costs of integration for wind and solar energygy g: large-scale

Costs of Integration for Wind and Solar Energy: Large-scale studies and gy gimplications

MIT Wind Integration Workshop

Michael Milligan, Ph.D.

National Renewable Energy Laboratory

Jan 21, 2011Jan 21, 2011

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Topics

Large-scale integration studies:Western Wind and Solar Integration StudyWestern Wind and Solar Integration Study Eastern Wind Integration and Transmission Study

What does it take to integrate variable generation (wind and solar)?(wind and solar)?

Wind and solar integration cost issuesWind and solar integration cost issues

National Renewable Energy Laboratory Innovation for Our Energy Future

Large-scale wind integration studies: WWSIS and EWITS

• Managed by NREL – large project teams• Western Wind and Solar Integration Study, inWestern Wind and Solar Integration Study, in

collaboration with WestConnect• Eastern Wind Integration and Transmission g

Study, in collaboration with the Joint Coordinated System Plan and Midwest I d d S OIndependent System Operator

National Renewable Energy Laboratory Innovation for Our Energy Future3

WWSIS www.nrel.gov/wwsis

• Large team, Debbie Lew project manager• NREL, GE Energy, 3Tier, Northern ArizonaNREL, GE Energy, 3Tier, Northern Arizona

University, Exeter, SUNY• Project partner: WestConnectj p• Large stakeholder group• Technical Review Committee


Overview

Goal – To understand the costs and

operating impacts due to the variability and uncertainty ofvariability and uncertainty of wind, PV and concentrating solar power (CSP) on the WestConnect grid

UtilitiUtilities– Arizona Public Service– El Paso Electric

NV E– NV Energy– Public Service of New Mexico– Salt River Project

Tri State G&T– Tri-State G&T– Tucson Electric Power– Xcel Energy– WAPA


WAPA

Scenario Overview

In Footprint Rest of WECCWind Solar Wind Solar10% 1% 10% 1%10% 1% 10% 1%20% 3% 10% 1%30% 5% 20% 3%

Baseline – no new renewablesIn-Area – each transmission area meets its target from sources within that areawithin that areaMega Project – concentrated projects in best resource areasLocal Priority – Balance of best resource and in-area sitesPlus other scenarios yet to be determined (high solar, high capacity value, high geographic diversity)Solar is 70% CSP and 30% distributed PV. CSP has 6 hours of thermal storage.


Penetrations are by energy.

6

Study Area Monthly Energy from Wind and Solar for 2004 – 2006 (30% In Area Scenario)

Study Area Total Monthly Wind and Solar Energy for 2004 2006Study Area Total Monthly Wind and Solar Energy for 2004 - 2006

8000

10000

12000

Wh)

05

06

4000

6000

Tota

l Ene

rgy

(GW 04

Study Area Percent Monthly Wind and Solar Energy for 2004 - 2006

0

2000

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Study Area Percent Monthly Wind and Solar Energy for 2004 2006

40%

50%

60%

ergy

55% of energy from wind and

solar

30% is not always

10%

20%

30%

% o

f Loa

d E

ne 30%


0%

10%

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecMonth of Year

Geographic Scenarios(high renewables case)(high renewables case)

In-Area - each state meets target from sources within that statesources within that state.

Mega Project - concentrated projects in best resource areas.


Local Priority - Balance of best resource and In-Area sites.

Wind data

• 3TIER Group ran weather model to recreate weather for 2004, 2005, 2006

• 10 minute wind power output for 960power output for 960 GW of wind sites in WECC. Po er profiles ere• Power profiles were based on VestasV90 3MW turbine at 100 h i ht100m height.

• Hourly day-ahead power output


forecasts.

How does the system operate in the high renewables case?

Mid JulyMid‐July Load

The operator formerly managed to load but now has to manage the net load.

Net Load = Load ‐Wind ‐ Solar


How does the system operate in the high renewables case?

Mid July Mid AprilMid‐July Mid‐April

Mid‐April shows the challenges of


p goperating the grid with 35% wind and solar.

This was the worst week of the 3 years studied.

