AA WindModel

download AA WindModel

of 99

Transcript of AA WindModel

  • 7/31/2019 AA WindModel

    1/99

    Arnold SchwarzeneggerGovernor

    REVIEW OF INTERNATIONALEXPERIENCE INTEGRATING VARIABLE

    RENEWABLE ENERGY GENERATION

    Prepared For:

    California Energy CommissionPublic Interest Energy Research Program

    PIERP

    ROJECTREPORT

    Prepared By:EXETER

    ASSOCIATES, INCApril 2007CEC-500-2007-029

  • 7/31/2019 AA WindModel

    2/99

    Prepared By:

    Exeter Associates, Inc.Kevin PorterColumbia, MarylandCommission Contract No. 500-02-004Commission Work Authorization No: MR-017

    Prepared For:Public Interest Energy Research (PIER) Program

    California Energy Commission

    Michael Kane, Dora Yen-Nakafuji, Ph.D.

    Contract Manager

    Dora Yen-Nakafuji, Ph.D.

    Project Manager

    Elaine Sison-Lebrilla, P.E.

    Manager

    Energy Generation Research Office

    Martha Krebs, Ph.D.

    Deputy Director

    ENERGY RESEARCH & DEVELOPMENTDIVISION

    B.B. Blevins

    Executive Director

    DISCLAIMERThis report was prepared as the result of work sponsored by the California Energy Commission. It does not necessarily representthe views of the Energy Commission, its employees or the State of California. The Energy Commission, the State of California, itsemployees, contractors and subcontractors make no warrant, express or implied, and assume no legal liability for the informationin this report; nor does any party represent that the uses of this information will not infringe upon privately owned rights. Thisreport has not been approved or disapproved by the California Energy Commission nor has the California Energy Commissionpassed upon the accuracy or adequacy of the information in this report.

  • 7/31/2019 AA WindModel

    3/99

    i

    Acknowledgments

    TheCaliforniaEnergyCommissionsPublicInterestEnergyResearchprogramfundedthework

    describedinthereport.TheauthorsthankDoraYenNakafujiandtheCaliforniaWindEnergy

    Collaborativeteamfortheirtechnicalsupport.TheauthorsalsothankThomasAckermanofthe

    RoyalInstitute

    of

    Technology

    in

    Sweden;

    Brendan

    Kirby

    of

    Oak

    Ridge

    National

    Laboratory;

    BrianParsonsandMichaelMilliganoftheNationalRenewableEnergyLaboratory;Jim

    BlatchfordandDavidHawkinsoftheCaliforniaIndependentSystemOperator;J.CharlesSmith

    oftheUtilityWindIntegrationGroup;HanneleHolttinenoftheVTTTechnicalResearchCenter

    inFinland;BernhardErnstoftheRheinischWestflischesElektrizittswerkAktiengesellschaft

    (RWE)TransmissionSystemOperatorinGermany;AlbertoCenaofAsociacinEmpresarial

    Elica(AEE)inSpain;LucyCraigofGarradHassaninSpain;DaveOlsenofWestWindWires;

    MarkAhlstromofWindLogicsInc.;TomMillerofPacificGasandElectric;AbrahamEllisof

    PublicServiceCompanyofNewMexico;andJohnKehleroftheAlbertaElectricSystem

    Operatorforansweringnumerousquestionsandprovidingusefulinsights.Anyremaining

    errors

    or

    omissions

    are

    our

    own.

    Pleasecitethisreportasfollows:

    KevinPorter,ChristinaMuddandMichelleWeisburger.2007.ReviewofInternationalExperience

    IntegratingVariableRenewableEnergyGeneration.CaliforniaEnergyCommission,PIER

    RenewableEnergyTechnologiesProgram.CEC5002007029.

  • 7/31/2019 AA WindModel

    4/99

    ii

    Preface

    ThePublicInterestEnergyResearch(PIER)Programsupportspublicinterestenergyresearch

    anddevelopmentthatwillhelpimprovethequalityoflifeinCaliforniabybringing

    environmentallysafe,affordable,andreliableenergyservicesandproductstothemarketplace.

    ThePIERProgram,managedbytheCaliforniaEnergyCommission(EnergyCommission),

    conductspublicinterestresearch,development,anddemonstration(RD&D)projectstobenefit

    theelectricityandnaturalgasratepayersinCalifornia.

    ThePIERprogramstrivestoconductthemostpromisingpublicinterestenergyresearchby

    partneringwithRD&Dorganizations,includingindividuals,businesses,utilities,andpublicor

    privateresearchinstitutions.

    PIERfundingeffortsarefocusedonthefollowingRD&Dprogramareas:

    BuildingsEndUseEnergyEfficiency

    EnergyInnovations

    Small

    Grants

    EnergyRelatedEnvironmentalResearch

    EnergySystemsIntegration

    EnvironmentallyPreferredAdvancedGeneration

    Industrial/Agricultural/WaterEndUseEnergyEfficiency

    RenewableEnergyTechnologies

    Transportation

    ReviewofInternationalExperienceIntegratingVariableRenewableEnergyGenerationisthefinal

    reportfor

    asubtask

    of

    Task

    3for

    the

    PIER

    Intermittency

    Analysis

    Project

    (IAP),

    contract

    number50002004,workauthorizationnumberMR017,conductedbytheIAPteamcomprised

    oftheCaliforniaWindEnergyCollaborative,ExeterAssociates,BEWEngineering,DavisPower

    Consulting,andGEEnergyConsulting(withassistancefromAWSTruewind,National

    RenewableEnergyLaboratory(NREL),OakRidgeNationalLaboratory(ORNL),andRumla

    Consulting).TheinformationfromthisprojectcontributestoPIERsRenewableEnergy

    Technologiesprogram.

    FormoreinformationonthePIERProgram,pleasevisittheEnergyCommissionswebsiteat

    www.energy.ca.gov/pierorcontacttheEnergyCommissionat(916)6545164.

  • 7/31/2019 AA WindModel

    5/99

    iii

    Table of ContentsAcknowledgements .................................................................................................................................... i

    Preface..........................................................................................................................................................ii

    ListofTables...............................................................................................................................................v

    ListofFigures ............................................................................................................................................ vi

    Abstract .....................................................................................................................................................vii

    ExecutiveSummary................................................................................................................................... 1

    1.0 Introduction ................................................................................................................................. 19

    1.1. WorldwideWindandSolarCapacity................................................................................... 22

    2.0 WindIntegrationStudiesintheUnitedStatesandWorldwide .......................................... 27

    2.1 SummaryofVariousAssessmentsoftheImpactsofWindonReserves ........................ 29

    2.2 SummaryofEstimatedCostImpactsforAdditionalReservesfromWindEnergy....... 31

    2.3 UnitCommitmentImpacts..................................................................................................... 36

    2.4 WindandNaturalGasStorage.............................................................................................. 37

    2.5 ChangestoReserveService.................................................................................................... 37

    2.6 Implications

    for

    California ..................................................................................................... 383.0 MarketStructureandCapacityCredit..................................................................................... 39

    3.1 MarketSchedulingandBalancingRequirements............................................................... 39

    3.2 ResourceDelivery(CapacityCredit) .................................................................................... 40

    3.3 ImplicationsforCalifornia ..................................................................................................... 43

    4.0 OperationalIssuestoDate......................................................................................................... 45

    4.1 MinimumLoad ........................................................................................................................ 45

    4.2 Ramping.................................................................................................................................... 46

    4.3 TransmissionRatingandGenerationOverflow ................................................................. 51

    5.0 MitigationandOperatingSolutionsToDate..........................................................................53

    5.1 WindForecasting ..................................................................................................................... 53

    5.2 GridCodes ................................................................................................................................ 59

    5.2 WindTurbineModelingandVerification............................................................................ 64

    5.4 DemandResponse ................................................................................................................... 67

    5.5 Storage....................................................................................................................................... 67

    5.6 WindPowerCurtailment ....................................................................................................... 68

    5.7 TransmissionPlanningandDevelopment........................................................................... 70

    6.0FindingsandImplicationsforCalifornia ....................................................................................... 73

    6.1 AncillaryServices .................................................................................................................... 73

    6.2 WindForecasting ..................................................................................................................... 74

    6.3 Transmission ............................................................................................................................ 746.4 ActiveManagementofWindGeneration ............................................................................ 75

    6.5 FlexibleGeneration.................................................................................................................. 75

    6.6 Storage....................................................................................................................................... 76

    6.7 DemandResponse ................................................................................................................... 76

    7.0 Conclusion ................................................................................................................................... 77

