HomeMy WebLinkAbout20170925Utah_OCS 1.10 NREL - Repowering #60535.pdfDNREL
Wind Power Project
Repowering:Financial
Feasibility,Decision Drivers,
and Supply Chain Effects
Eric Lantz,Michael Leventhal,
and Ian Baring-Gould
NREL is a national laboratory of the U.S.Department of Energy
Office of Energy Efficiency &Renewable Energy
Operated by the Alliance for Sustainable Energy,LLC
This report is available at no cost from the National Renewable Energy
Laboratory (NREL)at www.nrel.gov/publications.
Technical Report
NRELITP-6A20-60535
December 2013
Contract No.DE-AC36-08GO28308
EDINREL
NATIONAL RENEWABLE ENERGY LABORATORY
Wind Power Project
Repowering:Financial
Feasibility,Decision Drivers,
and Supply Chain Effects
Eric Lantz,Michael Leventhal,
and Ian Baring-Gould
Prepared under Task No.WE11.0630
NREL is a national laboratory of the U.S.Department of Energy
Office of Energy Efficiency &Renewable Energy
Operated by the Alliance for Sustainable Energy,LLC
This report is available at no cost from the National Renewable Energy
Laboratory(NREL)at www.nrel.gov/publications.
National RenewableEnergy Laboratory Technical Report
15013 DenverWest Parkway NREL/TP-6A20-60535Golden,CO 80401 December 2013303-275-3000www.nrel.gov
Contract No.DE-AC36-08GO28308
NOTICE
This report was prepared as an account of work sponsored by an agency of the United States government.
Neither the United States government nor any agency thereof,nor any of their employees,makes any warranty,
express or implied,or assumes any legal liability or responsibility for the accuracy,completeness,or usefulness of
any information,apparatus,product,or process disclosed,or represents that its use would not infringe privately
owned rights.Reference herein to any specific commercial product,process,or service by trade name,
trademark,manufacturer,or otherwise does not necessarily constitute or imply its endorsement,recommendation,
or favoring by the United States government or any agency thereof.The views and opinions of authors
expressed herein do not necessarily state or reflect those of the United States government or any agency thereof.
This report is available at no cost from the NationalRenewable Energy
Laboratory(NREL)at www.nrel.gov/publications.
Available electronically at http://www.osti.gov/bridqe
Available for a processing fee to U.S.Department of Energy
and its contractors,in paper,from:
U.S.Department of Energy
Office of Scientific and Technical Information
P.O.Box 62
Oak Ridge,TN 37831-0062
phone:865.576.8401
fax:865.576.5728
email:mailto:reports@adonis.osti.gov
Available for sale to the public,in paper,from:
U.S.Department of Commerce
National Technical Information Service
5285 Port Royal Road
Springfield,VA 22161
phone:800.553.6847
fax:703.605.6900
email:orders@ntis.fedworld.gov
online ordering:http://www.ntis.gov/helplordermethods.aspx
Cover Photos:(left to right)photo by Pat Corkery,NREL 16416,photo from SunEdison,NREL 17423,photo by Pat Corkery,NREL
16560,photo by Dennis Schroeder,NREL 17613,photo by Dean Arrnstrong,NREL 17436,photo by Pat Corkery,NREL 17721.
Printed on paper containing at least 50%wastepaper,including 10%post consumer waste.
Acknow ed ments
This report was funded by the U.S.Department of Energy's (DOE)Wind and Water Power
Technologies Office in accordance with the National Renewable Energy Laboratory's (NREL)
Annual Operating Plan,Subtask WE 6.1.3 Wind Turbine Repowering and Recycling
Assessments,Project 21115,Agreement 24944.The authors thank Cash Fitzpatrick and Jose
Zayas of DOE's Wind and Water Power Technologies Office for supporting this research.
The authors would also like to thank Charles Newcomb (Endurance Wind Power),Mark
Jacobson (NREL),Neil Habig (Iberdrola Renewables),and Randy Mann (Edison Mission
Energy)for their reviews and comments on early versions of this manuscript.Any remaining
errors or omissions are the sole responsibility of the authors.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)iii
at www.nrel.gov/publications.
L st o Acrony s and Abbrev at ons
BWE BundesverbandWindEnergie (German Wind
Energy Association)
DOE U.S.Department of Energy
GW gigawatt
IEC International Electrotechnical Commission
kW kilowatt
m meter
MW megawatt
MWh megawatt-hour
NPV net present value
NREL National Renewable Energy Laboratory
O&M operation and maintenance
PPA power purchase agreement
PTC production tax credit
SAM System Advisor Model
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)iv
at www.nrel.gov/publications.
xecut ve Summary
As wind power facilities age,project owners are faced with plant end-of-life decisions.This
report is intended to inform policymakers and the business communityregarding the history,
opportunities,and challenges associated with plant end of life actions,in particular,repowering.
Specifically,the report details the history of repowering,examines the plant age at which
repowering becomes financially attractive,and estimates the incremental market investment and
supply chain demand that might result from future U.S.repowering activities.
Repowering as defined here includes two types of actions.Full repowering refers to the complete
dismantling and replacement of turbine equipment at an existing project site.Partial repowering
is defined as installinga new drivetrain and rotor on an existing tower and foundation.Partial
repowering allows existing wind power projects to be updated with equipment that increases
energy production,reduces machine loads,increases grid service capabilities,and improves
project reliabilityat lower cost and with reduced permitting barriers relative to full repowering
and greenfield projects.
Repowering first emerged in the early 1990s in the California and Danish wind power markets
and was followed by the Dutch and German markets in the 1990s and 2000s.Although
repowering activity has occurred elsewhere,these locales remain the principal markets for
repowering investments.Historically,repowering has been viewed as a means of increasing
project productivitywhile offering an array of other potential attributes of interest.
Fundamentally,however,profitability for a given project is the primary driver of repowering
decisions.Given limited financing,the anticipated profitabilityat alternate greenfield sites is
also relevant.
Two distinct analyses were conducted to understand the plant age when repowering becomes
viable.These analyses utilized NREL's System Advisor Model (SAM),a tool that enables the
user to predict estimated cash flows from a varietyof electric power generation technologies.Net
present value calculations were utilized to enable comparisons across time.
The first analysis involved creating "proto-typical"wind plants of four different vintages,with
commissioning years of 1999,2003,2008,and 2012.Proto-typical plants are representativeof
the industryat the time of their construction and rely on market data from each of the
commissioning years to derive installation and equipment costs,power purchase agreement
revenue,net capacity factor,receipt of federal production tax credit payments,and operation and
maintenance expenses.For each of these four plants,future investment decisions to repower,or
build a nearby greenfield site,were evaluatedfor 2015,2020,2025,and 2030.
The second analysis focused on three actual wind plants operating in the United States.These
plants were chosen for varyingvintages and geographical diversityand include a Northeast wind
plant (15-20 years old),a Midwest wind plant (10-15 years old),and a West Coast wind plant
(20-25 years old).For these three plants,the decision to repower the current site or invest in a
nearby greenfield site was assumed to occur in the 2012-2013 timeframe.Estimated costs to
repower,expected revenue,and operational statistics were acquired from nearby sites that were
recently placed in service or are in active development.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)y
at www.nrel.gov/publications.
Both financial analyses concluded that repowering tends to become financiallyattractive,relative
to investing in a nearby greenfield site,after approximately 20-25 years of service.Plants less
than 20 years old are expected to be capable of generatinga favorable revenue stream for several
more years.However,there are a number of critical variables that could alter both the timing and
attractiveness of repowering.These variables include the rate of technological advancement,
availability of quality wind resource sites,wholesale market prices for electricity,durability and
reliabilityof turbine equipment over time,and the extent to which owners are able to reuse
existing infrastructure in the future.
A single partial repowering scenario-replacementof only the turbine drivetrain and rotor-was
also examined.As defined here,partial repowering was estimated to reduce the cost of
repowering by about 10%,while achieving about 50%of the energy production improvements
associated with full repowering.Under these conditions,partial repowering was determined to be
less economically attractive than full repowering.It should be noted however,that partial
repowering can take many forms.If other concepts for partial repowering can be developed and
offer reduced operation and maintenance costs or increased energy production at substantially
lower costs than full repowering,such opportunities may well be financiallysound.
Assuming a plant life of 20-25 years,demand for repowering is expected to be low over the next
decade,reach a few hundredmegawatts per year in the early 2020s,and achieve 1-3 gigawatts
per year by the late 2020s.Total estimated value of the repowering market segment is estimated
at $25 billion through 2030 (constant 2012 U.S.dollars)with the vast majority of this investment
occurring in the latter half of the 2020s.Policy support in the form of financial incentives,as
well as solutions to potential regulatory and contractual hurdles,would likelybe necessary if an
acceleration of repowering investment were desired.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)vi
at www.nrel.gov/publications.
