially designed for condition monitoring purposes, the wind
turbine SCADA system does not collect all the information needed for complete
condition monitoring. Moreover, a concern is that the values of the SCADA data
vary over broad ranges under varying operation conditions, and without an
adequate data analysis tool it is hard to detect an incipient fault from raw
SCADA data. According to Mrquez et al. [12] SCADA uses several signal
processing methods such as the following: statistical methods; trend analysis;
ltering methods; time-domain analysis; Cepstrum analysis; time synchronous
averaging; Fast Fourier transform; amplitude demodulation; order analysis;
wavelet transforms; hidden Markov models.
Operation and Maintenance of Floating Offshore Wind Turbines 189

The empirical mode decomposition (EMD) method and the ensemble EMD
(EEMD) method are important in diagnosing wind turbine blades crack and plan-
etary gearbox faults. However, the EEMD method needs a considerable number of
trials to effectively diagnose gearbox faults. The voltage- and current-based
methods in generators have the disadvantage of the efciency of articial recog-
nition on these data being quite low compared to vibration data. Thus, neural
networks and expert systems are introduced for quick and accurate diagnosing of
faults. The dynamic model analysis has been the main method to diagnose the faults
of the wind turbine components. The SCADA data is dependent on the health
condition of the turbine and on its operational conditions. The SCADA data value
changes over a wide range, increasing the difculty in condition monitoring. For
example, incipient faults are difcult to detect [13].
Gradually, the wind energy industry has been installing larger wind turbines in
more remote areas onshore and increasingly located offshore for optimal wind
conditions. The offshore wind farms are being moved farther from shore and into
deeper waters [3]. The location and size of the offshore wind turbines has been
responsible for maintenance challenges that are unique when compared with those
of traditional power generation systems. To deal with it, wind turbine CMS man-
ufacturers need to improve the existing techniques and/or develop more adequate
techniques. The future objective in CMS is the use of intelligent software algo-
rithms and automated analysis, i.e. the wind industry is moving toward intelligent
machine health management. The main goal is to have wind energy conversion
systems able to understand and make decisions without human intervention. This
requires using intelligent condition-based maintenance systems based on RCM
mechanisms. Thus, new tendencies in the wind turbine condition monitoring
industry include the following: smart monitoring, remote and e-monitoring,
in-service SHM, integration and interaction of monitoring and control systems, and
estimation of the remaining component life service [8].

2.3 Floating Offshore Wind Turbines

The concept of floating wind turbine has been introduced to harness the higher and
steadier wind speeds available away from coast, having the additional advantage of
reducing the visual impact. At these sites, the xed offshore wind turbines are not
economically viable given that the loading increases as depth increases requiring
larger structures [14].
Floating structures that are stable with low draft can be towed into port facilities
by tug boats to perform long-term maintenance, avoiding using expensive vessels
like the jack-up barges. The ease in which this is done will lower the maintenance
cost during critical overhaul cycles. Structures which are more difcult to float back
to shore, such as the TLP or a spar-buoy, can be more costly in terms of mainte-
nance [15].
190 F.P. Santos et al.

At the end of 2013, there were available only two experimental and two full
scale floating structures [3] and it is expected during the course of the next decade,
that deep offshore wind farms will consist of floating platform technologies [16].
The lack of experience on floating wind turbines dictates that in order to study them
one has to adapt the O&M models of xed offshore wind turbines.
The offshore wind turbines are subjected to higher environmental and power
utilization stresses than onshore turbines resulting from the marine environment.
Therefore, they have higher failure rates [17], which is the reason why onshore
failure models are not appropriate for modelling the O&M and performance of
offshore turbines. Given that public domain eld data of offshore wind turbines are
scarce [1, 17] power rating and environmental stress factors for mechanical systems
[18] have been used as an empirical approach to obtain the offshore failure models
of the turbines components from characteristic onshore failure distributions, as
done by [19].
However, the logistics of floating offshore wind turbines are typically more
complex and recent studies have developed simulation models to study their effect.
Shaee [20] did a comprehensive review of the state-of-the-art on maintenance
logistics, covering issues and challenges related to operation and maintenance
logistics of offshore wind farms classied according to a strategic-tactical-
operational framework. Strategic issues include among others wind farm design
for reliability and selection of wind farm maintenance strategy. Tactical topics
comprise spare parts inventory management and operational decisions include
scheduling of maintenance tasks and routing of vessels.
Other studies have addressed the maintenance optimization of offshore wind
turbines and farms [21, 22]. A quite relevant issue is the role of grouping in the
development of an overall maintenance optimization framework for offshore wind
turbines. This comes from the fact that wind turbines on the wind farm will behave
similarly as when they are located in close proximity, which can be exploited at the
wind farm level as done by Hameed and Vatn [23].
Markov approaches have been used in the optimization of wind farms [24, 25].
Recently, a Markov decision process was proposed to determine the optimal
operation and maintenance policy of an offshore wind farm considering the
stochastic wind and weather conditions [25].
In the last few years, Generalized Stochastic Petri nets (GSPN) with predicates
combined with Monte Carlo simulation (MCS) have been used to model and
simulate O&M strategies of xed offshore wind turbines [19, 26, 27].
A Petri net comprises three basic elements: places and transitions used to model
conditions and events of a system, respectively; directed arcs connected from places
to transitions and vice versa establishing relations between network places and
transitions [28]. Places represent the systems states (e.g. functioning) and are also
deposits of resources (e.g. maintenance team) called tokens or marks, whereas
transitions represent the events (e.g. failure) that manipulate those resources. The
transitions ring is responsible for the evolution of the tokens in the Petri net, i.e.
for the change in the systems states. This way the dynamic behaviour of a system is
modelled. According to Zio [29] Petri nets can properly represent the dynamic
Operation and Maintenance of Floating Offshore Wind Turbines 191