    7.1 BenefitstoCalifornia............................................................................................................... 79

    References......................................................................................................................................81

  • 7/31/2019 AA WindModel

    6/99

    iv

    Appendix A Review of International Experience Integrating Variable Renewable EnergyGeneration. Appendix A: Denmark

    Appendix B Review of International Experience Integrating Variable Renewable EnergyGeneration. Appendix B: Germany

    Appendix C Review of International Experience Integrating Variable Renewable EnergyGeneration. Appendix C: India

    Appendix D Review of International Experience Integrating Variable Renewable Energy

    Generation. Appendix D: Spain

  • 7/31/2019 AA WindModel

    7/99

    v

    List of Tables

    Table ES-1. Examples of wind power penetration levels, 2005..................................................... 2

    Table ES-2. Reserve definitions in Germany, Ireland and the United States ................................. 3

    Table ES-3: Estimated ancillary service costs from various wind integration studies inthe United States ..................................................................................................................... 6

    Table ES-4. Examples of wind capacity credit methods in the United States................................ 9

    Table ES-5. Examples of wind grid codes.................................................................................... 12Table 1. Examples of wind power penetration levels, 2005......................................................... 20

    Table 2. Global wind energy capacity by country, 2006 .............................................................. 23

    Table 3. Twenty largest grid-connected photovoltaic systems..................................................... 25Table 4. Reserve definitions in Germany, Ireland, and the United States.................................... 28

    Table 5. Estimated ancillary service costs from various wind integration studies in the United

    States..................................................................................................................................... 33

    Table 6. Estimated financial impacts on the Public Service Company of Colorados gas supply

    due to wind generation variability and uncertainty............................................................... 37Table 7. Market closing times in various electricity markets ....................................................... 39

    Table 8. Factors positively and negatively affecting the capacity credit of wind power.............. 41Table 9. Examples of wind capacity credit methods in the United States.................................... 43

    Table 10. Estimated capacity credit of various renewable energy technologies as compared to a

    medium-sized gas plant......................................................................................................... 44Table 11. Overview of operational short-term wind power forecast models in Europe............... 54

    Table 12. Examples of wind grid codes........................................................................................ 60

    Table 13. Power control requirements for wind turbines ............................................................ 62

    Table 14. Summary of performance tests and results for the Woolnorth Wind Farm.................. 66

  • 7/31/2019 AA WindModel

    8/99

    vi

    List of Figures

    Figure ES-1. Range of findings of additional reserve costs from wind generators ...................... 4

    Figure ES-2. Estimated increase in reserve requirements from wind from various studies in

    Europe..................................................................................................................................... 5Figure ES-3. Capacity credit values................................................................................................ 8

    Figure ES-4. Frequency control requirements by selected country.............................................. 13Figure 1: Worldwide PV installations in 2005 (MW) .................................................................. 24

    Figure 2. Range of findings of additional reserve costs from wind generators .......................... 32

    Figure 3. Estimated increase in reserve requirements from wind from various studies in Europe

    ............................................................................................................................................... 34Figure 4: Capacity credit values ................................................................................................... 42

    Figure 5: Simulated hourly wind generation changes in New York, 200103............................. 48

    Figure 6: Estimated total wind ramping requirements in California 2002 ................................... 50Figure 7: Estimated solar ramping requirements in California - 2002 ......................................... 51

    Figure 8: Frequency control requirements by selected country.................................................... 63Figure 9: Proposed transmission projects in the West.................................................................. 72

  • 7/31/2019 AA WindModel

    9/99

    vii

    Abstract

    ThisreportsummarizestheexperienceintheUnitedStatesandinternationallythrough2006

    withintegratingvariablerenewableenergygeneration,primarilywindgeneration,and

    discussespotentialoperatingandmitigationstrategiesforincorporatingvariablerenewable

    energygeneration.

    Initially,

    wind

    development

    in

    Europe,

    particularly

    in

    Denmark

    and

    Germany,consistedofsmallerbutnumerouswindprojectsinterconnectedtothedistribution

    grid,incontrastwithlarger,utilityscalewindprojectsinterconnectedtothetransmissiongrid

    intheUnitedStates.ThedifferencesbetweenEuropeandtheUnitedStatesarestartingto

    narrowasdevelopmentofvariablerenewableenergygeneration(e.g.windandsolar)increases

    andaswinddevelopmenttakesplaceinmorecountries.Inaddition,asmoreutilityscalewind

    projectsemerge,morecountriesarerelyingoncommonstrategies,suchasgridcodes,tohelp

    integratevariablerenewableenergygeneration.ThisreportisapartoftheIntermittency

    AnalysisProject(IAP),acomprehensiveprojectaimedatassessingtheimpactofincreasing

    penetrationofvariablerenewableenergygenerationinCalifornia.Areviewoftheinternational

    experience

    will

    provide

    perspective

    and

    insight

    to

    the

    IAP

    analysis

    team

    on

    various

    techniques

    formanagingintermittency.

    Keywords:windintegration,solarvariability,windforecasting,variablerenewableenergy

    generation,windforecasting,transmission,VARsupport,reserves,ramprates,gridcode,

    ancillaryservices.

  • 7/31/2019 AA WindModel

    10/99

    viii

  • 7/31/2019 AA WindModel

    11/99

    1

    Executive Summary

    IntroductionCaliforniasrenewablepolicytargetsof20percentrenewableenergyby2010and33percentby

    2020are

    likely

    to

    be

    met

    with

    significant

    amounts

    of

    variable

    renewable

    energy

    generating

    resourcessuchaswindandsolarpower.Theanticipatedgrowthintheserenewablesourcesis

    challengingdecisionmakerstolookathowtheCaliforniagridwillaccommodatethese

    resources.Someanswersarefoundbyexamininginternationalexperience,wherewind

    developmenthasbeengrowingsteadilyforseveralyears,andsolargeneratingcapacityis

    accelerating.Bytheendof2006,over74gigawatts(GW)ofwindpowercapacityhasbeen

    installedworldwide,withtwothirdsofthatinEurope.Bytheendof2005,aboutfiveGWof

    gridconnectedsolarpowerisinstalledworldwide,withoverhalfofthatcapacitylocatedin

    Germany.

    PurposeAlthoughtherearenumerousstudiesestimatingpotentialwindintegrationcoststhatrelyon

    modelsandpowersimulations,thereislittleinformationthatprovidesactualexperiencewith

    increasinglevelsofvariablerenewableenergygeneration.Thisreportwilldiscussresultsfrom

    bothactualexperienceandstudiesthatrelyonmodelsandsimulations,andwillattemptto

    distinguishbetweenthosetwothroughoutthedocument. Thisreportispartofthe

    IntermittencyAnalysisProject(IAP)andisfundedbytheCaliforniaEnergyCommissions

    PublicInterestEnergyResearch(PIER)Program.TheIAPisacomprehensiveanalysisproject

    aimedatassessingtheimpactofincreasingpenetrationofvariablerenewableenergygeneration

    inCalifornia.Areviewoftheinternationalexperiencewillprovideperspectiveandinsightto

    theIAPanalysisteamonvarioustechniquesformanagingintermittency.TheIAPwillmodel

    fourscenariosofincreasinglevelsofvariablerenewableenergygeneratingresources,and

    assessthepotentialgridimpactsandproposemarketandoperationstrategiestomitigate

    impacts,ifanyareidentified.

    MarketPenetrationWorldwidewindcapacityismorethan74GWbytheendof2006,withEuropeaccountingfor

    twothirdsofthatcapacity.Germanyhasthemostinstalledwindcapacitywithover20GW,

    followedbySpain(11GW),theUnitedStates(11GW),India(6GW)andDenmark(3GW).Byenergycontribution,Denmarkistheworldleader,withover18percentofitsenergycomingfromwind.Someregionswithincountrieshaveevengreaterpenetrationsofwindpower,as

    indicatedin

    Table

    ES

    1.

    Germanyaccountsformorethanhalfoftheworldsinstalledsolarcapacity,withtheUnited

    StatesandJapanthenextleadingcountries.Thereislessgridexperiencewithsolarcapacityas

    thereiswithwindpower,inpartbecauselargergridconnectedsolarfacilitiesarejustnow

    comingonline.Ofthe20largestsolarfacilitiesintheworld,onlyfourwereinstalledbefore

    2004.Forthatreason,thisreportwillmostlyfocusonwindpower.