Tab e of Contents
1 Introduction...........................................................................................................................................1
1.1 History and Status ......................................................1
1.2 Critical Variables Influencing Trends ..................................................3
2 General Economic and Financial Feasibility .....................................................................................5
2.1 Methods..................................................5
2.2 Results ..................................................9
3 CaseStudyAnaIys is ..........................................................................................................................1 5
3.1 Methods.................................................15
3.2 Results .................................................17
4 Market Opportunity ............................................................................................................................19
4.1 Methods.................................................19
4.2 Results .................................................19
5 Supp IyCha i n Impa cts ........................................................................................................................21
5.1 Methods..................................................2 1
5.2 Results............................................................................21
6 Policy and Other Relevant Considerations......................................................................................24
7 Summa ry a ndConc I us i ons ...............................................................................................................25
References .................................................................................................................................................27
Appendix A:Interview Guide for Wind Owner Operators.....................................................................29
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)vii
at www.nrel.gov/publications.
L st of F ures
Figure 1.Value addedto a 1999 wind plant as a result of investing in a new greenfield or
full repowering ..................................................................................................................10
Figure 2.Value addedto a 2003 wind plant as a result of investing in a new greenfield or
full repowering ......................................................................................................................11
Figure 3.Value addedor lost as a result of investing in a new greenfield or full repowering for a 2008
wind power plant ...................................................................................................................12
Figure 4.Value addedor lost as a result of investing in a new greenfield or full repowering for a 2012
wind power plant ................................................................................................................13
Figure 5.Value addedto a 2003 wind plant as a result of investing in a new greenfield,full repowering,
or partial repowering in 2025 ...............................................................................................14
Figure 6.Value addedto each case study wind plant as a result of investing in a new greenfield or full
repowering............................................................................18
Figure 7.Estimated annual capacity repowered by year.............................................................................20
Figure 8.Estimated turbine demandby nameplatecapacity resulting from potential repowering
activity................................................................................................................................21
Figure 9.Estimated blade demandby length (meters)resulting from potential repowering activity.........22
Figure 10.Estimated tower demandby height (meters)resulting from potential repowering activity ......23
L st of Tab es
Table 1.Key Capacity-Weighted Average Turbine Parameters for Technology Considered for
Repowering in This Analysis .....................................................................................................5
Table 2.Summary of Financial ModelingInputs for Existing and Future Facilities by Year of ProjectCommissioninga
Table 3.Financial ModelingConstants Across Scenarios.....................................................................9
Table 4.Key Characteristics and Assumptions for the Three Case Study Wind Plants.............................16
Table 5.Financial ModelingConstants Across Case Study Scenarios...................................................17
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)viii
at www.nrel.gov/publications.
ntroduct on
As wind power facilities age,project owners are faced with plant end-of-life decisions.In many
cases,technological advancement suggests that project owners might be able to increase their
profits by refurbishing or replacing older equipment.However,end of life decisions are often far
from straightforward.Older facilities are typically fully depreciated and paid for,leaving any
remaining revenues after operation expenditures essentially as profit.In addition,the potential
for new contractual or regulatory terms when making substantial alterations at an existing site
can also present hurdles.This report is intended to inform policymakers and the business
communityregarding the history,opportunities,and challenges associated with plant end of life
actions,in particular repowering.
Repowering as defined in this report entails two types of actions.Full repowering refers to the
complete dismantling and replacement of turbine equipment at an existing project site,including
the tower and foundation.With full repowering,some of the existing project infrastructure (e.g.,
roads,buildings,and interconnection equipment)is assumed to be utilized in the new project.
There is also the potential to offset repowering costs by recycling or selling the older equipment.
Partial repowering is defined as installinga new drivetrain and rotor on an existing tower (with
an allowance for some structural tower modifications);some peripheral components,such as the
power convertors and electronics,may also be replaced.Partial repowering allows existing wind
power projects to be updated with equipment that increases energy production,reduces machine
loads,increases grid service capabilities,and improves project reliability.Performance
improvements associated with partial repowering are assumed to be lesser than under full
repowering but greater than the original design of the machine.
The remainder of Section 1 entails a brief history of global wind plant repowering activities and
trends,and summarizes the findings that have emerged from past research.In Section 2 and
Section 3 financial analyses are conducted that examine the economic drivers of repowering with
historical and projected cost and performance estimates and case studies.Estimates are carried
out to better understand the expected lifetime of U.S.wind power plants.In Section 4 and
Section 5,these results are used to derive estimates of the total repowering market and
equipment demand resulting from repowering activities.Section 6 discusses policy
considerations and other relevant decision drivers around repowering.The report concludes with
a summary of key results and takeaways.
1.1 H story and Status
Repowering first emerged in the early 1990s in the California and Danish wind power markets
and was followed by the Dutch and German markets in the 1990s and 2000s (White and Gipe
1993;Knight 2004;Hulshorst 2008;Munksgaard and Morthorst 2008).There has also been
interest in repowering in India and other parts of the world (e.g.,Kharul 2008;Goyal 2010;
GlobalData 2012;Filgueira et al.2009).Based on the data reviewed,Denmark and Germany
have generally been the most active repowering markets,followed by California.
Denmark was the first countryto activelypromote repowering (Wiser 2007;Sperling et al.
2010).Public policy support for repowering was first instituted there in 1994 (Sperling et al.
2010).In 2001,policy support was adjusted to provide an additional premium on top of the
standard feed-in tariff for repowered projects that previouslyused turbines smaller than 100
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)1
at www.nrel.gov/publications.
kilowatts (kW)(Sperling et al.2010;Wiser 2007).Additional incentives beyond the standard
feed-in tariff remain in place for repowered projects that previouslyused turbines smaller than
450 kW (Sperling et al.2010).Under the policy scheme in place from 2001 through 2003,it has
been estimated that 1,208 turbines were replaced,resulting in an increase in capacity of 202
megawatts (MW)(Sperling et al.2010).The stated goal under the policy scheme in place since
2004 (although modified in 2008)is for an additional 175 MW of existing capacity to be
replaced by 350 MW of new capacity (Sperling et al.2010).At least one source indicates that
Denmark was the largest repowering market in 2011 with an estimated 213 MW of repowered
wind generation(GlobalData 2012).
Suggested barriers to repowering in Denmark have been anecdotally observed to be greater
capital requirements of modern wind power projects and shiftingownership models from
community-based local ownership to larger corporate or utility players.Wiser (2007)also
suggests that,similar to California,industrystakeholders see relatively limited economic
incentive to replace a project that continues to operate and generate revenue.
Germany has been estimated to have the largest market potential for repowering,estimated at
nearly 6,000 MW by 2015 (BWE 2011).However,in spite of more favorable treatment than
greenfield projects due to the country'sfeed-in tariff laws,which incentivize repowering,
repowering activities to date have still been limited (Wiser 2007;BWE 2008;BWE 2011).
Through 2005,an estimated 59 MW of wind turbines were removed from service and replaced
by 169 MW of new capacity (Wiser 2007).Since then,repowering activities have continued but
remain modest.The German Wind Energy Association (BWE)reports (BWE 2008;BWE 2011)
that in 2007,an additional 108 turbines (41 MW)were replaced with 45 new turbines (203 MW).
In 2010,183 MW of new equipment were installed at previouslydeveloped sites;in 2011,170
turbines (123 MW)were replaced by 95 turbines (238 MW)(GlobalData 2012).The above
annual repowering statistics represent a relatively small share of the potential installations that
could be repowered in Germany (BWE 2008),and under the current energy policies,repowering
appears not to be profitable until an installation has been operating for at least 15 years
(Knight 2004).
Repowering in Germany has been limited by local hub and total turbine height restrictions as
well as requirements for setbacks from existing residential areas (Wiser 2007).In addition,the
BWE has noted that the existing incentives for wind power are insufficient to drive large-scale
shifts in the repowering market in Germany (BWE 2008).
Despite its status as a pioneer in the repowering space,California has struggled to develop a
robust repowering market due to a varietyof policy and regulatory challenges.White and Gipe
(1993)observed that repowering was likely affected by the contractual requirements negotiated
in the original sales agreements under the Public Utility Regulatory Policy Act standard offer
contracts of the 1980s.Stipulations around the federal production tax credit (PTC)also
apparently stifled repowering efforts in California (Wiser 2007;Wiser et al.2008).'Under all but
the most conservativeassumptions,prior analysis suggests that there is little economic incentive
i The literature occasionally refers to the "CaliforniaFix,"whereby projects are ineligiblefor the PTC if after
repowering they remain on their existing standard offer contract entered into prior to 1987.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)2
at www.nrel.gov/publications.
for functioningCalifornia projects to pursue repowering (Wiser et al.2008).2 Through 2007,it
was estimated that 365 MW of capacity had been repowered in California (Wiser et al.2008)
with more than 70%of that occurring before 1999 (Wiser 2007).Total repowered capacity
through 2007 was equivalent to about 20%of the installed wind power capacity in the state in
1994,further suggesting that significant technical potential for repowering in the California
market remains (Wiser et al.2008).
Other repowering activities have been observed in a limited number of cases in the Netherlands
and India (e.g.,Grontmij;RWE Innogy;Ramesh),and analysis of the potential economics of
repowering has been conducted in Spain (Filgueira et al.2009).However,repowering activities
in these countries have not been widespread.