interactions and dependencies between different systems components which

influence the systems behaviour and its maintenance. Stochastic performance
values can be an output of such nets when combined with MCS.
The GSPN with predicates are a recent extension of the original Petri nets. Their
transitions can re deterministically, stochastically and be conditioned by predi-
cates, i.e. by guards and assignments. The guards are pre-conditions that enable or
inhibit the ring of transitions, whereas assignments are post-conditions messages
which update variables used in the model (e.g. in transitions).
GSPN with predicates coupled with MCS were used by Santos et al. [26] to
model the planning of O&M activities of an offshore wind turbine considering
degraded components. Corrective maintenance based on replacements and age
imperfect preventive maintenance (PM) were modelled and compared in terms of
the wind turbines performance (e.g. availability and loss production) and of the
O&M costs. Santos et al. [27] have compared a PM repair cost function dependent
on an age reduction effort and on the age of the components is compared with an
age independent cost function, in terms of their effect on the O&M costs and
revenues. Failure models are based on onshore ones and derived through an
empirical approach based on stress factors for mechanical systems. O&M activities
consider logistic resources, times and costs, and weather constraints. Santos et al.
[19] performed a parametric study on how the variation of failure and repair models,
vessels logistic times, weather windows and waiting times affect the wind turbine

3 Conclusion

Currently, the floating wind turbine industry is in its infancy; thus, no much
experience has been accumulated so far. In terms of the maintenance of the floating
structure knowledge is available from the long experience of the offshore oil and
gas industry. In which concerns the floating wind turbine, the O&M modelling may
be adapted from the xed offshore wind turbines by using for example, GSPN with
predicates and MCS. In the future, as the development of condition monitoring
systems is moving towards intelligent machine health management, i.e. to intelli-
gent software algorithms and automated analysis, taking decisions without human
intervention may be possible.

Acknowledgements This work was performed within the Strategic Research Plan of the Centre
for Marine Technology and Ocean Engineering, which is nanced by the Portuguese Foundation
for Science and Technology (Fundao para a Cincia e Tecnologia-FCT).
192 F.P. Santos et al.