Operations during mid-April

No Wind/Solar High renewables case


12

Operating Cost Savings perOperating Cost Savings per MWh of Renewable Energy ($/MWh) - WECC - 2006

Perfect forecast cases State of the art forecast cases

86

88

90

80

82

84

s ($

/MW

h)

74

76

78

Savi

ng

70

72

74

Pre P I 10 P I 20 P I 30 P Pre R I 10 R I 20 R I 2020 R I 30 R


EWITS www.nrel.gov/ewits

• Built on Joint Coordinated System Plan• Large project team, Dave Corbus projectLarge project team, Dave Corbus project

manager• NREL, MISO, EnerNex, AWS True Power,

University of Stuttgart, Riso (stochastic model)• Representation from RTOs/ISOs, utilities

T h i l R i C itt• Technical Review Committee


What is Needed to Integrate 20% Wind in the Eastern Interconnection?

• Evaluate the power system operating impacts and transmission associated with increasing windassociated with increasing wind energy to 20% and 30%

• Impacts include operating with the variability andwith the variability and uncertainty of wind

• Build upon prior wind integration studies and related technical work;studies and related technical work;

• Coordinate with current regional power system study work;

P d i f l b dl• Produce meaningful, broadly supported results

• Technical Review Committee


EWITS Analysis Provides Detailed Information on

• Wind generation required to produce 20% and 30% of the projected electric energy demand over the U.S. portion of the Eastern Interconnection in 2024

• Transmission concepts for delivering energy economically for each scenarioeconomically for each scenario

• Economic sensitivity simulations of the hourly operation of the power system with windoperation of the power system with wind generation, future market structures and transmission overlay

• The contribution made by wind generation to resource adequacy and planning capacity margin


Scenario 1: 20% high capacity factor Scenario 2: Hybrid w/off-shore

Scenario 3: 20% local, aggressive offshore Scenario 4: 30% aggressive


Conceptual Transmission Overlays


Transmission - Why are We Always Jumping Through Hoops


Key Task – Wind Integration

• ApproachH l i l ti f ti l l i d• Hourly simulation of operational planning and power system operation using PROMOD• Synchronized wind and loadSynchronized wind and load

• Day-ahead unit commitment and scheduling based on load and windscheduling based on load and wind generation forecasts• Real-time operations (hourly• Real-time operations (hourly

simulations)• Operational structures


• Operational structures

Assumed operational structure for the Eastern Interconnection in 2024 (white circles represent ( pbalancing authorities)


Additional Reserve Requirements by Region and Scenario

7000

6000

7000

uiremen

t

4000

5000

gulation

Req

MW)

Load Only

Reference Case

Scenario 1

2000

3000

e Hou

rly Reg (M

Scenario 1

Scenario 2

Scenario 3

Scenario 4

0

1000

Average

MISO ISO‐NE NYISO PJM SERC SPP TVA

Operating Region

Increased reserve does not imply increased installed or committed capacity over and


Increased reserve does not imply increased installed or committed capacity over and above the no-wind case. It implies some generation that would be used for energy in the no-wind case is now used for reserve (or possibly different mix of units in the stack).

Total Annualized Scenario Costs

180,000

200,000

)

140,000

160,000

,

2009

USD

M$)

WindCapital Cost

80 000

100,000

120,000

nario Co

st (2

Wind Capital Cost

New Generation Capital Cost

Transmission Cost

Integration Cost

40,000

60,000

80,000

nualized

Sce

g

Wind Operational Cost

Production cost

‐

20,000

Reference Scenario 1 Scenario 2 Scenario 3 Scenario 4

Ann


e e e ce Sce a o Sce a o Sce a o3 Sce a o

EWITS Conclusions

• 20 and 30% wind penetrations are technically feasible with significant expansion of the transmission i f t tinfrastructure.– New transmission will be required for all the future wind

scenarios in the Eastern Interconnection

• Without transmission enhancements, substantial curtailment of wind generation will occurInterconnection ide costs for integrating large amo nts• Interconnection-wide costs for integrating large amounts of wind generation are manageable with large regional operating pools, where benefits of load and wind diversity can be exploited and large numbers of supply resources are efficiently committed and dispatched.