Repowering efforts have overwhelminglyresulted in full turbine replacements as opposed to
partial repowering (White and Gipe 1993;Knight 2004;Fairley 2009;Wiser et al.2008;
Ramesh;RWE Innogy;Grontmij)."This is a result of the rate at which wind power technology
has improved.Even if existing equipment were refurbished,new turbines are often capable of
generatingmuch more energy than the existing equipment (Filgueira et al.2009;Knight 2004).
Moreover,the replacement of smaller turbines could allow some increase in the potential
generatingcapacity from a given land area;as a result,countries with relativelyfew high-value
wind resource sites are likely to prefer full repowering at these sites when wind power-or
climate-related policy provisions are under consideration.
The literature has indicated two solutions for equipment removed from an existing site.
California projects repowered in the early 1990s were largelyforced to scrap the existing
equipment due to dramatic improvements in technology (White and Gipe 1993;Knight 2004).
Repowering efforts in the early 2000s in Europe (for projects using equipment that was
approximately 10 years old)were more successful in their attempts to refurbish and sell the
dismantled turbine equipment into emerging markets in eastern Europe and other parts of the
world (Knight 2004).
1.2 Cr t ca Var ab es nf uencing Trends
Historically,repowering has been viewed as a possible means of increasing project productivity,
improving grid support and grid interactions,extending the use of existing infrastructure
investments,more efficiently utilizinghigh-value resource areas,and in some cases (e.g.,single
turbines or a single linear array),enabling greater installed capacity in a given land area (White
and Gipe 1993;Kharul 2008;Hulshorst 2008;Fairley 2009;Filgueira et al.2009;Wiser et al.
2008).Repowering has also been cited as a means of reducing the visual impact or clutter of
wind power projects because for a given capacity rating,it reduces the number of turbines in a
specific location (Knight 2004;Wiser et al.2008;Filgueira et al.2009;Meyerhoffet al.2010).
At the same time,it has been observed to potentiallyincrease visual and aesthetic impacts by
introducing new machines that are significantlylarger and taller than their predecessors
(Meyerhoffet al.2010;Sperling et al.2010).Repowering is generally anticipated to reduce
2 Presently,no data are available detailing functioningversus nonfunctioningwind power plants in California.
3 It should be noted,however,that partial repowering can in fact take many forms.It is feasible that more basic
turbine upgrades incorporated through power electronics or other means could occur with much greater frequency
than forms of partial repowering that involvemore extensive overhaul and replacement of turbine components.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)3
at www.nrel.gov/publications.
operation and maintenance (O&M)costs by deploying newer,more reliable equipment and by
reducing the number of moving parts when moving to larger turbines (White and Gipe 1993;
European Wind Energy Association 2009).4 Repowering has also been noted to offer the
possibility of reduced avian and wildlife impacts,at least for some species.However,broad-
based data on the empirical change in impacts before and after repowering are not available,
making it difficult to confirm this potential characteristic of repowered facilities (Wiser et al.
2008;Hotker 2006;Smallwood and Neher 2010).
Fundamentally,however,profitability for a given project is the primary driver of repowering
decisions.Given limited financing,the anticipated profitabilityat alternate greenfield sites is also
relevant.One's ability to extend,amend,or re-enter power purchase agreements (PPAs)with a
creditworthyoff-taker and to secure potential additional sources of revenue such as renewable
energy certificates is another factor.These themes are consistent throughoutmarkets for wind
power around the globe.
Detailed financial modeling conducted by Wiser et al.(2008)suggests that the profitability of
repowering is highly dependent on project-specific performance and annual operation
expenditures.Wiser et al.(2008)also found that older projects can be expected to continue to
operate profitably for many years,as long as the projects continue to function within a
reasonable range of performance and are able to avoid dramatic upticks in annual operation
expenditures.They conclude that in the absence of additional financial incentives,it is often in
the project owner's interest to continue to produce power at an existing facility rather than to
repower.Experience from Germany confirms this finding (Knight 2004).
4 The parts counts of modern turbines relativeto their predecessors are generally comparable;however,by reducing
the number of machines required to achieve a specific rated capacity,one can reduce plant-wide movingparts by
simply installing fewer turbines that are far greater in their rated capacity than the original equipment.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)4
at www.nrel.gov/publications.
2 Genera conom c and F nanc a Feas ty
An initial assessment of the economics of repowering was developed from aggregate U.S.wind
industrydata.This analysis extends and updates prior work by Wiser et al.(2008),which
focused on the California market.Althoughthis approach does not capture the complete array of
local and regional variabilities among projects,it provides a basic level of visibility-from the
project owner's perspective-regardingthe point at which repowering might be considered an
attractive mvestment opportumty.
2.1 Methods
Repowering opportunities are examined through 2030 for existing plants.The analyses consider
both full and partial repowering.The analyses presume capital is available to invest in wind
infrastructure.Repowering is analyzed as a potential investment opportunityrelative to
developing a new adjacent greenfield.If neither of these options results in added value to the
project owner,it is assumed the owner will simply maintain the existing plant and invest
available capital in other opportunities.
Adjacent greenfields were chosen as the point of reference for alternative wind power investment
opportunities based on anecdotal evidence and feedback from industryobtained through semi-
structured interviews.These data sources indicated that new greenfield opportunities in
proximate or adjacent locations to older facilities do exist.This suggests that the comparison to
an adjacent greenfield is a realistic tradeoff considered by the development community.This
approach also simplifies the analysis by eliminating the need to estimate wind resource
conditions for the "typical"greenfield project at future points in time.It should be noted,
however,that the ability to place a new greenfield in an adjacent site with essentially the same
wind resource as the existing facility increases the attractiveness of the greenfield relative
to repowermg.
To understand how repowering opportunities might vary as technology and the industryhave
evolved,repowering is examined independently for plants commissioned at four points in the
past:1999,2003,2008,and 2012.Table 1 details the assumed plant parameters for projects
completed in each of these years.Plant costs and performance are based on historical industry
data (Table 2 and Table 3)reportedby Wiser and Bolinger (2012).By emphasizing aggregate
industrydata,the analysis is representative of facilities commissioned in each of the respective
years but not necessarily indicative any single plant's characteristics.
Table 1.Key Capacity-Weighted AverageTurbine Parameters forTechnologyConsideredforRepoweringinThisAnalysis
Year Turbine Rating Hub Height Rotor Diameter
Commissioned (MW)(m)(m)
1999 0.7 56 48
2003 1.2 66 64
2008 17 79 79
2012 2.1 85 95
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)5
at www.nrel.gov/publications.
Full repowering is considered as a possibilityfor each of the existing plants at four points in the
future:2015,2020,2025,and 2030.In addition,for 2003-commissionedplants,the potential for
partial repowering is also considered.Partial repowering is examined explicitlyfor this 2003
plant because of its relativelymodern technology compared with 1999 plants (Table 1),and
unlike plants commissioned later (e.g.,2008 and 2012),it was determined to be a viable
candidate for full repowering within the time period considered in this analysis.For the 2003
facility,partial repowering is analyzed for 2025,the year in which full repowering becomes the
preferred investment relative to an adjacent greenfield.
In order to analyze the viabilityof repowering,conceptual projects with specific cost and
performance characteristics were evaluatedand defined (for full repowering and new greenfield
projects)at each of the prescribed future investment dates (2015,2020,2025,and 2030).Partial
repowering costs and performance were estimated for 2025 (Table 2).Future plant costs,
performance,and estimated PPA prices are based on historical trends,National Renewable
Energy Laboratory (NREL)technological advancement projections detailed in Chapman et al.
(2012),and semi-structured interviews with plant owners.Defining future technology cost and
performance is critical,as repowering decisions are functions of current plant performance and
the value added by investing in new state-of-the-art technology.Key financial modeling inputs
are summarized in Table 3.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)6
at www.nrel.gov/publications.
Table 2.Summary of Financial Modeling Inputs for Existing and Future Facilities by Year of Project Commissioninga
Operations
Installed Capital Cost Net Capacity Factorb PPA Pricee ExpendituresPlant(Year 1)Commission
Date Repower Partial Fixed VariableGreenfieldPartial2012PTC2012RepowerGreenfieldRepower201220122012$lkW Repower $lMWh Available$lkW 2012 $lkW $lkW-yr $lMWh
1999 1,672 28.1%50 Yes 12.5 10.2
2003 1,402 29.8%35 Yes 12.5 8.2
2008 1,998 33.7%57 Yes 12.5 6.1
2012 1,890 35.6%52 Yes 12.5 6.1
2015 1,862 1.769 41.0%41.0%63 No 12.5 6.1
2020 1,816 1,725 42.0%42.0%57 No 12.5 6.1
2025 1,770 1,681 1,504 43.0%43.0%37.2%53 No 12.5 6.1
2030 1,712 1,626 43.0%43.0%51 No 12.5 6.1
aHistorical data are derived fromWiser and Bolinger(2012);future data are derivedfrom NREL cost projection analyses (e.g.,Chapman et al.2012;U.S.