1. Tavner P (2012) Offshore wind turbines: reliability, availability and maintenance. Renewable
Energy Series 13. The Institution of Engineering and Technology, London
2. Nielsen JJ, Srensen JD (2011) On risk-based operation and maintenance of offshore wind
turbine components. Reliab Eng Syst Saf 96:218229
3. The European Wind Energy Association (2014) The European offshore wind industrykey
trends and statistics 2013.
European_offshore_statistics_2013.pdf . Accessed 17 June 2014
4. Santos FP, Teixeira AP, Guedes Soares C (2015) Review of wind turbine accident and failure
data. In Guedes Soares C (ed) Renewable energies offshore. Taylor & Francis Group, London.
pp 953959
5. Caithness Windfarm Information Forum (2014) Wind turbine accident compilation. http:// Accessed 30 Sept 2014
6. Carswell W, Arwade SR, DeGroot DJ, Lackner MA (2015) Soilstructure reliability of
offshore wind turbine monopile foundations. Wind Energy 18(3):483498. doi:10.1002/we.
7. Ciang CC, Lee J-R, Bang H-J (2008) Structural health monitoring for a wind turbine system: a
review of damage detection methods. Meas Sci Technol 19(12):120. doi:10.1088/0957-0233/
8. Tchakoua P, Wamkeue R, Ouhrouche M, Slaoui-Hasnaoui F, Tameghe TA, Ekemb G (2014)
Wind turbine condition monitoring: state-of-the-art review, new trends, and future challenges.
Energies 7:25952630. doi:10.3390/en7042595
9. Garrad Hassan GL (2013) A guide to UK offshore wind operations and maintenance. Scottish
enterprise and the crown estate. Accessed 11 Sept 2015
10. Spinato F, Tavner PJ, van Bussel GJW, Koutoulakos E (2009) Reliability of wind turbine
subassemblies. Renew Power Gener IET 3(4):387401. doi:10.1049/iet-rpg.2008.0060
11. Faulstich S, Hahn B, Tavner PJ (2011) Wind turbine downtime and its importance for offshore
deployment. Wind Energy 14(3):327337. doi: 10.1002/we.421
12. Mrquez FPG, Tobias AM, Prez JMP, Papaelias M (2012) Condition monitoring of wind
turbines: techniques and methods. Renew Energy 46:169178. doi:10.1016/j.renene.2012.03.
13. Liu WY, Tang BP, Han JG, Lu XN, Hu NN, Hu ZZ (2015) The structure healthy condition
monitoring and fault diagnosis methods in wind turbines: a review. Renew Sustain Energy
Rev 44:466472
14. Kaldellis JK, Kapsali M (2013) Shifting towards offshore wind energy-Recent activity and
future development. Energy Policy 53:136148
15. Buttereld S, Musial W, Jonkman J, Sclavounos P (2007) Engineering challenges for floating
offshore wind turbines. Conference paper NREL/CP-50038776.
fy07osti/38776.pdf. Accessed 17 Sept 2015
16. The European Wind Energy Association (2011) Wind in our sailsthe coming of Europes
offshore wind energy industry.
reports/Offshore_Report.pdf. Accessed 17 June 2014
17. Karyotakis A, Bucknall R (2010) Planned intervention as a maintenance and repair strategy for
offshore wind turbines. J Mar Eng Technol 9:2735
18. Davidson J (1994) The reliability of mechanical systems (Institute of Mechanical Engineering
(IMechE) guides for the process industries). Wiley, New York
19. Santos FP, Teixeira AP, Guedes Soares C (2015) Modelling and simulation of the operation
and maintenance of offshore wind turbines. Proc Inst Mech Eng, Part O: J Risk Reliab 229
(5):385393. doi:10.1177/1748006X15589209
20. Shaee M (2015) Maintenance logistics organization for offshore wind energy: current
progress and future perspectives. Renew Energy 77:182193
Operation and Maintenance of Floating Offshore Wind Turbines 193

21. Besnard F, Patriksson M, Stromberg A-B, Wojciechowski A, Bertling L (2009) An

optimization framework for opportunistic maintenance of offshore wind power system. In:
Proceedings of IEEE Bucharest PowerTech conference, Bucharest, 28 June2 July 2009.
IEEE, New York. doi:10.1109/PTC.2009.5281868
22. Besnard F, Patriksson M, Stromberg A-B, Wojciechowski A, Fischer K, Bertling L (2011) A
stochastic model for opportunistic maintenance planning of offshore wind farms. In:
Proceedings of IEEE Trondheim PowerTech conference, Trondheim, 1923 June 2011. IEEE,
New York. doi:10.1109/PTC.2011.6019376
23. Hameed Z, Vatn J (2012) Role of grouping in the development of an overall maintenance
optimization framework for offshore wind turbines. Proc Inst Mech Eng, Part O: J Risk Reliab
226(6):584601. doi: 10.1177/1748006X12464616
24. Byon E, Ntaimo L, Ding Y (2010) Optimal maintenance strategies for wind turbine systems
under stochastic weather conditions. IEEE Trans Reliab 59(2):393404
25. Castanier B, Pehlivan C, Yeung TG (2015) Optimization of maintenance and operational
policies of an offshore wind farm subject to stochastic wind conditions. In: Nowakowski T,
Mlynczak M, Jodejko-Pietruczuk A, Werbiska-Wojciechowska S (eds) Safety and reliability:
methodology and applications. Taylor & Francis Group, London, pp 11411146
26. Santos FP, Teixeira AP, Guedes Soares C (2013) Maintenance planning of an offshore wind
turbine using stochastic Petri nets with predicates. In: Proceedings of the 32nd international
conference on ocean, offshore and arctic engineering, Nantes, France. doi:10.1115/
27. Santos FP, Teixeira AP, Guedes Soares C (2015) An age-based preventive maintenance for
offshore wind turbines. In: Nowakowski T, Mlynczak M, Jodejko-Pietruczuk A,
Werbinska-Wojciechowska S (eds) Safety and reliability: methodology and applications.
Taylor & Francis Group, Oxford, pp 11471155. doi: 10.1201/b17399-161
28. Murata T (1989) Petri nets: properties, analysis and applications. Proc IEEE 77(4):541580.
29. Zio E (2009) Reliability engineering: old problems and new challenges. Reliab Eng Syst Saf
94:125141. doi:10.1016/j.ress.2008.06.002