  • 7/31/2019 AA WindModel

    12/99

    2

    Table ES-1. Examples of wind power penetration levels, 2005

    Country or region Installed wind

    capacity

    (MW)

    Total installed

    power capacity

    (MW)

    Average

    annual

    penetration

    levela(%)

    Peak

    penetration

    levelb

    (%)

    Western Denmark 3,128 7,488 ~23 >100Germany: 18,428 124,268 ~5 n.a.

    Schleswig-Holstein 2,275 _________c ~28 >100

    Spain 10,028 69,428 ~8 ~25%

    Island systems:

    Swedish island of

    Gotlandd

    90 No local generation

    in normal state

    ~22 >100

    n.a. = Not availableaWind energy production as share of system consumption

    bLevel at high wind production and low energy demand, hence, if peak penetration level is >100%

    excess energy is exported to other regions.cGerman coastal province

    d2002 data. The island of Gotland has a network connection to the Swedish mainland.

    Source: Adapted from Soder, Lennart and Ackerman, Thomas (2005). Wind Power in Power Systems: AnIntroduction, In T. Ackerman (Ed.), Wind Power in Power Systems(pp. 25-51). England: John Wiley andSons, Ltd. Updated and adapted by the author. Reproduced with permission.

    MarketOperationsEuropeusesdifferentterminologyindescribingtheancillaryservicesnecessarytomaintain

    gridreliabilitythantheUnitedStates(TableES2).InEurope,primaryreservesassistwiththe

    shortterm,

    minute

    to

    minute

    balancing

    and

    control

    of

    the

    power

    system

    frequency,

    and

    is

    equivalentintheUnitedStatestoregulation.SecondaryreservesinEuropetakeoverfor

    primaryreserves10to30minuteslater,freeingupcapacitytobeusedasprimaryreserves.

    LongertermreservesinEuropearecalledtertiaryreservesandareavailableintheperiodsafter

    secondaryreserves.Sincewearefocusedoninternationalexperiencewithintegratingvariable

    renewableenergygeneration,wewillusethetermsprimaryandsecondaryreservesforthis

    report.

    Todate,gridreliabilityhasbeenmaintainedaswindandsolarcapacityhasbeenincorporated.

    Thelargestimpactofwindappearstobeonsecondaryreserves.Windhashadlittleeffecton

    primaryreserves,asthevariationsinwindpowerarerandom.Whenaggregatedwithloadand

    generationvariations,

    the

    variations

    from

    wind

    power

    tend

    to

    be

    small

    or

    cancel

    each

    other

    out.

    Sofar,Denmark,GermanyandSpainhavenotchangedtheamountofprimaryreserves

    requiredtomaintainsystemreliability,andwindintegrationstudiesconductedinGermany

    andtheUnitedStudieshavealsofoundthatonlysmallamountsofadditionalregulating

    reservesarerequired.

  • 7/31/2019 AA WindModel

    13/99

    3

    Table ES-2. Reserve definitions in Germany, Ireland and the United StatesShort-term

    reserves

    Medium-term

    Reserves

    Long-term

    reserves

    Germany Primary reserve:

    available within 30

    seconds, releasedby transmission

    system operator

    Secondary

    reserve: available

    within 5 minutes,released by

    transmission

    system operator

    Minute reserve:

    available within

    15 minutes,called by

    transmission

    system operator

    from supplier

    n/a

    Ireland Primary operating

    reserve: available

    within 15 seconds

    (inertial response/

    fast response)

    Secondary

    operating reserve:

    operates over

    timeframe of 15-

    90 seconds

    Tertiary

    response: from

    90 seconds

    onwards

    (dynamic or static

    reserve)

    n/a

    United States Regulation horizon:

    1 minute to 1 hour

    with 1- to 5-second

    Load-following horizons: 1 hour within

    increments 5- to 10 -minute

    increments (intra-hour) and several

    hours (inter-hour)

    Unit-

    commitment

    horizon: 1 day to

    1 week with 1-

    hour time

    increments

    Source: Gul, T. and Stenzel, T. 2005. Variability of Wind Power and Other Renewables: ManagementOptions and Strategies. Paris: International Energy Agency

    Includingbothprimaryandsecondaryreservecosts,itappearsthatthecostofintegratingwind

    isless

    than

    $6/MWh

    at

    energy

    penetration

    levels

    of

    up

    to

    20

    percent

    (Figure

    ES

    1).

    Caution

    shouldbeusedininterpretingFigureES1,asthestudiesemploydifferentmethodologies,data,

    timescales,andtools.Forexample,theE.OnNetzdatainFigureES1measuresreserve

    impactsofwindonadayaheadbasis,whileotherstudiesmeasurereserveimpactsduringthe

    hour;theresultsillustratethatwindcannotbeforecastedasaccuratelyonadayaheadbasisas

    onetotwohoursahead.

    Factorsthataffectwindintegrationcostsinclude:

    Howthevariabilityinwindgenerationinteractswiththevariabilityinelectricity

    demand

    Thegeographic

    concentration

    of

    wind

    projects

    Howfarinadvancethepowerschedulesmustbesubmittedtosystemoperators.

  • 7/31/2019 AA WindModel

    14/99

    4

    Figure ES-1. Range of findings of additional reserve costs from wind generatorsSource: Adapted from Gross, Robert; Heptonstall, Philip; Anderson, Dennis; Green, Tim; Leach, Matthew;and Skea, Jim. (2006). The Costs and Impacts of Intermittency. London: United Kingdom Energy ResearchCenter. Available at http://www.ukerc.ac.uk/content/view/258/852. British currency converted to U.S. $ usinga conversion of $1.8717 per British pound, as of May 25, 2006. Denmark 2002 from Ackerman, Thomas;Morthorst, Poul Erik. 2005. Economic Aspects of Wind Power in Power Systems. In T. Ackerman (Ed.),Wind Power in PowerSystems (pp. 384-410). England: John Wiley and Sons, Ltd. National Grid numbersfrom National Grid Transco. 2004. Submission to the Enterprise and Culture Committee: Renewable Energyin Scotland Inquiry. Available at www.scottish.parliament.uk.. Sustainable Energy numbers from SustainableEnergy Ireland. 2004. Operating Reserve Requirements as Wind Power Penetration Increases in the IrishElectricity System. Available at http://www.sei.ie/uploadedfiles/InfoCentre/IlexWindReserrev2FSFinal.pdf.See Reference for details.

    Country Comments Reference

    1 UKLower bound estimates based on analysis from NEMCO (Australia), Lewis Dale of National

    Grid, SCAR Study and Millsborrow 2002

    Mott MacDonald,

    2003.

    2 Nordic Based on data collected in Finland, Sweden, Norway and Denmark Holttinen, 2004.

    3 UK Dale, Milborrow SCAR, PIU studies Dale et al 2003.

    4 UK Based on modeling efforts Ilex & Strbac, 2002.

    5 Ireland Numbers derived from analysis of international experience, specifically, Denmark, US (BPA) Millborrow, 2004.

    6 Ireland Study conducted for Sustainable Energy Ireland, estimates based on modeling analysis Ilex et al, 2004.

    7 Denmark Actual costs to Eltra, Danish grid operator Pedersen et al, 2002

    8 UK Estimates based on the technical standards of the National Grid Company Milborrow, 2001a

    9a Spain Low market costs of procuring the difference between predicted and actual generation Fabbri et al, 2005.

    9b Spain High market costs of procuring the difference between predicted and actual generation Fabbri et al, 2005.

    10 UK Estimates based on 2001 market data for imbalances Dale, 2002

    11 Germany Figures derived from analysis of E.On Netz study Milborrow, 2005a

    12a Denmark Low estimate based on Nord Pool balancing market (2002 prices Ackerman et al, 2005

    12b Denmark High estimate based on Nord Pool balancing market (2002 prices) Ackerman et al, 2005

    13a ScotlandNational Grid estimates for balancing costs with 10 % penetration of wind in the UK, asreported to the Scottish Parliament

    National GridTransco, 2004

    13b ScotlandNational Grid estimates for balancing costs with 20 % penetration of wind in the UK, as

    reported to the Scottish Parliament

    National Grid

    Transco, 2004

    1

    2

    34

    5

    6

    8

    7

    9a

    9b

    10

    11

    12a

    12b13a

    13b

    0

    2

    4

    6

    8

    10

    12

    14

    16

    0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45

    Intermittent generation level penetration level (% of total system energy)

    Reservecost

    ($/MWh)

    1

    2

    3

    4

    5

    6

    7

    8

    9a

    9b

    10

    11

    12a

    12b

    13a

    13b

  • 7/31/2019 AA WindModel

    15/99

    5

    Submittingschedulesclosertotherealtimemarketwillallowformoreaccuratepredictionsof

    windgeneration,althoughsometradeoffsareinvolved.Havingashorterperiodoftimebefore

    thestartofrealtimemarketoperationsmayleadtoaneedformoreflexibleoperatingreserves,

    orperhapshighercostsfromtheincreasedstartingandstoppingofconventionalunits.The

    shorterperiodsoftimemaynotallowsufficienttimetochangeunitcommitmentdecisionsfor

    conventionalgenerating

    units.