Department of Energy 2008),current industry trends,and semi-structured interviews with owner operators.
b Future net capacity factors are equivalent for greenfield and repowering developments,as the analysis assumes a proximate or adjacent facility in a comparable
wind resource area;the apparent discontinuity between 2012 and 2015 is the result of recent technology advancements that have brought larger rotor and taller
tower machines to the market.Historical data are intended to reflect the performance of plants installed in a specific year.
°The historical PPA prices shown in this table are reported contract values (Bolinger2013).As a result,the PTC should be considered as an additional revenue
stream above and beyond that reported here.This analysis assumes a one-year delay between PPA contract execution date and project commissioning.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)7
at www.nrel.gov/publications.
Underlyingdata sources,trends,and justificationsfor the project cost,performance,and revenue
characteristics summarized in Table 2 are described in greater detail here:
Installed capital cost:Costs declined from 1999 to 2003 but then increased 2004-2010
(Wiser and Bolinger 2012).Today,however,costs have once again begun to decline with
projects in development anticipated to have 20%-30%lower installed costs than during
2009 and 2010 (Wiser et al.2012).Future assumptions include a continual,but slower,
decline in installed costs in line with other independentprojections in the literature (Lantz
et al.2012;Chapman et al.2012).
Fullyrepoweredplants are assumed to have slightly lower (5%)total installed capital
costs as a result of the ability to reuse or repurpose existing plant infrastructure noted
above.Although actual savings will vary from project to project,semi-structured
interviews completed as part of this effort and the literature reviewed above suggest a 5%
cost savings is feasible.Partial repowering is estimated to result in an approximately 15%
cost savings as a result of reusing existing infrastructure as well as cost savings on towers
and foundations.
Net capacity factor:Historical net capacity factors were estimated for plants
commissioned in a specific year and groundedin historical averages reported by Wiser
and Bolinger (2012).However,larger rotors and taller towers,along with other
technological advancements,have resulted in significant increases in net capacity factors
for utility-scalewind plants over the past 13 years (Lantz et al.2012).Changes are
particularlyevident when examined across fixed wind resource areas with an emphasis
on performance resulting from the latest turbine models rather than fleet-wide data
(Wiser et al.2012).Future installations assume net capacity factor improvements as a
result of increased turbine optimization for a given site as well as continued incremental
technology improvements.
Both repowered and new adjacent greenfield facilities are anticipated to bring about
substantial (Wiser et al.2012;Chapman et al.2012)and equivalent improvements in net
capacity factor.The latter is assumed as a result of comparing repowered facilities with
adjacent greenfields where a comparable wind resource is likely.Partial repowering is
assumed to achieve approximately 50%of the anticipated energy production
improvement at a given site that results from full repowering.Energy production gains
from partial repowering are assumed to be limited by the original assets that continue to
be used (e.g.,the tower and foundation).
PPA prices:Existing plant revenues are approximated from historicallybundled
(renewable energy certificates and electricity)PPA price data reported by Bolinger
(2013)and assume the PTC is utilized.Future project revenues assume current federal
policy (PTC expiration at year-end 2013)and PPA pricing estimates that are aligned with
wind power levelized cost of energy projections (Lantz et al.2012).PPA escalation is
estimated to increase about 1%(nominally)per year (Bolinger 2013).The PPA escalation
rate is based on a composite mix of PPA contracts that escalate in line with inflation and
others that have no stated escalation rate.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)8
at www.nrel.gov/publications.
Operationsexpenditures:Historical data (Wiser and Bolinger 2012)indicate that
variable O&M costs have dropped over the past 13 years.Analyses conducted here
assume year one total operation expenditureshave fallen in the past but are constant into
the future (Houston 2013).As facilities age,total operation expendituresare assumed to
escalate at 2%above inflation (a 4.5%nominal escalation rate)."Operating costs are
considered to be equivalent for new greenfield and repowered facilities (both full
and partial).
Table 3.Financial Modeling Constants Across Scenarios
Inflation 2.5%
Discount rate (nominal)9%
Cost of financing (nominal)9%
PPA escalation rate (nominal)1%
O&M escalation rate (nominal)4.5%
Data Sources:Wiser et al.(2012),Bolinger(2013),Tegen et al.(2012),and the Bureau of Labor Statistics (2012)
The evaluation of investment options is based on a comparison of the net present value (NPV)of
future after-tax cash flow.Specifically,the change in projected NPV for repowered facilities is
compared with the change in projected NPV from future cash flows of an existing facility plus
the NPV from a new adjacent greenfield plant.This comparison is appropriatebecause a
greenfield investment decision allows the existing plant to continue generatingrevenue,while
repowering replaces the existing plant's revenue stream with an alternate one (the repowered
revenue stream).After-tax cash flows are included in the NPV as long as they remain positive or
until 30 years of operation.6 Potential positive cash flows estimated after 30 years of operation
are not considered for any projects (existing,repowered,or greenfield)as a result of potential
step functions for operating costs that are not captured by applying a generic operating expense
escalation rate (Table 3).
NREL's System Advisor Model (SAM)was used to calculate expected future cash flow.Post
analysis processing of these data allowed calculation of the NPV for a project's remaining cash
flow and development of a uniform comparison in 2012 dollars for each of the potential future
mvestment pomts.
2.2 Resu ts
The data presented here represent the change in estimated NPV (2012 dollars)of after-tax cash
flow resulting from the addition of a greenfield plant (i.e.,the combined cash flows of the
existing plant and the new adjacent greenfield)or repowering of each representative historical
wind plant detailed above.In effect,the data illustrate value gained or lost as a result of a
specific investment decision.As each of these plants is modeled at an equivalent size,plant-
specific NPVs can be compared across time.However,caution is advised against any direct
assessment of wind plant profitability or return on investment as the overall magnitude of NPV is
highly correlated to plant size.Moreover,as the data presented represent the change in NPV
'This is the approximate annual inflationrate necessary to achieve a levelized pre-tax O&M cost comparable to the
value assumed by Tegen et al.(2012).
6 Negativeafter-tax cash flows are excluded based on the premise that plants will not be operated at a loss.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)9
at www.nrel.gov/publications.
relative to simply maintaining the existing facility,they do not represent the full NPV from a
given investment.Results from both the repoweredand greenfield facilities assume a common
reference project size of 100 MW.
Based on the inputs and analyses completed,wind power plants built in 1999 appear to be
reasonablyprofitable moving into their 14th ÿCRT Of Operation.In part,this is the result of
relativelyfavorable PPAs signed during this time and relativelylow capital costs.Nevertheless,
the age of this facility and the advancements in technology that have occurred to date suggest
that action in the near term is likelymerited.Data analyzed here indicate that in 2015,after 15
full years of operation,overall profitability for a plant built in 1999 can be enhanced by
repowering or by buildingan adjacent facility (Figure 1).If asked to choose the more lucrative
option,this analysis indicates that developing an adjacent greenfield adds more value than
simplyrepowering.By 2020,after 20 years of operation,the result is relatively similar with both
alternatives adding value,but the combined assets of the new greenfield and the existing facility
appear to be most profitable.
Full repowering becomes more attractive than an additional new greenfield sometime between
20 and 25 years of operation,for plants built in 1999;it is an increasingly attractive alternative as
time progresses.
$30 ----------------------------------------------------------------------------
o .
Existing +Repower Existing +Repower Existing +Repower Existing +Repower
Green Green Green Green
2015 2020 2025 2030
Figure 1.Value added to a 1999 wind plant as a result of investing
in a new greenfield or full repowering
Note:Assumes common reference plantsize of100 MW
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)10
at www.nrel.gov/publications.
Figure 2 illustrates,in the same manner,the results for a wind plant commissioned in 2003.For
this facility,analysis suggests that building an adjacent greenfield plant in 2015 and 2020 (after
11 and 16 years of operation,respectively)is again the preferred alternative relative to full
repowering.However,in this case it is between 16 and 21 years of operation (2020-2025)that
repowering becomes financiallypreferable,somewhat earlier in the life of the plant than for the
1999 plant.In part,the somewhat shorter anticipated life for the 2003 plant is a function of its
lower estimated PPA price and the relativelyslight improvement in net capacity factor observed
for the 2003 facility relative to the 1999 facility.Combined with lower profitabilityoverall for
this plant,increasing operational costs begin to erode the value of 2003 vintage plants earlier in
their life than plants commissioned in 1999 or later.
$30 ----------------------------------------------------------------------------
LI.I
Existing +Repower Existing +Repower Existing +Repower Existing +Repower
Green Green Green Green
2015 2020 2025 2030
Figure 2.Value added to a 2003 wind plant as a result of investing
in a new greenfield or full repowering
Note:Assumes common reference plantsize of100 MW
Results for the plants commissioned in 2008 and 2012 are presented in Figure 3 and Figure 4,
respectively.For both the 2008 and 2012 plants,building an adjacent new greenfield project is
the financiallypreferable option through 2030.In fact,throughoutthis time period (as many as
21 and 17 years of operation,respectively),full repowering results in a reduction in the NPV of
future after-tax cash flows.For the 2008 facility,this is in part the result of historicallyhigh
7 The authors note,however,that even though this analysis indicates that repowering might be financially attractive
after only 16-21 years,owners/operators might not ultimately choose to repower within this timeframe as a result of
project-specific financing,depreciation,or other constraints.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)11
at www.nrel.gov/publications.