EWITS Conclusions

• Transmission helps reduce the impacts of the variability of the wind and….y• Reduces wind integration costs• Reduces need for building conventional generation

I li bilit f th l t i l id• Increases reliability of the electrical grid• Helps make more efficient use of the available generation

resources • Costs for aggressive expansions of the existing grid

are significant, but they make up a relatively small i f th t t l li d t i f thpiece of the total annualized costs in any of the

scenarios studied• Wind generation displaces carbon based fuels


• Wind generation displaces carbon-based fuels, directly reducing carbon dioxide (CO2) emissions

What is needed for large-scale integration g gof wind and solar?


TechnicalTechnical


Physical Sources of Flexibility

Alternative generation mix with more flexibilityy

– Less base-load (or modified for flexibility)

– Aero derivative/fast-response turbines (GE, Siemens)

– Reciprocating engines (Wartsilla)(Wartsilla)

– Pumped storageWind/solar provide regulationResponsive loadElectric vehiclesForecasting

Courtesy: WindLogics, Inc. St. Paul, MN


Forecasting

29

InstitutionalInstitutional


Energy scheduling rules and other institutional factorsinstitutional factors

• Transmission protocols/scheduling inprotocols/scheduling in the Western Interconnection

• Fast energy markets• Ancillary services market

(and possible new AS)(and possible new AS)• Smarter about reserves• Smarter about wind

i

forecasts and how to use/visualize them

www.osei.noaa.gov


Smarter Transmission Usage


Non-spin and Supplemental reserves are 10 to 20 times cheaper than regulation and better match wind ramping characteristicscharacteristics

2002 2003 2004 2005 2006 2007 2008 Annual Average $/MW-hr

California (Reg = up + dn) ( g p )Regulation 26.9 35.5 28.7 35.2 38.5 26.1 33.4

Spin 4.3 6.4 7.9 9.9 8.4 4.5 6.0 Non-Spin 1.8 3.6 4.7 3.2 2.5 2.8 1.3

Replacement 0.90 2.9 2.5 1.9 1.5 2.0 1.4Replacement 0.90 2.9 2.5 1.9 1.5 2.0 1.4ERCOT (Reg = up + dn)

Regulation 16.9 22.6 38.6 25.2 21.4 43.1 Responsive 7.3 8.3 16.6 14.6 12.6 27.2

Non-Spin 3.2 1.9 6.1 4.2 3.0 4.4Non Spin 3.2 1.9 6.1 4.2 3.0 4.4New York

Regulation 18.6 28.3 22.6 39.6 55.7 56.3 59.5 Spin 3.0 4.3 2.4 7.6 8.4 6.8 10.1

Non Spin 1 5 1 0 0 3 1 5 2 3 2 7 3 1Non Spin 1.5 1.0 0.3 1.5 2.3 2.7 3.130 Minute 1.2 1.0 0.3 0.4 0.6 0.9 1.1

New England (Reg +”mileage”) Regulation 54.64 30.22 22.26 12.65 13.75

Spin 0 27 0 41 1 67


National Renewable Energy Laboratory Innovation for Our

Spin 0.27 0.41 1.6710 Minute 0.13 0.34 1.21 30 Minute 0.01 0.09 0.06

Larger Balancing Areas (real or virtual)

500

1000

r)

Operating separate balancing areas causes extra ramping compared to combined operations. Blue: up-rampGreen: down-rampYellow: combined ramp

-500

0

Ram

p (M

W/h

-1000

500

ing

(MW

/hr)

Some areas are ramping up nearly 1000 MW/hr while other areas are ramping down nearly 500 MW/hr

-200

0

200

400

Exc

ess

Ram

p

Ramping that could be eliminated by combining operations

53045298529252865280Hour of Year (one day)

-400

200


Kirby, B.; Milligan, M. (2008). Impact of Balancing Area Size, Obligation Sharing, and Energy Markets on Mitigating Ramping Requirements in Systems with Wind Energy. Wind Engineering. Vol. 32(4), 2008; pp. 399‐414;

Inter-BA Scheduling and Communication

800

600

400

200Gen

erat

ion

(MW

)

One Week

0

Win

d

200

-200

0

ast E

rror

(MW

)

-400

Fore

ca

48.248.047.847.647.4

x103

Wind Energy 30-minute forecast error 10-minute forecast error 2-hour persistence forecast error

Milligan and Kirby (2009) Capacity Requirements to Support Inter Balancing Area Wind Delivery 29 pp ; NREL