    This

    problem

    can

    be

    simply

    addressed

    with

    awind

    plant

    scheduleupdate.

    FigureES2illustratestheestimatedpercentageincreaseinreservesfromwindfromseveral

    windintegrationstudiesinEurope.Themethodologydifferssignificantlybystudy,making

    theseresultsnotdirectlycomparable.Forexample,thedenastudyinGermanyestimated

    reserverequirementsonadayaheadbasis,whiletheUnitedKingdomandSwedenstudies

    estimatedreserverequirementsfourhoursahead.Theotherstudiesestimatedtheimpacton

    reservesfromwindvariabilityduringtheoperatinghour.Generally,FigureES2suggeststhat

    anincreaseinreservesislikelywithhigherlevelsofwindpenetration.

    Figure ES-2. Estimated increase in reserve requirements from windfrom various studies in EuropeSource: Holttinen, Hannele, Pete Meibom, Antje Orths, Frans Van Hulle, Cornel Ensslin,Lutz Hofmann, John McCann, Jan Pierik, John Olav Tande, Ana Estanqueiro, LennartSoder, Goran Strbac, Brian Parsons, J. Charles Smith and Bettina Lemstrom. Design andOperation of Power Systems with Large Amounts of Wind Power: First Results ofInternational Energy Agency Collaboration. Global Wind Power Conference, Adelaide,Australia. September 18-21, 2006.http://www.ieawind.org/AnnexXXV/Meetings/Oklahoma/IEA%20SysOp%20GWPC2006%20paper_final.pdf. (accessed November 8, 2006).

    WindintegrationstudiesconductedintheUnitedStateshaveoftenfocusedonunitcommitment,thetimeframewheregeneratorsarecommittedinadvancetomeetexpected

    demand(TableES3).Thisiswhereimprovementsinwindforecastingarelikelytohavethe

    greatestimpact.Ingeneral,theEuropeanstudiesdidnotfocusasmuchonunitcommitment

    issues.

    Increase in reserve requirement

    0 %

    1 %

    2 %

    3 %

    4 %

    5 %

    6 %

    7 %

    8 %

    9 %

    10 %

    0 % 5 % 10 % 15 % 20 % 25 %

    Wind penetration (% of gross demand)

    Increaseas%o

    fwindcapacit Nordel: SE, NO, FI, DK

    Finland

    Sweden

    Ireland

    UK

    Sweden 4 hours ahead

    dena Germany

  • 7/31/2019 AA WindModel

    16/99

    6

    Table ES-3: Estimated ancillary service costs from various wind integrationstudies in the United States

    Study Wind

    Penetration

    (%)

    Regulation

    $/MWh

    Load

    Following

    $/MWh

    Unit

    Commitment

    $/MWh

    Gas

    Supply

    Cost

    ($/MWh)

    Total

    $/MWh

    UWIG/Xcel 3.5 0 0.41 1.44 NA 1.85

    PacifiCorp 20 0 1.64 3.00 NA 4.64

    BPA/Hirst 7 0.19 0.28 1.00-1.80 NA 1.47-2.27

    PJM/Hirst 0.06-0.12 0.05-0.30 0.70-2.80 N/A NA 0.75-3.10

    We

    Energies I

    4 1.12 0.09 0.69 NA 1.90

    We

    Energies II

    29 1.02 0.15 1.75 NA 2.92

    Great River

    Energy I

    4.3 NA NA NA NA 3.19

    Great River

    Energy II

    16.6 NA NA NA NA 4.53

    CA RPS

    Phase III

    4 0.46 NA NA NA NA

    MN

    DOC/Xcel

    15 0.23 0 4.37 NA 4.60

    Xcel-PSCo 10 0.20 NA 3.32 1.26 3.72

    Xcel-PSCo 15 0.20 NA 3.32 1.45 4.97

    Sources: Parsons, Brian, et al: Grid Impacts on Wind Power Variability: Recent Assessments from aVariety of Utilities in the United States. Paper given to Nordic Wind Power Conference, May 22-23, 2006,Finland; and Smith, J.C.; DeMeo, E.; Parsons, B.; and Milligan, M. Wind Power Impacts on Electric-Power-

    System Operating Costs: Summary and Perspective on Work to Date. March 2004. Presented to theAmerican Wind Energy Conference, Chicago, Illinois. www.nrel.gov/docs/fy04osti/35946.pdf. (accessedJune 2, 2006).

    AlthoughpresentoperatingpracticesinEuropehavesuccessfullyintegratedwindpower,

    currentinitiativesindicatethatchangesmaybenecessaryasmorewindpowercomesonline.

    Amongotherinitiatives:

    TheEuropeanTransmissionSystemOperators(TSO),theassociationoftransmission

    systemoperatorsinEurope,isconductingaEuropewidewindintegrationstudy,with

    resultsdueby2008.

    TheInternational

    Energy

    Agency

    (IEA)

    is

    sponsoring

    an

    annex,

    Design

    and

    Operation

    ofPowerSystemswithLargeAmountsofWindPowerProduction,thatbeganin

    mid2006.

    InAsia,thesituationisdifferentinChinaandIndia,asthelackofgridinfrastructureseverely

    handicapsnotonlywinddevelopmentandoperationsbutalsotheeconomyasawholeinboth

    countries.

  • 7/31/2019 AA WindModel

    17/99

    7

    CapacityCreditofWindAreviewofvariousstudiesestimatingthecapacitycreditofwindpowerinEuropeindicated

    thatwindhasacapacitycreditgreaterthanzero,andalsothatthecapacitycreditdecreasesas

    thelevelofwindgenerationrises.ThesefindingsareillustratedinFigureES3.Capacitycredit

    studies

    for

    wind

    in

    the

    United

    States

    have

    not

    generally

    measured

    the

    capacity

    credit

    of

    wind

    versusthemarketpenetrationofwind.Instead,thesestudieshavefocusedmoreonthe

    methodsandmechanicsofdeterminingthecapacitycreditforwind.Avarietyofapproaches

    havebeenusedintheUnitedStatesfordeterminingthecapacitycreditofwind,rangingfrom

    determiningtheequivalentloadcarryingcapabilityofwind;usingaproxyvalue;applyingthe

    capacityfactorofwindduringpeakdemandhours;andusingthecapacityvalueofwind

    duringafractionofthetoppeakdemandhours(TableES4).

    AswithFigureES1,cautionshouldbeusedininterpretingFigureES3andTableES4,as

    differentstudymethodologies,assumptionsanddatawereusedinseveralofthesestudies.

    OperatingIssuestoDateMinimumLoad:Definedsimply,minimumloadisthesmallestamountofloadonthegrid

    duringadefinedperiodoftime.Windproductionmaycoincidewithtimesofminimumload

    andaddtosystemchallengesinmanagingthegrid.

    WindintegrationinDenmarkandGermanyhasbeeneasedconsiderablybytheextensive

    interconnectionsthetwocountrieshavewithneighboringcountries.AttimesinDenmark,

    hourlywindproductioncanexceedloaddemand,andconventionalpowerplantshaveto

    reducetheirproductionuntilthesupplyanddemandbalanceisrestored.Ontheseoccasions,

    spotpricesmaydroptozero,asoccurredfor83hoursinDenmarkin2003.GeneralElectrics

    windintegrationstudyfortheNewYorkStateEnergyResearchandDevelopmentAuthority

    (NYSERDA)found

    that

    minimum

    load

    is

    not

    asignificant

    issue

    with

    10

    percent

    wind

    penetration,asNewYorkisanenergyimporterwithoutwindandremainsanimporterwith

    wind.

    Californiahasthepotentialforminimumloadissues.Theseissuesinclude:

    MustrunqualifyingfacilitycontractsunderthePublicUtilityRegulatoryPoliciesAct.

    Increasedprocurementofcombinedcyclenaturalgasprojectsthatoperatebaseloadand

    aroundtheclock.1

    1Anotherpotentialneartermcontributortominimumloadissuesisthearoundtheclockenergy

    procurementcontractsthattheCaliforniaDepartmentofWaterResourcessignedduringtheelectricity

    crisisof2000and2001.However,thesecontractsexpirebetween2009and2011,likelybeforevariable

    renewablesmayreachhighlevelsofmarketpenetrationinCalifornia.