PPAs signed during this period.Also playing a role are the assumed declines in PPA pricing
associated with future technology advancements and cost reductions.Regardless of drivers,more
recent plants appear to have substantial value remaining beyond 20 years.Accordingly,these
projects could delay repowering investments until 25 years of operation or beyond.
$40 -----------------------------------------------------------------------------
.$(50)------------------------------------------------------------------------
$(70)-------------------------------------------------------------------------
Existing +Repower Existing +Repower Existing +Repower Existing +Repower
Green Green Green Green
2015 2020 2025 2030
Figure 3.Value added or lost as a result of investing in a new greenfield or
full repowering for a 2008 wind power plant
Note Assumes common re erence plantsize o 100 MW
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)12
at www.nrel.gov/publications.
$40 ----------------------------------------------------------------------------
.$(100)-----------------------------------------------------------------------
$(120)----------------------------------------------------------------------------
Existing +Repower Existing +Repower Existing +Repower Existing +Repower
Green Green Green Green
2015 2020 2025 2030
Figure 4.Value added or lost as a result of investing in a new greenfield
or full repoweringfor a 2012 wind power plant
Note:Assumes common reference plantsize of100 MW
Partial repowering is examined for the 2003 plant only.Figure 5 demonstrates the results for
partial repowering relative to full repowering as well as building a new adjacent greenfield
facility.Results are shown specifically for 2025 when full repowering was determined to be the
preferred alternative.Based on the inputs summarized in Table 2 and Table 3,partial repowering
offers less added value to an owner/operator than either full repowering or building an adjacent
greenfield.This is primarily due to the lower net capacity factor resulting from not replacing the
tower,and hence not accessing stronger winds aloft,coupled with the modest additional cost
savings to be gainedby keeping the existing tower and foundation intact.It should be noted that
this conclusion is based on the concept of partial repowering that was identified and defined
earlier.Were innovations to emerge that boost the productivityof aging equipment at less cost
than estimated here,partial repowering could be more attractive.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)13
at www.nrel.gov/publications.
$14 -----------------------------------------------------------------------------
o $12 ------------------------------------------------------------------
ELI
E -$10 ------------------------------------------------------------------
ON
$8 --------------------------------------------------------
-O
z $4 ---------------------------------------------
Existing +Green Repower Partial Repower
Figure 5.Value added to a 2003 wind plant as a result of investing in a new greenfield,
full repowering,or partial repowering in 2025
Note:Assumes common re erence plantsize o 100 MW
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)14
at www.nrel.gov/publications.
3 Case Study Ana ys s
The financial modeling completed within the general economic feasibilityassessment is
supplemented by additional modeling and semi-structured interviews with project owners.These
efforts focus on repowering opportunities within three specific regions of the United States:the
Northeast,Midwest,and West Coast.This approach offers a more nuanced perspective on the
repowering opportunity.In addition,this supplemental work serves as a check on the more
general conclusions drawn from industry-wide analysis covered in Section 2.
3.1 Methods
Analytically,the case study analysis was carried out in an equivalent manner as the more general
economic analysis.Modeling was performed using SAM,and the change in NPV of after-tax
cash flows was compared across the two specific investment opportunities (repowering or
developing an adjacent greenfield).Notably,the case study analysis focuses only on full
repowering,as partial repowering was determined to offer less added value than full repowering
(Figure 5).However,unlike the more general assessment,which emphasized aggregated industry
data,the case study approach developedmodeling inputs for existing plants that were specific to
actual projects within the three regions considered.Table 4 summarizes the general
characteristics and modeling input variables for the plants chosen for analysis.
Criteria for plant selection included geographic diversity,an operation history of at least
10 years,and the presence of recent,or ongoing,development at an adjacent greenfield site.
Choosing existing wind plants with new greenfield sites in active developmentprovided a more
refined look at real-world factors such as installation costs,expected performance,and PPAs that
were appropriate for those locations at the present time.In principle,this approach ties the case
study analysis to more realistic conditions faced by project owners.
To gather relevant data for these sites,a three-step process was undertaken.First,Web-based
inquiries were conducted with the intent of extracting key project characteristics for the existing
plants and the adjacent greenfield projects.Characteristics such as number of turbines,nameplate
capacity,and expected initial net capacity factor were collected from project websites and
archived news articles.Second,spreadsheet forms were distributed to contacts at all of the
companies that own or operate the respective wind plants or those who are developing the nearby
greenfield sites.Owner/operator contacts were asked to provide supplemental data to fill in the
gaps that were missing from publicly available sources.Third,relevant modeling input data were
solicited from the industrydatabases maintained by Lawrence Berkeley National Laboratory and
summarized by Wiser and Bolinger (2012).Where data were not available,estimates based on
comparable projects (both in terms of age and technology type)were developedto allow for
completion of the financial modeling.Given some persistent data gaps,follow-up semi-
structured interviews with plant owners/operators were conducted to obtain reactions and review
of results.Final modeling input data and assumptions are summarized in Table 4 and Table 5.
*Note that Table 4 contains ranges to protect the confidentialityand anonymity of the specific plants
being analyzed.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)15
at www.nrel.gov/publications.
Table 4.Key Characteristics and Assumptions for the Three Case Study Wind Plants
Avera e Year i OpExCommercialTurbinePPAInstallationFederal
Operation Size Plant Rated Annual Net Pricel Cost Incentive Fixed Variable
Date (MW)Capacity Capacity MWh 2012$lkW Utilized (2012$/(2012$/Factor kW-yr)MWh)
Northeast Wind Plant
Existing 1995-2000 0.5-1.0 5-10 MW 20%-25%91 2,400-2,900 PTC 12.5 11.2
Nearby greenfield 2010-2015 1.5-2.5 25-35 MW 32%-37%88 2,400-2,900 PTC 12.5 6.1
Repower existing"2010-2015 1.5-2.5 25-35 MW 32%-37%88 2.300-2.800 PTC 12.5 6.1
West Coast Wind Plant
Existing 1988-1993 <0.5 45-55 MW 17%-22%135 2,300-2,800 None 12.5 13.3
Nearby greenfield 2010-2015 1.5-2.5 95-105 MW 25%-30%b 115 2,900-3,400 PTC 12.5 6.1
Repower existing"2010-2015 1.5-2.5 95-105 MW 28%-33%115 2,700-3,200 PTC 12.5 6.1
Midwest Wind Plant
Existing 1997-2002 0.5-1.0 10-15 MW 32%-37%69 1,700-2,200 PTC 12.5 10.2
ITC/1603
Nearby greenfield 2010-2015 1.5-2.5 25-35 MW 35%-40%68 1,700-2,200 Cash 12.5 6.1
Grant
11011603
Repower existing"2010-2015 1.5-2.5 25-35 MW 35%-40%68 1,600-2,100 Cash 12.5 6.1
Grant
a Installation costs for "Repowerexisting"were again assumed to be 5%less than building a nearby greenfield site based on semi-structured interviews with
developers/owners.
b For the West Coast case study,interviews with owners/operators indicated that finding a nearby greenfield site with as strong a wind resource as the existing
wind site was unlikely giventhe large number of wind plants that have already been installed in this location.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)16
at www.nrel.gov/publications.
Table 5.Financial Modeling Constants Across Case Study Scenarios
Inflation 2.5%
Discount rate (nominal)9%
Cost of financing (nominal)9%
PPA escalation rate (nominal)a 1%
O&M escalation rate (nominal)"4.5%
a The Midwest plant PPA's escalation rate was assumed to be 0.2%for the existing plant,and 0%escalation for the
greenfield and repower scenarios per guidance fromsemi-structured interviews.Moreover,as reported by Bolinger
(2013)there are many PPAs signed today that are consistent with this approach and are constant in nominal terms
with time.
b O&M escalation rate for the existing West Coast wind plant was assumed to be 5.5%givenits earlier
commissioning date.
3.2 Resu ts
For the Northeast wind plant,which has 15-20 years of operation,both building a new
greenfield site and repowering would increase the NPV of projected after-tax cash flows (Figure
6).Moreover,these two possible investment opportunities result in very similar NPVs,within
1%.Thus,although repowering does appear to be a viable option,one cannot conclude it is
clearly favorable to building a new greenfield site.
