Reducing Cost of Wind Energy Opportunities to reduce the cost of wind energy 2nd NREL Wind Energy Systems Engineering Workshop Denver, Colorado, 29-30 January 2013

Henk-Jan Kooijman GE Power & Water, Wind Farm Engineering

Reducing cost of wind energy

2

• Levelized cost of energy to study the feasibility of subsidy-free wind energy

• LCOE reduction scenarios for onshore and offshore wind energy

• Topical aspects for turbine load design, site assessment, and farm design

• Summary and conclusions

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

Fantanele-Cogealac, Romania 240 GE 2.5 MW wind turbines; 600 MW farm power

Installation was completed in November 2012

Progress ratio vs. economies of scale

3

Progress ratio: cost reduction with doubling in cumulative volume

Learning rate = 1 – PR

Ex. 1: Unit #100 = 1500 $/kW, turbine PR = 90%

=> Unit #200 = 1350 $/kW

Progress ratio refers to cost reduction over time

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

Economies of scale: lower unit cost for a larger wind farm or turbine

Ex.: 10MW farm = 2000 $/kW, EOS = 90% for doubling in size

=> 20MW farm = 1900 $/kW

Economies of scale refers to cost reduction with size

Progress ratio - turbine availability

4

• Upward trending improvement in turbine availability for consecutive

model year introductions.

• Improved design. Improved services. Continuously resolving top issues.

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

2007 2008 2009 2010 2011 2012 2013

GE 1.5/1.6 MW availability trends

2011

2010

2009

2008

2007

model year introduction

LCOE = levelized cost of energy

5

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

+

• LCOE reduction is importantly driven by economies of scale for BOP and O&M

• Turbine economies of scale doesn’t work well because of square-cube law.

New technologies and relative share of electrical cost can compensate this.

70%

80%

90%

100%

110%

120%

100% 150% 200% 250% 300%

[relative LCOE ]

relative turbine name plate rating (P / Pref ) [-]

Relative LCOE vs. nominal power (constant cap. factor)

130%

120%

110%

100% (baseline)

90%

turbine economies of scale, i.e. relative price per MW with doubling in turbine

New technologies to drive down LCOE

6

Individual new technologies can importantly bring down LCOE, like more use

of condition monitoring to reduce OPEX, distributed turbine control to

improve farm AEP and design loads, or advanced aero to grow rotor size.

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

LCOE benefit [$ct/kWh]

Cumulative effort (risk, time, and investment)

1+ years 2+ years 3+ years

Onshore LCOE reduction scenario

30% reduction in onshore LCOE can be realized through a combination of:

• a 20% reduction in CAPEX, e.g. due to new technologies and economies of scale;

• a 10% higher capacity factor for the same CAPEX per kW;

• and 1pt lower real interest rate.

7

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

0%

20%

40%

60%

80%

100%

-3.0 -2.0 -1.0 0.0

relative change

LCOE change [$ct/kWh]

LCOE reduction entitlement onshore wind

O&M

interest rate

turbine DM

CAPEX

The offshore case • Diverse site characteristics (water depth, facilities, distance to harbour and grid)

• Realization cost ($/kW) roughly twice as high as onshore.

• No visual impact.

• Important levers to reduce LCOE are:

– Economies of scale (go bigger)

– Progress ratio (do better)

– Reduce DM cost (smarter design)

8

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

• Still, additional success criteria are:

– Long-term government committment

– Continuing research funds

– Adequate service hubs and grid infrastructure

• Possible game changing concepts: VAWT or Kites.

Wind turbines,

30%

O&M, 25%

Foundations,

14%Cables

including installation,

8%

Install.

foundations and

substation, 8%

Substation, 7%

Installation turbines, 5%

Other, 5%

Courtesy www.skysails.de

Offshore LCOE reduction entitlements

Maximum of ~50% reduction in offshore LCOE when grouping all entitlements.

9

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

-30%

20%

-28% -80%

-25%

5%

-24% -17%

-10% -8% -5% -5%

-40%

-30%

-20%

-10%

0%

10%

20%

LCOE impact for assumed entitlement

relative entitlement

impact on LCOE

GE 4.1MW in Gothenburg, Sweden

Offshore LCOE reduction scenario

More realistically, 35% reduction in offshore LCOE is realized through a combination of:

• 20% reduction in CAPEX, e.g. through new technologies and economies of scale;

• No change in logistics cost per kW, improved turbine reliability: Availability up 3 pts;

• Net cap factor up by 7 pts, e.g. bigger rotor and less wake losses;

• 50% reduction in contingency and 15% reduction in O&M.

10

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

Arklow: 7 x GE 3.6 MW World’s first 3+ MW

offshore turbines Since 2002

50%

100%

150%

200%

0% 100% 200% 300% 400%

[turbine / BOP cost]

Relative turbine rating [-]

Change in offshore wind cost breakdown with growth in turbine size

high

indicative

Economies of scale:

Topical aspects: turbine design loads

11

• There are many projects in Europe in area-constrained, complex terrains

with a high design turbulence and challenging design load assessment.

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

Example:

Power density [MW/km2] for a

TC-A design relative to TC-B is

equal to: (7D / 5D) 2 factor 2

10%12%14%16%18%20%22%24%

0 1 2 3 4 5 6 7

TI @ 15 m/s

Turbine spacing (rectangular array) [Diameters]

Wind turbine class and TI effective

m = 12 (blade)

m = 4 (tower)

TC-A

TC-B

TC-C

• The global installed wind power, e.g. GE’s 20,000+ unit fleet, offers

valuable potential for experience-based loads analysis.

Topical aspects: site assessment

12

Accurate site design wind conditions are key for accurate load assessment.

For example, GE newly determines Vref based on meso-scale data combined

with site measurements combined with using Bayesian analysis.

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

Source: EWEA 2013 poster #490, author Joerg Winterfeldt et al., GE Power & Water. Data by courtesy of Bonneville Power Administration (BPA)

25

27

29

31

33

35

37

39

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Vref [m/s]

year

Benchmark

Method of Independent Storms Gumbel, least squares

Method of Independent Storms Bayesian LL 29 m/s, UL 34 m/s

Topical aspects: farms

NREL 2nd workshop Jan 2013 13

Henk-Jan Kooijman

Operated as single units in a wind farm

Designed as stand alone unit Accurate

wind farm

flow models Co-operative, ‘distributed’ farm control for max AEP

Better site lay out with diverse turbine options in a farm

Turbine-centered view Holistic farm-level view

• Curtailment losses on a regional scale can be significant,

e.g. strong growth of installed wind base or delayed expansion of the grid.

• Industry is taken on a more wind power plant-centred perspective.

Because only the OEM knows the turbine design limits and aero-elastic

model, it has an essential role in minimizing the project-specific LCOE.

Summary

14

• Offshore wind energy LCOE will remain ~twice the value of onshore wind.

• Economic growth, investors, and green policy agenda like Europe’s

‘Horizon 2020’ are essential for the expansion of wind energy.

• LCOE reduction is importantly driven by:

– Up-scaling turbine and farm size to drive down BOP and O&M over AEP

– Better design modelling to understand and deploy margins for more AEP

– New technologies and operating methods for turbines and farms to:

o Increase AEP and limit turbine cost per MW

o Reach the goal of a purely fatigue-driven turbine:

mitigate extreme loads and reduce blade static moment.

Henk-Jan Kooijman NREL 2nd workshop Jan 2013

Thank You

GE 2.5-100