  • 7/31/2019 AA WindModel

    18/99

    8

    Country Comments Reference

    1 Ireland Estimate of capacity credit values for an island system Watson 2001

    2 UK

    Estimates based on analysis from a three different

    sources, Central Electricity Generating Board, National

    Grid, and System Costs of Additional Renewables

    (SCAR Report)

    Mott

    MacDonald

    2003

    3 Germany Dena project steering group Dena 2005

    4 UKExamines the CEGB and SCAR reports and adjusts

    them for greater penetrations of wind

    Dale, et al.,

    2003

    5 UK Based on modelingIlex and Strbac,

    2002

    6 N. EuropeEstimates based on reanalysis data collected from

    operating wind facilitiesGiebel, 2000

    7 UK Early assessment of capacity of wind projects in the UK Grubb 1991

    8 Germany E. On NetzE. On Netz

    2005

    9 UK Study Commissioned by UK Government Sinden 2005

    Figure ES-3. Capacity credit valuesSource: Adapted from Gross, Robert; Heptonstall, Philip; Anderson, Dennis; Green, Tim; Leach,Matthew; and Skea, Jim. (2006). The Costs and Impacts of Intermittency. London: United KingdomEnergy Research Center. Available at http://www.ukerc.ac.uk/content/view/258/852. See Reference fordetails.

    TheCaliforniaIndependentSystemOperator(CAISO)notedthatminimumloadconditionscan

    beexacerbatedinAprilandMaywhenhydroelectricitygeneration,consideredmusttake,

    increasesbecauseofsnowmeltandwhenwindgenerationcorrespondinglyisatahighlevelas

    well.

    1

    2

    3

    4 567

    8

    9

    0

    5

    10

    15

    20

    25

    30

    35

    40

    0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39

    Intermittent generation penetration level (% of total system energy)

    CapacityCredit(%

    ofinstalled

    intermittentgenerationcapacity)

    1

    2

    3

    45

    6

    7

    8

    9

  • 7/31/2019 AA WindModel

    19/99

    9

    Table ES-4. Examples of wind capacity credit methods in the United States

    Region/Utility Method Note

    CA/CEC ELCC Rank bid evaluations for RPS (low 20s)

    PJM Peak Period

    Jun-Aug from 3 p.m.-7 p.m., capacity factor using 3-

    year rolling average (20%, fold in actual data when

    available)

    ERCOT 10%May change to capacity factor, 4 p.m.-6 p.m., Jul

    (2.8%)

    MN/DOC/Xcel ELCC Sequential Monte Carlo (26-34%)

    GE/NYSERDA ELCC Offshore/onshore (40%/10%)

    CO PUC/Xcel ELCC

    PUC decision (30%) and Current Enernex study

    possible follow-on, Xcel using MAPP approach (10%)

    in internal work

    RMATS Rule of thumb 20% all sites in RMATS

    PacifiCorp ELCC Sequential Monte Carlo (20%)

    MAPP Peak Period Monthly 4-hour window, median

    PGE 33% (method not stated)

    Idaho Power Peak Period 4 p.m.-8 p.m. capacity factor during July (5%)

    PSE and Avista Peak PeriodPSE will revisit the issue (lesser of 20% or 2/3 Jan

    C.F.)

    SPP Peak Period Top 10% loads/month; 85th

    percentile

    Source: Milligan, Michael, and Kevin Porter (2005). Determining the Capacity Value of Wind: A Survey ofMethods and Implementation. Golden, CO: National Renewable Energy Laboratory. Available atwww.nrel.gov/docs/fy05osti/38062.pdf.

    Ramping:Attimes,windgenerationcanrampupanddownquickly,particularlyinresponseto

    storms.In

    general,

    ramping

    events

    are

    of

    more

    concern

    to

    smaller,

    weaker

    grids

    with

    few

    externalinterconnectionsandgridswithlargeconcentrationsofwindprojectsinoneregion.

    Gridswiththesefeaturestypicallydonothaveadeepstackofgeneratingresources,

    connectionstootherregionsorthelargegeographicdiversityofwindresourcestomanage

    rampingevents.Forthisreason,theTSOsthathaveproposedorimplementedrampinglimits

    onwindturbineshavetendedtobesmallergridsorgridswithfewexternalinterconnections.

    OneexceptionisinGermany,wheretheTSOslimitthepositiveramprateofwindgenerationto

    10percentofratedpowerperminute.Someexamplesincludethefollowing:

    EirGridinIrelandlimitsthepositiveramprateto130MWperminute

    Scotland,

    where

    the

    positive

    ramp

    rate

    is

    limited

    to

    110

    MW

    per

    minute,

    depending

    on

    thecapacityofthewindproject,andthedownwardramprateto3.3percentofpower

    outputperminute

    TheAlbertaElectricSystemOperatorhasproposedlimitingsystemwiderampratesfor

    windprojectsto4MWperminute.

  • 7/31/2019 AA WindModel

    20/99

    10

    TheIAPwillassesstherampingimpactsofvariableresourcesontheCaliforniagrid.Asastate,

    CaliforniahasarelativelydeepresourcestackandinterconnectionswiththePacificNorthwest

    andtheSouthwest.Californiaisnotintheextremesituationasislandsorsmallergrids.In2006

    theCaliforniaWindEnergyCollaborative(CWEC),underaconsultingagreementtotheEnergy

    Commission,examinedrampingcapabilityintheCAISObasedonpubliclyavailabledata.

    CWECdetermined

    that

    the

    CAISO

    had

    sufficient

    ramping

    capability

    to

    accommodate

    load

    variabilityandcurrentlevelsofvariablerenewableenergygeneration.

    TransmissionRatingandUnscheduledGeneration:Attimes,thecombinationofwindfrom

    DenmarkandGermanycanresultinunscheduledpowerflowsontheEuropeantransmission

    grid,especiallyduringtimesofhighwindproductionandlowdemand.Thelackofsufficient

    northtosouthtransmissioninGermanyresultsinwindgenerationfromNorthernGermany

    beingtransmittedtocustomersinSouthernGermanyviathetransmissionnetworksofthe

    Netherlands,BelgiumandFrance.

    In2005theElectricPowerGroup(EPG),underconsultingagreementtotheEnergy

    Commission,suggested

    that

    the

    frequency

    response

    of

    generating

    resources

    in

    California

    and

    throughouttheWesternElectricityCoordinatingCouncil(WECC)hasdecreasedinrecentyears

    becauseofseveralgeneratingresourcesoperatingatbaseloadwithlimitedupwardcapability.

    That,inturn,couldleadtoreducedtransmissionpathratingsintoCaliforniaandthroughout

    WECC.Furthermore,theEPGfoundthatasignificantresourceshifttomorerenewable

    resourcesinWECC,withoutcorrespondingattentiontothethermalcapabilityofgenerators,

    voltagesupport,andhowgeneratorsperformduringcontingencyevents,couldcompoundthis

    issue.Theimpact,ifany,wouldarisemostlikelyduringnonpeakhours.

    MitigationandOperatingSolutionstoDateSeveral

    strategies

    have

    been

    proposed

    and

    implemented

    to

    integrate

    variable

    renewable

    energy

    generation,primarilywind.Theseincludewindforecasting,gridcodes,curtailment,wind

    turbinemodelingandverification,demandresponse,andtransmissionplanningand

    development.

    WindForecasting:Ingeneral,windgenerationcanbepredictedmoreaccuratelythecloserit

    occurstoactualoperation.Windgenerationcanbepredictedwithabout90percentorgreater

    accuracyonehourahead,with70percentaccuracyninehoursaheadbutonlyabout50percent

    accuracy36hoursahead.Themeanabsoluteerrorbyinstalledcapacityforwindforecastingin

    Denmarkistypicallybetween8and9percent,whichisequivalenttoa38percentforecasterror

    byenergy.InGermany,therootsquaremeanerror(RSME)ofwindforecastsis5to8percentof

    installedwind

    capacity

    with

    maximum

    errors

    ranging

    from

    30

    to

    40

    percent

    of

    installed

    wind

    capacity.OnafourhouraheadbasisinGermany,theRSMEis3.8percent,withamaximum

    errorrangingfrom28to36percent.