For the West Coast plant,which is 20-25 years old,both buildinga new greenfield site and
repowering also increase the NPV of projected cash flows above and beyond those resulting
from simply maintaining the existing plant.However,in this case,repowering the existing wind
plant was notablymore lucrative than building a new greenfield site (Figure 6).In part,this is a
result of the slightly lower net capacity factor assumed for a nearby greenfield site relative to a
repowered facility.This assumption was applied in this specific case as a result of interviewees
observing that the West Coast plant would have difficultyacquiring land to build an adjacent
greenfield site with a similar high-qualitywind resource.However,the more critical factor is
likelythe overall age of the facility and its relativelylow remaining cash flow.Under these
conditions,the incremental cost savings from use of a portion of the existing infrastructure
coupled with the slightly better wind resource at the repowered site result in repowering being
the most attractive investment opportunity.
As in the previous two cases,building a new greenfield site and repowering the existing wind
plant both resulted in an increase in the NPV of projected after-tax cash flows for the Midwest
wind plant.However,as this facility has been in operation for only 10-15 years,and there are
presumably many years of substantial profitability from this plant,building a new greenfield site
is more lucrative than repowering the existing plant.
This location is already extensively developed,making it relativelydifficult to acquire adjacent land.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)17
at www.nrel.gov/publications.
$70 -----------------------------------------------------------------------------
Northeast:$40 ---Existing plant -------------------------------------
o 15-20 yrs old West Coast:
$30 -----------Existing plant -
>0-20-25 yrs old
$0
Existing +Repower Existing +Repower Existing +Repower
Green Green Green
Figure 6.Value added to each case study wind plant as a result of
investing in a new greenfield or full repowering
Note:Comparing the NPV across the three case studies is not appropriate.The absolute magnitude of the NPV is
highly correlated with the size of the wind plant,as larger wind plants require higher levels of investment.Within
each case study,it was always assumed that both greenfield and repowering decisions wouldbe of the same size
(i.e.,same rated capacity)and thus can be fairly compared.
The results of these case studies support the findings generated by the economic analysis
summarized in Section 2.Only after a wind plant has been in operation for more than 20 years
does repowering provide a clear financial advantage relative to building a nearby greenfield site.
Interestingly,the simulated results compiled here appear to align relativelywell with current
owner/operator financial assessments.Only one of these three plants-theWest Coast plant-
has been repowered to date.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)18
at www.nrel.gov/publications.
4 Market pportun ty
Both the general economic assessment and the more specific regional case study analysis suggest
repowering is likely to become attractive after 20-25 years of operation.'°An anticipated plant
life of 20-25 years is generally consistent with the typical contract life for wind facilities.
Historically,wind power contracts have been 20 years in length with 15-and 25-year contracts
signed as well (Bolinger 2013).Having identified the expected timeframe in which projects will
consider repowering,the next element of this analysis is to quantifythe market opportunity
provided by repowering activity through 2030.
4.1 Methods
The market opportunitywas quantified in terms of annual repowered capacity and expected
value in constant 2012 dollars.Estimates are premised on 25%of existing facilities repowering
after 20 years of operation and 50%repowering after 25 years of operation.The remaining 25%
of the existing fleet is assumed to either continue to operate after 25 years or be
decommissioned.As empirical data on actual repowering behavior are extremely limited,these
estimates are intended for hypothetical purposes only.Given this timeframe,only existing plants
will be making repowering decisions by 2030.Accordingly,this task relied on installations
contained in the U.S.industryprojects database (American Wind Energy Association 2012)to
determine the potential installed capacity that might be repowered on an annual basis over the
next two decades.Annual installed capacity estimates are calculated assuming a 1:1 replacement
of capacity.However,if repowering results in increased capacity at a given site,as has been
observed in Germany and Denmark,additional demand could result.The total repowering
market value is estimated based on the calculated plant lifetime,projected turbine pricing,and
installed cost estimates.
This approach represents a high-levelestimate of the potential repowering opportunity.The
actual magnitude of the repowering market will likelyvary;nevertheless,the results presented
here are anticipated to capture the general order of magnitude of the market segment in terms of
both installed capacity and dollar value.
4.2 Results
The estimated plant life applied here suggests that repowering could be considered for about
40 gigawatts (GW)of the operating wind power fleet within the next two decades (i.e.,plants
commissioned by year-end2010).Moreover,it suggests that there will be a substantial increase
in repowering activity in the early 2020s as turbines installed in the late 1990s and early 2000s
begin to approach the 20-to 25-year threshold where repowering begins to be attractive.
Assuming 25%of the existing fleet repowers at or around year 20 of operation and 50%
repowers at or around year 25 of operation,annual repowering activity in terms of megawatts
repowered through 2030 are highlighted in Figure 7.With these conditions the effect is that 75%
of the fleet does not repower before 25 years of operation;accordingly,the cumulative
repowered capacity by 2030 is estimated to be slightly less than 14 GW.
10 If technological advances are more rapid,costs for repowering are substantially lowerrelativeto a new greenfield
plant,or if capacity factors at new greenfields are notably lower than at repowered facilities,repowering could occur
earlier than projected in Section 2 and Section 3.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)19
at www.nrel.gov/publications.
3,500 -----------------------------------------------------------------16,000
14,0003,000 --------------------------------------------------------------
--Annual MW Repowered -Left Axis -12,0002,500 ------------
--Cumulative MW Repowered -Right Axis e
-10,000 y
no 2,000 ----------------------------------------------------------
-8,000 i
§1,500 -------------------------------------------------------.g
-6,000
1,000 -----------------------------------------------------
-4,000 3
"500 -------------------------------------------------
-2,000
0 iiiiiiiiiiiiiiiii O
2013 2015 2017 2019 2021 2023 2025 2027 2029
Figure 7.Estimated annual capacity repowered by year
Note:Results assume 1 MW of existing capacity is replaced by 1 MW of repowered capacity.
The total future value of repowering activity through 2030 is estimated at $20-$25 billion.Once
repowering begins to take an active foothold in the market around 2023,the annual value of the
market segment is anticipated to be roughly $1 billion per year through 2025 and increases to
about $5 billion per year in 2028.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)20
at www.nrel.gov/publications.
5 Supp y Cha n mpacts
Based on the potential range of incremental new demand resulting from repowering activities,
impacts to the supply chain are calculated in terms of demand for new blades,drivetrains,and
turbines.As a result of the uncertainties noted above-in terms of the percentage of eligible sites
that actuallyrepower and the percentage of partial versus full repowering-thisanalysis seeks
simplyto provide a basic order of magnitude estimate for changes in demand,given the
estimated levels of repowering and assuming that all repowering activity is full repowering.
5.1 Methods
Projected technological advancement"is coupled with anticipated repowering activity to
estimate total repowering demand and demand for specific component sizes.Given the relatively
strong wind resource areas where those facilities projected to repower over this timeframe are
located (Wiser and Bolinger 2012),the majority of demand is expected to be for International
Electrotechnical Commission (IEC)Class II turbines.The results are briefly compared with the
existing U.S.wind manufacturing base to inform the ability of the current supply chain to serve
the repowering market.
5.2 Resu ts
Total turbine demand is estimated from the installed capacity estimates presented in Section 4.
Assuming 2-to 3-MW turbines constitute the bulk of capacity during the late 2010s and early
2020s,an average of approximately 60 turbines per year will be needed to serve repowering
investment from 2018 to 2021.Assuming that 3-to 4-MW turbines become more prevalent after
2021,an average of approximately 230 turbines per year will be needed to serve the repowering
demand from 2022 to 2027.From 2028 to 2030,an average of about 730 turbines per year will
be needed (Figure 8).
900 ------------------------------------------------------------------------
-o 800 -------3-4 Mw Turbines ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
E 700 -----·----------------------------------------------
,2-3 MW Turbines
500 ------------------------------------------------------------------
400 ----------------------------------------------------------
.300 ----------------------------------------------------------
2014 2016 2018 2020 2022 2024 2026 2028 2030
Figure 8.Estimated turbine demand by nameplate capacity
resulting from potential repowering activity
Based on extrapolation of historical trends as well as analysis and research summarized by Cohen et al.(2008)
and Chapman et al.(2012).
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)21
at www.nrel.gov/publications.
Blade and tower demand by size are presented in Figure 9 and Figure 10,respectively.Based on
the technology projections applied here,the majorityof blades could be on the order of 50-60
meters (m)with some IEC Class III turbines being utilized in the late 2020s and subsequently
requiring longer 70-to 80-m blades.Similarly,the majority of towers required for sites that are
repowered (i.e.,Wind Power Class 5 and above wind resource sites)are expected to be 80-to
100-m towers with some 120-to 130-m towers in the later years.
3,000 ------------------------------------------------------------------------------
2,500 ------------------------------------------------------------------------------
o 2,000 --------------------------------------------------------------------
1,500 --------------------------------------------------------------------
LLI
1,000 ------------------------------------------------------------
70-80m Blades
-50-60m Blades
2014 2016 2018 2020 2022 2024 2026 2028 2030
Figure 9.Estimated blade demand by length (meters)resulting from potential repowering activity
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)22
at www.nrel.gov/publications.