806080408020800079807960794079207900

-600


Milligan and Kirby (2009). Capacity Requirements to Support Inter‐Balancing Area Wind Delivery. 29 pp.; NREL Report No. TP‐550‐46274, and An Examination of Capacity and Ramping Impacts of Wind Energy on Power Systems, Kirby & Milligan, ElecJ Aug./Sept. 2008, Vol. 21, Issue 7, pp 30‐42

35

Better use of existing flexibility

• Tap into maneuverable generation that may be “behind the wall”1 4000%

5000%

oad

Nee

d CA ISO PJM WAPA 42,400 55,600 3,087 MW Peak Load Generation 21,900 47,000 2,911 Fossil (measured)behind the wall

• Provide a mechanism (market, contract, other) that benefits system operator and 2000%

3000%

4000%

Ram

ping

Cap

acity

vs

Lo

PJM

13,100 2,500 700 Hydro 4,600 13,500 0 Nuclear 3,700 600 11 Other 2002 Hourly Data and Generation Capacity

WAPAbenefits system operator and generator/loads

• Fast energy markets help provide needed flexibility2 0%

1000%

Mea

sure

d Fo

ssil

R

CA ISO

provide needed flexibility2 and can often supply load following flexibility at no cost3

0 2000 4000 6000 8000

Hours/Year

Generators and loads should be able to participate in these markets equally

1Kirby & Milligan, 2005 Methodology for Examining Control Area Ramping Capabilities with Implications for Wind http://www.nrel.gov/docs/fy05osti/38153.pdf2Kirby &Milligan, 2008 Facilitating Wind Development: The Importance of Electric Industry Structure.

participate in these markets equally


Kirby & Milligan, 2008 Facilitating Wind Development: The Importance of Electric Industry Structure. http://www.nrel.gov/docs/fy08osti/43251.pdf3Milligan & Kirby 2007, Impact of Balancing Areas Size, Obligation Sharing, and Ramping Capability on Wind Integration . http://www.nrel.gov/docs/fy07osti/41809.pdf

Do Markets Incent Flexibility?

Short-run– Ramp products

3200

Energy Price $10 /MW hE nergy Price $10 /MW hE nergy Price Increases

to $90/MW h becausebase unit can't rampp p

necessary to supplement energy markets?

3000

2800

Pea king -$90/MW h

fast en ough

– Low LMPs and capital cost recovery

2600

Long-run– Do prices induce

long-term

2400

4:00 AM 6:00 AM 8:00 AM 10:00 AM 12:00 P M 2:00 PM 4:00 PM 6:00 PM

Base Load - $10 /MW h

long term development of flexible supply/loads?


pp y

37

Integration CostsIntegration Costs


Integration cost of wind and solarC it b d? 100 103

10

• Can it be measured?• If so, how is it defined?• What is proper benchmark

it?

100x103

50

0

-50Del

ta M

arke

t Val

ue ($

)

-5

0

5

Over or U

nder-estim

(Flat block value) - (wind value)

unit?• How are cost and value

untangled?What about units in one

0

20

40

et V

alue

($)

-10

mate of W

ind Value ($/MW

h)

0

5

10

Daily flat energy block ($52.33/MWh) Daily flat block difference $/MWh (right) 6-Hour flat energy block ($48.59/MWh) 6_Hour flat energy block difference $/MWh (right)

(Wind: $48.98/MWh)

• What about units in one region that economically respond to needs in another region?

160140120100806040200

-80x103

-60

-40

-20

Del

ta M

arke

-10

-5

0

g• Are there integration costs for

other units?– Do all AGC units follow the

i l?

Milligan, M.; Kirby, B. (2009). Calculating Wind Integration Costs: Separating Wind Energy Value from Integration Cost Impacts. 28 pp.; NREL Report No. TP‐550‐46275. http://www nrel gov/docs/fy09osti/46275 pdfsignal?