    Contributorstowindforecastingerrorsincludephaseerrors,whichoccurwhenwind

    forecastspredictstorms.Inpractice,thestormmayoccurafewhoursaheadorfewhours

    behindthewindforecast.Anothercontributortowindforecastingerrorsistherelativelylow

    spatialandtemporalqualityofmeteorologicaldata.Mostforecastinghasbeenfocusedon

  • 7/31/2019 AA WindModel

    21/99

    11

    weatherattributessuchasprecipitationandtemperature,withalowerspatialandtemporal

    resolutionthanisrequiredforwindgeneration.Manybusinessandgovernmentalentitiesare

    becominginterestedinfiner,morepreciseforecasting,andthatinturnmaycorrespondto

    betterdataforimprovingwindforecasting.

    In

    2002,

    the

    CAISO

    became

    the

    first,

    and

    to

    date

    the

    only,

    regional

    transmission

    operator

    in

    the

    UnitedStatestooffercentralizedwindforecastingtopredicttheoutputofvariablerenewable

    energygeneration.TheParticipatingIntermittentResourceProgram(PIRP)isvoluntary.To

    date,onlywindgenerationisenrolledinPIRP,althoughwithseveralproposedlargescalesolar

    projectsinCalifornia,itispossiblethatsolarwilljoinwindinthePIRPprogram.InPIRP,the

    positiveandnegativeimbalancesassociatedwiththe10minuteschedulesofwindpower

    generatorsarenettedoutandsettledonamonthlybasis,withthenotionthattheseimbalances

    willcanceloutoverthemonth.Anynetimbalancesattheendofthemonth,positiveor

    negative,aresettledattheweightedaveragezonalmarketclearingprice.TheCAISOisallowed

    tochargepenaltiesforexcessivedeviationsofageneratorcomparedtoadvanceschedulesbut

    doesnotatthistime.IftheCAISOchargesthispenalty,participatingintermittentresourcesin

    PIRPwould

    be

    exempt.

    Initially,PIRPwashandicappedbymissingtelemetrydatacausingvariationsinthewind

    forecast;however,mostofthistypeoferrorhasbeencorrected.Therearesomemarket

    participantconcernsregardingthereallocationofcostsfromwhichparticipatingintermittent

    resourcesareexempt.TheCAISOisexploringmakingseveralenhancementsandchangesin

    hopesofreducingthesecostconcerns.Theseenhancementsincludeincreasingtheforecasting

    feesforbeinginPIRPandsubjectingpowerexportsfromparticipatingintermittentresourcesto

    higherfees.InDecember2006,theFederalEnergyRegulatoryCommission(FERC)approved

    theCAISOspetitiontochargeanexportfeetoPIRPfacilitiesthatexportpoweroutofthe

    CAISO

    control

    area.

    GridCodes:Acommonapproachtakenbymanytransmissionsystemoperatorstoincorporate

    wind,istoadoptgridcodesspecifictowindgenerators.Germanyintroducedtheirwindgrid

    codein2003,followedbyDenmarksTSOsinlate2004.Britain,Ireland,andtheUnitedStates

    havesincefollowedwithwindgridcodesin2005.

    Theintentistoensurethatwindprojectsdonotnegativelyimpactreliability.Alargeamountof

    windcapacitytrippingofflineinresponsetoagriddisturbancecouldleadtoafallinvoltage

    and/orfrequency.That,inturn,couldcontributetoothergeneratorstrippingoffthegridand

    couldresultinnothavingenoughgenerationtomeetload.Thegridcodeshaveemergedona

    transmissionoperatororcountrybasis,anddifferencesbetweenthegridcodeshavenaturally

    resulted.To

    date,

    wind

    specific

    grid

    codes

    have

    required

    wind

    power

    facilities

    to

    address

    one

    ormoreofthefollowingconditionsto:

    Ridethroughgridfaults

    IncreaseordecreasepowergenerationattheTSOsrequest

    Supplyreactivepower

    Adjustpowergenerationinresponsetofrequencychanges

  • 7/31/2019 AA WindModel

    22/99

    12

    Controlorlimitrampingincreases.

    Generally,allwindgridcodeshaveafaultridethroughrequirementspecifyingthatwind

    generatorsmuststayconnectedforaperiodoftimewhenfaultsoccuronthetransmission

    systemandvoltagedrops.AsindicatedinTableES5,faultridethroughrequirementsdifferby

    country.

    Table ES-5. Examples of wind grid codes

    Grid Code Fault Duration

    (Milliseconds)

    Voltage Drop

    During Fault

    (% Nominal)

    Voltage Recovery

    (Milliseconds)

    Denmark 100 25 1000

    Germany(E.On) 150 0 1500

    Ireland(EirGrid) 625 15 3000

    UK(NGT) 140 0 1200

    Spain 500 20 1000

    UnitedStates 150 0* NA

    *As of 2008. For 2007 and for normally cleared three-phase faults, wind turbines must be able to ridethrough voltages down to 15 percent at the point of interconnection for 150 milliseconds. Source:Milborrow, David. 2005b. Going Mainstream at the Grid Face. Windpower Monthly, September 2005,p. 49. Reproduced by permission. United States provisions drawn from Federal Energy RegulatoryCommission. December 12, 2005. Order No. 661-A. Interconnection for Wind Energy.

    Asmallernumberofcountriesalsorequirewindturbinestoprovidefrequencyresponsein

    ordertomaintainthefrequencyat50Hz(thelevelinEurope).Windturbineshavealimited

    abilitytoprovidefrequencycontrolascomparedtoconventionalunits.Tomeetthis

    requirement,windturbinesmustbeoperatedatlessthanfulloutput,suchthatbladepitchcan

    beadjustedtoincreasegenerationwhencalledupon.Thisisanoptiononnewerpitch

    controllableturbines.

    Ireland

    requires

    wind

    generators

    to

    provide

    primary

    frequency

    control

    of

    35percentofpoweroutputandtoprovidesecondaryfrequencycontrolifcalledupon.

    DenmarkandtheUnitedKingdomrequirewindgeneratorstoprovidefrequencycontrolaftera

    systemfaultorifpartofthegridisisolated.Similarly,transmissionsystemoperatorsarealso

    requiringwindgeneratorstostayonlineduringfrequencydeviations,asindicatedinFigure

    ES4.

    Gridcodesalsogenerallyrequirewindturbinestooperatecontinuouslyatratedoutputin

    normalvoltageranges,tostayonlineduringvoltagechangeswithinaspecifiedrange,andto

    supplyreactivepower.Forinstance,E.OnNetzinGermanyrequireswindturbinestocontinue

    tosupplyreactivepowerforuptothreesecondsafteravoltagedrop.Sweden,Norwayand

    Spainalso

    have

    provisions

    for

    wind

    turbines

    and

    reactive

    power.

  • 7/31/2019 AA WindModel

    23/99

    13

    Figure ES-4. Frequency control requirements by selected countrySource: Van Hulle, Fran. 2005. Large Scale Integration of Wind Energy in the EuropeanPower Supply. Brussels, Belgium: European Wind Energy Association. Available athttp://www.ewea.org/fileadmin/ewea_documents/documents/publications/grid/051215_Grid_report.pdf.

    Inthe

    United

    States,

    FERC

    adopted

    agrid

    code

    in

    2005

    for

    wind

    turbines.

    A

    WECC

    task

    force

    is

    alsoconsideringpossiblechangestoWECCscurrentlowvoltageridethroughstandardto

    lowertheminimumvoltagetoleranceperiodtozeroatthepointofinterconnectionfor12cycles

    (about1/5ofasecond).

    WindTurbineModelingandValidation:Acommonissuewithwinddevelopmentistheneedto

    improvethemodelingofwindprojectsfordeterminingthepotentialimpactsonsystem

    reliabilityduringtheevaluationofinterconnectionapplications.Lackofknowledgeby

    transmissionsystemoperatorsaboutwind;theincreasingsizeofwindprojects;andtheoften

    weaktransmissionnetworkthatwindprojectswereattemptingtointerconnecttohavemade

    interconnectionmodelingachallenge.TheWECCWindGeneratorModelingGroupis

    preparingwind

    turbine

    generator

    models.

    In

    Europe,

    continued

    growth

    of

    wind

    energy

    in

    some

    countriesmaybeconditionedonnotonlyresolvinguncertaintiesaboutthegridimpactsof

    windturbinesbutalsoontheavailabilityofvalidatedanalyticaltoolsandmodels.ESBin

    Irelandhasinstitutedcertificationrequirementsforwindturbinemodelstobeusedinsystem

    interconnectionstudiesaspartofIrelandsgridcode.