900 ------------------------------------------------------------------------------
800 ----------------------------------------------------------------------------
700 --------------------------------------------------------------------------
e 600 ------------------------------------------------------------------------
500 -----------------------------------------------------------------------
o
400 -------------------------------------------------------------------
120-130m Towers
-300 -------------------------------------------------80-100m Towers
2014 2016 2018 2020 2022 2024 2026 2028 2030
Figure 10.Estimated tower demand by height (meters)resulting from potential repowering activity
U.S.wind installations topped 13 GW per year in 2012.With incremental demand from
repowering expected to be below 500 MW through the early 2020s and only about 3 GW by the
late 2020s,supply chain constraints are not expected to be a major impediment to repowering.
Were partial repowering to become more viable than observed here,production capacity for
somewhat smaller components geared toward refurbishing older equipment might need to be
developed.However,as this analysis suggests that repowering will be dominated by full
repowering efforts utilizingthe same state-of-the-art technology being installed on greenfields,
such additional investment is not foreseen.At the same time,repowering cannot be expected to
create notable supplemental demand in the U.S.supply chain until at least the early to mid-
2020s.Were the U.S.supply chain forced to downsize dramatically between today and the mid-
2020s,supply chain constraints could emerge as repowering demand picks up once again in the
latter half of the 2020s.Assuming a 20-to 25-year life suggests that repowering demand will
jumpagain in the early 2030s.Nevertheless,as these increases can be anticipated and predicted,
they are not expected to result in significant supply chain hurdles.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)23
at www.nrel.gov/publications.
6 Po cy and ther Re evant Cons derat ons
The analyses conducted here suggest that repowering activity will remain modest in the United
States through the early to mid 2020s;however,repowering decisions are not often driven by
economic considerations alone.The followingdiscussion highlightsthe perspectives of wind
industryprofessionals with regard to the impacts of current policy on repowering decisions as
well as areas where future policy may be applied to facilitate or discourage repowering.
Semi-structured interviews indicated that current federal policy has only a modest impact on
repowering decisions.In some instances,repowering investment may be accelerated to take
advantage of the PTC or accelerated depreciation provisions similar to other wind power
development activity,but generally,no substantial impact was observed.Similarly,
Environmental Protection Agency rules under development that may encourage the retirement of
existing generationcapacity could create more demand for wind power generally but are unlikely
to directly affect decisions specific to repowering.
Multipleinterview respondents observed that repowering could offer the possibilityfor
expeditedpermitting because development is occurring at a previouslypermitted site and these
sites have a much longer track record of historical environmental data.Existing plants also tend
to have increased local familiarity,which might also facilitate the permitting process.The ability
to expedite the permitting process would suggest,perhaps,that repowering could take place
somewhat sooner than envisioned by the financial analyses above.However,development costs,
includingpermitting,are low relative to the total capital expendituresassociated with modern
wind facilities (Tegen et al.2011)suggesting that this effect may be relatively modest.It was
also observed that where endangered species live or other environmental damage has occurred,
permitting would likelybe more complex and drawn out,with the likelyeffect of postponing
repowering activity.Repowering activity was also noted to be likelyto occur later in a plant's
life if local siting requirements become more stringent over time.
Along with more general policy considerations,respondents noted that the contractual language
around an existing PPA could potentiallyraise a number of red flags when considering
repowering.For example,wind professionals expressed concern about potential outage clauses
and minimum generationthresholds as the physical act of repowering would likelyinvolve
substantial plant downtime.Respondents were also concerned about the contractual implications
resulting from potential changes in production associated with putting new more productive
technology at a specific site or from the potential addition of more capacity at a site.It was also
noted that past PPAs tended to be more generous,so project owners are hesitant to take any
action that would jeopardize an existing set of terms,conditions,and payments.Such a condition
is comparable to that described earlier as the "California Fix,"whereby repoweredplants that
take the federal PTC are anticipated to result in the loss of relativelygenerous standard offer
contracts for many California wind projects built in the 1980s.
Within this context,if the various attributes of repowering are valued,repowering activities
could be accelerated.However,encouraging and accelerating repowering investment is likelyto
require a shift in policy.Policy changes would likelyneed to address the current economic
incentives aroundrepowering,but may also consider the various permitting,regulatory,and
contractual risks associated with repowering.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)24
at www.nrel.gov/publications.
7 Sumrnary and Conc us ons
In line with prior work,this report's principal conclusion is that wind power plants that continue
to generate a positive after-tax cash flow are unlikelyto pursue repowering in the immediate
future.Based on the analysis completed here,it is estimated that existing plant viabilityremains
strong up until 20-25 years of operation,at which point repowering becomes more attractive
than a new greenfield addition.Before this time,the continuing revenue stream from the existing
plant augments any new greenfield site's future revenue stream,making repowering less viable.
This timeframe for repowering was supported both by our aggregated representativewind plant
financial analysis and by our case studies.Once the after-tax cash flow falls below a given
threshold (after 20-25 years),the installed cost savings associated with repowering and resulting
from the ability to use existing infrastructure,includingroads,buildings,and electrical
equipment,shifts the balance in favor of repowering.
As defined here,partial repowering appears to offer less value than full repowering.Estimated
cost savings of approximately 10%relative to full repowering that result from reusing the
existing tower and foundations did not offset the reduction in energy generation due to the lower
hub height and smaller rotor anticipated for the partially repoweredturbine.Accordingly,partial
repowering is unlikelyto be pursued unless innovations emerge that can boost plant productivity
or reduce operating costs with less investment than was anticipated here.
In brief,wind industryinvestors looking to expand their profitability over the next decade are
likelyto see a greater increase in NPV from developing adjacent new greenfield sites than fromrepowering.12 Nevertheless,there is an array of variables that could affect the attractiveness of
repowering and either extend or shorten operating plant life including:
Technologicaladvancement:More rapid advancement will encourage repowering while
slower advancement will reduce the viabilityof repowering.
Wind resource regime for greenfieldplants:A wind resource that is lower qualitythan
that of existing facilities will encourage repowering while higher-qualitywind resource
areas potentiallyopened up by new transmission will encourage further greenfield
mvestment.
PPA prices:Both higher prices for future repoweredplants and,alternatively,lower PPA
prices for existing plants would encourage repowering to occur earlier.
Operationexpenditures:More rapid cost escalation as facilities age will make
repowering more attractive earlier.
Repoweringcost savings (relativeto a greenfieldproject):The ability to capture >5%
cost savings from repowering activities will encourage repowering;potentiallyhigher
costs (than modeled here)for repowering will discourage repowering investment.
12 Assuming,of course,that new greenfield plants are viableinvestments independent of the repowering
considerations.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)25
at www.nrel.gov/publications.
Assuming a plant life of 20-25 years,repowering demand is expected to be low over the next
decade,reach a few hundred megawatts per year in the early 2020s,and achieve 1-3 GW by the
late 2020s.The total estimated value of the repowering market segment is estimated at $25
billion through 2030 with the vast majorityof this investment occurring in the latter half of the
2020s.Turbine demand is anticipated to range from well below 100 turbines per year in the early
2020s to slightly more than 700 turbines per year by the late 2020s.Tower and blade demand is
anticipated to be concentrated on the state-of-the-art IEC Class II turbine designs suggesting that
repowering could support an incremental increase in supply chain activity.
Policy support in the form of financial incentives,as well as solutions to potential regulatory and
contractual hurdles,is likelyto be necessary if there is value placed on the accelerationof
repowermg mvestment.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)26
at www.nrel.gov/publications.
efe rences
American Wind Energy Association (AWEA).(2012).Projects Database.Washington,DC:
American Wind Energy Association.
Bolinger,M.(2013).Revisiting the Long-Term Hedge Value of Wind Power in an Era of Low
Natural Gas Prices.LBNL-6103e.Berkeley,CA:Lawrence Berkeley National Laboratory.
BundesverbandWindEnergie (German Wind Energy Association)(BWE).(2008).Germany:
Newsletter Article.Accessed March 19,2012:http://www.gwec.net/index.php?id=77&L=0&tx_ttnews[backPid]=76&tx_ttnews[pointerl=8&tx
ttnews[tt news]=135&cHash=4a5a383387.
BWE.(2011).GWEC:Germany.Accessed July 17,2013:http://www.gwec.net/index.php?id=129.
Chapman,J.;Lantz,E.;Denholm,P.;Felker,F.;Heath,G.;Mai,T.;Tegen,S.(2012)."Wind
Energy Technologies,"Chapter 11.Renewable Electricity Futures Study,Vol.2,Golden,CO:
National Renewable Energy Laboratory;pp.11-1 -11-63
Cohen,J.;Schweizer,T.;Laxson,A.;Butterfield,S.;Schreck,S.;Fingersh,L.;Veers,P.;
Ashwill,T.(2008).Technology Improvement Opportunities for Low Wind Speed Turbines and
Implications for Cost of Energy Reduction.NREL/TP-500-41036.Golden,CO:National
Renewable Energy Laboratory.
European Wind Energy Association.(2009).Economics of Wind Energy.Brussels:European
Wind Energy Association.
Fairley,P.(2009).Europe Replaces Old Wind Farms -IEEE Spectrum.IEEE.Accessed
September 2012:http://spectrum.ieee.org/green-tech/wind/europe-replaces-old-wind-farms.