– Are there efficiency costs of adding conventional generators?

http://www.nrel.gov/docs/fy09osti/46275.pdf


Milligan, M.; Ela, E.; Lew, D.; Corbus, D.; Wan, Y. H. (2010). Advancing Wind Integration Study Methodologies: Implications of Higher Levels of Wind. 50 pp.; NREL Report No. CP‐550‐48944. http://www.nrel.gov/docs/fy10osti/48944.pdf

Not all units can follow AGC2 l it h diff t bilit f f ll i AGC2 coal units show very different ability of following AGC.Unit on the right increases the need for regulation on the system by 31 MWsystem by 31 MW

720

740

STED

T (M

W)

ACTUAL440

450

ESTE

DT

(MW

)

ACTUAL

660

680

700

CTU

AL

AN

D R

EQU

EN

ERAT

OR

OU

TPU

T

AGC SIGNAL410

420

430

CTU

AL

AN

D R

EQU

EEN

ERAT

OR

OU

TPU

T

31 MW

97094A

620

640

0 15 30 45 60

TIME (minutes)

AC

GEN

AGC SIGNAL

97094

390

400

0 15 30 45 60

TIME (minutes)A

CG

E

AGC SIGNAL


Milligan, M.; Ela, E.; Lew, D.; Corbus, D.; Wan, Y. H. (2010). Advancing Wind Integration Study Methodologies: Implications of Higher Levels of Wind. 50 pp.; NREL Report No. CP‐550‐48944. http://www.nrel.gov/docs/fy10osti/48944.pdf

Integration cost of wind

• Large nuclear units set the contingency reserve obligation for power pools, resulting in real g gcosts

• When any new generator is added to the generation mix, it potentially impacts all of the units that are above it in the dispatch stackE l h b l d• Example: new cheap baseload– Units formerly run as base load are pushed up the

stack: lower capacity factors higher cyclingstack: lower capacity factors, higher cycling– More cycling higher O&M costs– Lower capacity factor less revenue


p y

Integration costs: wind and solar

Solar and wind integration issues are similar– Wind is becoming reasonably well understood

S l– Solar• PV has high potential for short-term variability from cloud

variations, but the impact of large PV plants is largely unknownunknown

• CSP without storage has some thermal inertia and is likely less variable than PV

• CSP with storage is thought to be much less of an integration• CSP with storage is thought to be much less of an integration challenge but still unknown

Variability and uncertaintyC li ffi iCycling efficiencyAre not unique to wind or solar


Large BAGeographically Dispersed Wind and Solar

Wind/Solar Forecasting Effectively Integrated Into System Operations

Accommodating Wind and Solar Integration

Wind/Solar Forecasting Effectively Integrated Into System OperationsSub-Hourly Energy Markets

Fast Access to Neighboring MarketsNonSpinning and 30 Minute Reserves for Wind/Solar Event Response

Regional Transmission Planning For Economics and Reliabilityg g yRobust Electrical Grid

More Flexible Transmission ServiceFlexibility in Generation

Responsive LoadOverall

Example Utility Structures10 8 7 10 7 2 7 6 7 7 3 7 Large RTO with spot markets

6 6 6 3 3 2 6 4 7 2 2 4 Smaller ISO6 6 6 3 3 2 6 4 7 2 2 4 Smaller ISO

1 3 2 1 2 1 2 3 2 2 2 2 Interior west & upper Midwest (non-MISO)

7 6 6 2 2 2 5 4 2 5 2 4 Large vertically integrated utility

1 3 2 1 2 1 2 4 2 2 2 2 Smaller Vertically Integrated Local Utilityy g y

8 Unconstrained hydro system3 Heavily fish constrained hydro system

1 1 1 1 1 1 1 1 1 1 1 11 Weightings Factors


Adapted from Milligan, M.; Kirby, B.; Gramlich, R.; Goggin, M. (2009). Impact of Electric Industry Structure on High Wind Penetration Potential. 31 pp.; NREL Report No. TP‐550‐46273. http://www.nrel.gov/docs/fy09osti/46273.pdf

Selected NREL Integration Projects

• Wind wind and solar• Scoping EWITS 2, WWSIS 2• Reserves analysis: UCD, MISO• WECC’s Efficient Dispatch Toolkit• Development of variable time-step production

simulation model (includes AGC)Coal cycling (GE WECC)• Coal cycling (GE, WECC)

• Wind Turbine Frequency control (EPRI, GE, others)others)

• Stochastic unit commitment and forecast error characterization


Questions?

45

costs of integration for wind and solar energygy g: large-scale

Documents