    DemandResponse:Demandresponsemayhelpintegratelargeramountsofwindpowerby

    movingconsumptionfromwhenwindproductionislowtotimesofhigherwindproduction,

  • 7/31/2019 AA WindModel

    24/99

    14

    therebylesseningtherequirementforreservesfromconventionalpowerplants.Oneexample

    researchedinDenmarkistouseelectricityproductionfromwindgenerationduringoffpeak

    hoursfordistrictwaterheatinginsteadofotherfuels.Sofar,participationindemandresponse

    programshasbeenrelativelysmallinEuropeandintheUnitedStates,althoughregulatoryand

    industryinterestisgrowing.Californiahassettargetsforutilitiestomeet3percentofitsannual

    peakdemand

    with

    demand

    response,

    increasing

    1percent

    per

    year

    to

    5percent

    by

    2007

    and

    favorsdemandresponseandenergyefficiencyoverotherresourcesinmeetingnewelectricity

    demand.

    WindPowerCurtailment:Maximumwindproductioncanbeseveraltimeslargerthanaverage

    windproduction,meaningthatat20percentwindpenetrationbyenergy,windproductionmay

    equalconsumerdemandforsomehours.Curtailmentofwindgenerationmaybenecessaryif

    theamountofwindgenerationataspecifictimeismorethanwhatthegridcanreliablyhandle.

    Infact,forgridswithsmallcontrolareasthataredominatedbythermalgenerationthatmaynot

    beveryflexible,windcurtailmentscouldoccuratpenetrationsaslowas10percent.

    InNorthern

    Germany,

    E.

    On

    Netz

    implemented

    curtailment

    policies,

    or

    generation

    managementasdescribedbyE.OnNetz,forwindgeneratorsintheSchleswigHolsteinregion

    inmid2003,covering700MW(about1/3ofthewindcapacityinthatregion),andexpandingit

    toLowerSaxonyin2005.Ifoverloadconditionsarepresent,E.OnNetzidentifiestheregionof

    concernandsendsasignaltowindprojectstoadjustoutputaccordingly,definingthe

    maximumactiveoutputthattheregionswindprojectscanprovidetothegrid.Untilnew

    transmissioncapacityisadded,E.OnNetzwillnotinterconnectnewwindprojectsin

    SchleswigHolsteinunlessthewindgeneratorsparticipateinE.OnNetzsgeneration

    managementprogram.Spainalsocurtailedwindgenerationin2004whenwindpower

    penetrationexceeded12percentofdemand,duetolocalgridlimitations.Thesewind

    curtailments

    occurred

    less

    frequently

    in

    2005.

    TransmissionPlanningandDevelopment:Stronggridinterconnectionshaveplayedapartin

    helpingDenmarkmanageitshighlevelofwindproduction.Ingeneral,though,thereislimited

    interconnectionbetweennationalandregionalelectricitymarketsinEurope,andcurrenttrans

    countryinterconnectionscanbeheavilyloaded.TheInternationalEnergyAgencypredictsthat

    $1.8trillionoftransmissionanddistributioninvestmentsarenecessaryby2030simplytomeet

    demandgrowthandtoupgradeexistingassetsinEurope.Californiahasextensive

    interconnectionswiththePacificNorthwestandwiththeDesertSouthwest,andthestateis

    workingonnewtransmissionthatwillbenecessaryifCaliforniaisgoingtomeetits20percent

    RPSby2010.Anumberoftransmissionplanningactivitiesareoccurringbothinsideand

    outsideof

    California.

    In

    August

    2006,

    the

    CAISO

    Board

    of

    Governors

    approved

    the

    Sun

    Path

    projectthatwilladd1,000MWoftransmissioncapacitytoSouthernCaliforniaprovidingaccess

    togeothermalandsolarresourcesintheImperialValley.TheCAISOBoardofGovernorsis

    consideringproposedtransmissionprojectsinTehachapiandtheLakeElsinoreAdvanced

    PumpStorage(LEAPS)project.OutsideofCalifornia,morethanadozentransmissionprojects

    havebeenproposed,withsomeoftheseproposalstargetingCaliforniaastheultimatemarket.

    Manyoftheseproposalsareataveryearlystage,andnotallofthemmaybeconstructed.

  • 7/31/2019 AA WindModel

    25/99

    15

    ConclusionsNearlytwothirdsoftheworldswindinstalledcapacityisinEurope,withGermany,Spain,

    andDenmarkaloneaccountingforonehalfoftheworldsinstalledwindcapacity.Wind

    developmentinEurope,atleastinitially,differedfromthelargerutilityscaleprojectsinthe

    United

    States,

    particularly

    in

    Denmark

    and

    Germany,

    where

    wind

    development

    consisted

    of

    smaller(butnumerous)windprojectsinterconnectedtothedistributiongrid.Thattypeofwind

    developmentinDenmarkandGermanytookadvantageofthegeographicdiversityofwind

    resourcestosmoothsomeofthevariabilityinwind.

    SimilarmanagementstrategiesbetweentheUnitedStatesandEuropehavebeguntoemergeas

    winddevelopmenthasexpandedtoothercountrieswithlessrobustgridinfrastructure,as

    comparedtoDenmarkandGermany,andaswinddevelopmenthastendedtowardsutility

    scaleprojectsthatarecommonintheUnitedStates.Theimplementationofgridcodes(although

    varyinginspecificsfromcountrytocountry)isonesuchexample.Theneedfortransmissionin

    bothEuropeandtheUnitedStates,notjustforwindgenerationbutforalltypesofgeneration,

    isanother

    similarity.

    Considerable

    transmission

    planning

    and

    activity

    is

    underway

    in

    both

    EuropeandtheUnitedStates.

    Theparticularcircumstancesineachcountry,stateorregionwilldeterminetheeaseof

    integratingvariablerenewableenergygeneration.Thesefactorsincludethegeneratingmix;the

    flexibilityofresourcesinmix;whethertherearerobustdayaheadmarketswithdeepresource

    stacks;thelocationofwindresources;transmissionavailability;andthesizeofcontrolareas.

    Windintegrationwillalmostcertainlybemorechallenginginsmallcontrolareas,inareaswith

    limitedinterconnections,orinareaswithasmallloadand/orsmallresourcestacksascompared

    toregionswithlargercontrolareas,extensiveinterconnectionsorlargeloadsand/ordeep

    resourcestacks.Becausethesecircumstancescanvarydramatically,cautionshouldbeusedin

    comparingcountries

    or

    regions

    with

    each

    other.

    Thisreportexaminedhowcountriesoverseashaveincorporatedvariablerenewableenergy

    generation,whatoperatingstrategieshavebeenusedtointegratevariablerenewableenergy

    generation,whatlessonshavebeenlearned,andwhetherthatexperienceistransferableto

    California.Foravarietyofreasons,thereportfocusedmostlyonwind,giventhatthereismore

    gridconnectedwindcapacityworldwidethansolar;theexperiencewithwindismorewidely

    reported;andthedevelopmenttodateofsolarsystemshasbeenofsmall,distributedsystems

    and,atleastasofnow,doesnotfacethesamesystemintegrationissuesaswindpower.

    Somehighlightsofintegrationstrategiesandfindingsfromvariouscountryreportsinclude:

    Strategiesimplementedtoincorporatewindincludewindforecasting,gridcodes,

    curtailment,windturbinemodelingandverification,demandresponse,and

    transmissionplanninganddevelopment.

    Todate,gridcodeshavefeaturedthesemajorthemes:requiringwindturbinestoride

    throughgridfaults;increasingordecreasingpowergenerationattheTSOsrequest;

    supplyingreactivepower;adjustingpowergenerationinresponsetofrequencychanges;

    andcontrollingorlimitingrampingincreases.

  • 7/31/2019 AA WindModel

    26/99

    16

    VariousEuropeantransmissionsystemoperatorshaveimplementedmorecontrol

    requirementsforwindthanhavebeenseenintheUnitedStatessofar,suchasramprate

    limitsandtherequirementtoprovidereservesandfrequencycontrol.Ingeneral,these

    controlrequirementshavebeenafunctionofsmallcontrolareasorlimitedtransmission

    interconnections,orboth.

    Someof

    the

    more

    stringent

    wind

    control

    strategies

    have

    been

    proposed

    in

    countries

    that

    havelittleornogridinterconnections,andtheseparticularcircumstancesneedtobe

    keptinmindwhencomparinginternationalwindintegrationexperiences.Ramping

    eventswillbeofmoreconcerntosmallgrids,orgridswithfewexternal

    interconnections,orgridswithalargeconcentrationofwindprojectsinoneregion.