Filgueira,A.;Seijo,M.A.;Munoz,E.;Castro,L.;Piegari,L.(2009)."Technical and Economic
Study of Two Repowered Wind Farms in Bustelo and San Xoan,24.7 MW and 15.84 MW
Respectively."International Conference on Clean Electrical Power (pp.545-549).IEEE.
GlobalData.(8 March 2012).Global Data Press Releases.(GlobalData,Producer).Accessed
March 20,2012:www.globaldata.com/PressRelease/Details.aspx.
Goyal,M.(2010).Repowering-NextBig Thing in India.Renewable and Sustainable Energy
Reviews (14);pp.1400-1409.
Grontmij.(n.d.).Grontmij Highlights.Grontmij.Accessed March 20,2012:
http://www.grontmij.com/highlights/water-and-energy/Pages/Repowering...d-turbines-in-the-
Netherlands-produces-more-sustainable-energy.aspx.
Hotker,H.(2006).The Impact of Repowering of Wind Farms on Birds and Bats.Michael-Otto-
Institute within NABU -Research and Education Centre for Wetlands and Bird Protection.
Bergenhusen:NABU.
Houston,C.(March 2013).The Real Truth About O&M Costs in the U.S.Oxford,Conneticut:
North American Wind Power.
Hulshorst,W.(2008).Repowering and Used Wind Turbines.Econ International.Brussels:
Leonardo Energy and The European Copper Institute.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)27
at www.nrel.gov/publications.
Kharul,R.V.(26 November 2008)."Repowering of Old Wind Farms in India."Pune,India:
World Institute of Sustainable Energy.
Knight,S.(May 2004)."The Art of Selling RepoweredTurbines."WindPower Monthly,pp.61-
63.
Lantz,E.;Wiser,R.;Hand,M.(2012).The Past and Future Cost of Wind Energy.NREL/TP-
6A20-53510.Golden,CO:National Renewable Energy Laboratory.
Meyerhoff,J.;Ohl,C.;Hartje,V.(2010)."Landscape Externalities from Onshore Wind Power."
Energy Policy (38);pp.82-92.
Munksgaard,J.;Morthorst,P.E.(2008)."Wind Power in the Danish Liberalised Power Market-
Policy Measures,Price Impact,and Investor Incentives."Energy Policy (36);pp.3940-3947.
Ramesh,M.(n.d.)."Repowering May Become the Game-Changer for Wind Sector."The Hindu
Business Line.
RWE Innogy.(n.d.).Wind Onshore in the Netherlands.Accessed March 20,2012:
http://www.rwe.com/web/cms/en/580970/rwe-innogy/sites/wind-onshore/netherlands/volkerak/.
Smallwood,K.S.;Neher,L.(28 April 2010)."Siting Repowered Wind Turbines to Minimize
Raptor Collisions at the Tres Vaqueros Wind Project,Contra Costa County,California."
Accessed March 27,2012:
http://www.efsec.wa.gov/Whistling%20Ridge/Adjudication/Intervenor's%20pre-
filed%20testimony/Ex%2022.04.pdf.
Sperling,K.;Hvelplund,F.;Vad Mathiesen,B.(2010)."Evaluation of Wind Power Planning
Denmark -Towards an Integrated Perspective."Energy (35);pp.5443-5454.
Tegen,S.;Hand,M.;Maples,B.;Lantz,E.;Schwabe,P.;Smith,A.(April 2012).2010 Cost of
Wind Energy Review.Golden,CO:National Renewable Energy Laboratory.
U.S.Bureau of Labor Statistics (BLS).(2012)."CPI Inflation Calculator."Accessed July 17,
2013:http://www.bls.gov/data/inflation_calculator.htm.
U.S.Department of Energy (DOE).(2008).20%Wind Energy by 2030:Increasing Wind
Energy's Contribution to U.S.ElectricitySupply.DOE/GO-102008-2567.Washington,DC:
DOE.
White,P.;Gipe,P.(1993).Repowering California Wind Power Plants.Washington,DC:
American Wind Energy Association.
Wiser,R.;Lantz,E.;Bolinger,M.;Hand,M.(2012).Recent Developments in the Levelized Cost
of Energy From U.S.Wind Power Projects.Berkeley,CA:Lawrence Berkeley National
Laboratory.
Wiser,R.(13 March 2007)."Barriers and Incentives for Wind Repowering in Europe and
Elsewhere."Sacramento,CA:California Energy Commission.
Wiser,R.;Bolinger,M.(2012).2011 Wind TechnologiesMarket Report.DOE/GO-102012-
3472.Washington,DC:DOE Office of Energy Efficiency and Renewable Energy.
Wiser,R.;O'Connell,R.;Bolinger,M.(2008).A Scoping-Level Study of the Economics of
Wind-Project Repowering Decisions in California.Burlington,MA:KEMA,Inc.California
Energy Comission.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)28
at www.nrel.gov/publications.
Append x A:nterv ew Gu de for nd wner
perators
Semi-structured interviews with plant owners and operators provided insights into their reasons
for having repowered or not,as well as an opportunityto acquire feedback on earlier modeling
inputs and results.This appendix includes the interview template and summarizes the themes that
emerged from these discussions.In total,extended interviews were completed with eight wind
power professionals,representingthe financial community,owner/operator/developer firms,and
utility owners.
Template
Part 1:Experiences and Perspective on Repowering
1.Project Overview
A.Explanation of our project
B.Why we are calling
C.What we mean by repowering
2.Have you been involved in any projects involvingrepowering of older wind plants?(if
no,move down to #3)
(If ves),please describe what was done in the repowering process.
A.How many turbines were removed?Rated capacity of each?
B.How many new ones were installed?Rated capacity of each?
C.Was the plant'stotal rated capacity increased?How much?
D.Were the towers replaced?
E.What other components,if any,of the balance of plant was modified?
F.Which of the following,if any,were reasons supporting this repowering decision?
You may choose more than one response.
i.Increasing total plant rated capacity
ii.Lowering O&M costs
iii.Increasing net capacity factor
iv.Adhering to environmental regulations
v.Technology was outdated
vi.Equipment was at the end of its useful life
G.Was a new environmental impact study conducted before the repowering was
approved?
H.What was done with the old nacelle,rotors,and towers when the plant was
repowered?
i.Refurbished and sold?
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)29
at www.nrel.gov/publications.
ii.Recycled?
iii.Landfill?
3.What are the primary benefits of repowering an existing wind plant?
4.What major obstacles,if any,should a wind plant owner consider when evaluating the
feasibility of repowering?
5.When evaluating the decision whether to invest in a new nearby greenfield site versus
repowering an existing wind plant,what are the major factors that an owner/operator
should consider?
I.Describe the advantage of repowering an existing wind plant relative to investing
in a new greenfield site.
J.Conversely,describe the advantage to investing in a new nearby greenfield site
relative to repowering an existing wind plant.
6.At what age do you think it is most appropriate to repower a wind plant?
7.Please discuss what impact,if any,the structure of current government incentives
supporting wind energy have on the decision to repower an older wind plant?
8.What impact,if any,does a wind plant's current PPA agreement for an existing wind
farm have on the decision to repower?
9.What should be done with older wind plant equipment (nacelle,rotors,and tower)when
taken down?
Part 2:Case Study Examples
Now we would like to hear your perspective on the three case studies we prepared.We compared
the financial projections for three possible investment decision paths for each case study:
1.Maintainingan existing wind plant
2.Building a new greenfield site in addition to maintaining the existing plant
3.Repowering an existing site.
We created three case studies (Northeast,West Coast,Midwest),all of which were run with
these three investment decision paths.If we could not find publicly available information for key
inputs,we used industryaverages described in the Wind TechnologyMarket Report 2011 (Wiser
and Bolinger 2012).Modeling was performed with SAM.
4.Have you had a chance to review our preliminaryresults of the case studies file we sent
you on Friday evening?(if no,ask them to review it)
K.What are your initial thoughts?
L.Do these results match what you would expect?Why or why not?
5.In the case study results we highlighted three important input assumptions that drove
repowering's NPV advantage over building a new greenfield site.Please consider each of
these three assumptions and describe whether you feel they are reasonable or not and why
you feel that way.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)30
at www.nrel.gov/publications.
M.5%better wind resource for repowered sites versus greenfield
N.5%-10%lower installation costs due to already existing infrastructure for
repowermg
O.Costs of removal of old equipment can be offset by selling it.
6.Does your company have any employees devotedto assessing the market potential of
repowering?Please describe.
7.What other thoughts do you have on the feasibility of repowering that you would like to
share?
8.What questions would you like answered to help your company make informed decisions
concerning repowering older wind plants?
9.Is there anything else you would like to tell us to inform this project?
10.Are there any other wind energy professionals,either within or outside of your company,
that you would recommend we speak with on this topic?Request specific contact
details.
This report is available at no cost from the
National Renewable Energy Laboratory (NREL)31
at www.nrel.gov/publications.