    Countrieswithmusttakerequirementsintheirrenewableenergyfeedinlawstendto

    havethetoughestgridcodeprovisionswithregardstowindcurtailment.

    Indescribingvariousancillaryservices,EuropeandtheUnitedStatesusedifferent

    terminology.InEurope,primaryreservesassistwiththeshortterm,minutetominute

    balancing

    and

    control

    of

    the

    power

    system

    frequency,

    and

    is

    equivalent

    in

    the

    United

    Statestoregulation.SecondaryreservesinEuropetakeoverforprimaryreserves10to

    30minuteslater,freeingupcapacitytobeusedasprimaryreserves.Theclosest

    terminologyintheUnitedStatesforsecondaryreservesiseitheroperatingreservesor

    loadfollowingreserves,whichmayincludebothspinningandnonspinning

    components.LongertermreservesinEuropearecalledtertiaryreservesandare

    availableintheperiodsaftersecondaryreserves.Tertiaryreservesareclosestto

    supplementalreservesintheUnitedStates,althoughthetimescalesmaybedifferent

    betweenEuropeandtheUnitedStates.

    Reconstitutingexistingreserveservicesmaybenecessaryashigherlevelsofvariable

    renewable

    energy

    generation

    is

    added.

    Submittingscheduleswithshorterperiodsoftimebeforetherealtimemarketbegins

    willallowformoreaccuratepredictionsofwindgeneration,althoughsometradeoffs

    areinvolved.

    Variouswindintegrationstudiesandtransmissionsystemoperatorshavereported

    someoperatingissueswithwindgeneration,suchasminimumloadandhighramp

    rates.ANewZealandwindintegrationstudyusedminimumloadtodeterminehow

    muchwindcouldbeaccommodatedonitsgrid.

    Forramping,variousstudiessuggestthatwindwillrampupanddownwithin10

    percentofcapacitymuchofthetimeoveranhour.Handlingwindrampingcouldbe

    managedwith

    sufficient

    regulation

    or

    load

    following

    generation;

    wind

    forecasting

    to

    predictvariabilityandrampingevents;performancelimitsonthewindgenerationsuch

    asrampratelimits;orsharingreservesorenergyimbalancesovermultiplecontrol

    areas.

    Effortsarealsounderwayonimprovingthemodelingofwindprojectsfordetermining

    thepotentialimpactsonsystemreliabilityduringtheprocessofevaluating

    interconnectionapplicationsfromwindgenerators.

  • 7/31/2019 AA WindModel

    27/99

    17

    Intermsofwindintegrationcosts,theresultsofvariousstudiesconductedtodateintheUnited

    Statesandoverseashavebeenreasonablyconsistent.Overall,thefindingscanbesummarized

    asfollows:

    Thecostforintegratingwindisnonzeroandincreasesastheproportionofwind

    generation

    to

    conventional

    generating

    resources

    or

    peak

    load

    increases;

    Reservecostsattributedtowindintegrationarerelativelysmallatwindpenetration

    levelsoflessthan20percent.Howthevariabilityanduncertaintyofwindgeneration

    interactswithvariationsinloadandloadforecastinguncertaintyhasalargeimpacton

    thelevelofwindintegrationcosts.

    Levelofgeographicconcentrationofwindprojectsalsoaffectswindintegrationcosts.

    Unitcommitmentimpactshavebeenamajorfocusofwindintegrationstudiesinthe

    UnitedStatesbuthavenotbeenaddressedasextensivelyintheEuropeanstudiesto

    date.

    BasedonseveralEuropeanstudiesthatestimatedthecostsofadditionalreserveswith

    windgeneration,costsweregenerallylessthan$6/MWhatwindenergypenetrationlevelsupto20percent,althoughthecostsvariedsignificantlyamongtheindividual

    studies.

    Reservecostsforwindgenerationaredependentonthecharacteristicsofthegridthatis

    integratingwind,theadequacyandcharacteristicsoftheexistingreserves,andthe

    specificreserverequirementsforeachgrid.

    StudiesestimatingthecapacitycreditofwindpowerinEuropedeterminedthatwind

    hasacapacitycreditgreaterthanzero,andalsothatthecapacitycreditdecreasesasthe

    levelofwindgenerationrises.

    Factorsthat

    affect

    the

    capacity

    credit

    of

    wind

    include

    present

    levels

    of

    wind

    generation

    onthegrid;thequalityofthewindresource;thecapacityfactorofthewindprojects;

    whetherdemandandwindgenerationarecorrelatedoruncorrelated;thedegreeof

    systemsecurity;andthestrengthofthetransmissioninterconnections.

    Astimegoeson,moresimilaritiesthandifferencesareapparentbetweenEuropeandthe

    UnitedStatesasvariablerenewableenergygenerationincreasesinmarketpenetration.These

    similaritiesaresparkinginformationexchangeandtransferthroughforumssuchastheIEA,the

    InstituteofElectricalandElectronicsEngineersandtheUtilityWindIntegrationGroup

    (UWIG).That,inturn,canhelpelevateprominentissuesandmakethetaskofdeveloping

    solutionsandoptionsforintegratingvariablerenewableenergygenerationeasier.

    BenefitstoCaliforniaCaliforniahasperhapsthemostsignificantanddiverseRPSintheUnitedStatesintermsofthe

    level(20percent),timeframe(2010)andtheamountofrenewableenergycapacitythatmaybe

    requiredtomeetthetarget.Transmissionandtheintegrationofvariablerenewableenergy

    generationremainchallengesthatneedtobeaddressedinorderforCaliforniatomeetitsRPS

    goals.VariouscountriesinEuropehaveexperiencewithintegratinghighlevelsofvariable

  • 7/31/2019 AA WindModel

    28/99

    18

    renewableenergygeneration.Byreviewingandhighlightingstrategiesandpracticesthathave

    beenusedtointegratewindinotherstatesandinothercountriesinthisreport,theIAPmay

    incorporatesomeofthesestrategiesandpracticesasoptionstotestpotentialeffectivenessin

    integratingvariablerenewableenergygenerationinthestate.ThehopeisthatCalifornia

    projectsandutilitiescanbegintoevaluateandincorporatesomeoftheseapproachesandtotest

    theireffectiveness

    in

    integrating

    renewables.

  • 7/31/2019 AA WindModel

    29/99

    19

    1.0 Introduction

    Growthinwindandsolarhasbeensurginginrecentyears.Windcapacityworldwideincreased

    by25%in2006ascomparedto2005,andEuropereachedits2010goalof40,000MWinstalled

    windcapacityfiveyearsearly(GlobalWindEnergyCouncil2006).Solarcellproductionhas

    been

    increasing

    at

    over

    25%

    annually,

    and

    shortages

    in

    materials

    for

    solar

    cells

    and

    solar

    cells

    themselveshavebeenreported(EarthPolicyInstitute2004).

    Withgrowthcomeconcernsoverhowtheelectricitygridwillintegratevariablerenewable

    energyresourcessuchaswindandsolar.Thisreportreviewsthecurrentstudies,practiceand

    experienceintegratingvariablerenewableenergygeneration.Theapproachforthispaperhas

    beentoreviewnumerousreports,presentationsandconferencepapersandtofocusonissues

    identifiedwithintegratingvariablerenewables.Foravarietyofreasons,thispaperwill

    primarilyciteexamplesforwindgiven:

    thereismoregridconnectedwindcapacityworldwidethansolar;

    theexperience

    with

    wind

    is

    more

    widely

    reported;

    and

    thedevelopmenttodateofsolarsystemshasbeenpredominantlyofsmall,distributed

    systemsand,atleastasofnow,doesnotfacethesamesystemintegrationissuesaswind

    power.

    Withanumberofincentiveprogramsforsolar,particularlyinGermanyandSpain,grid

    connectedsolargenerationisstartingtoincrease.Ofthelargest20gridconnectedphotovoltaic

    (PV)powerplantsintheworld,16havebeeninstalledin2004orlater(PVResources.com2006).

    Twothirdsofthe74GWofworldwidewindcapacityislocatedinEurope,makingEuropean

    interestingcasestudyforstudyingthegridimpactsofwind.Althoughwindprovidesabout3%

    of

    Europes

    electricity,

    some

    regions

    have

    considerably

    higher

    wind

    penetrations

    as

    indicated

    in

    Table1,suchasWesternDenmark(>20%)andSchleswigHolsteininGermany(~30%)

    (Holttinen2004).Ultimately,someestimatesindicatethatwindmayprovide12%ofEuropes

    electricitydemandby2020and30%by2030(VanHulle2005).

  • 7/31/2019 AA WindModel

    30/99

    20

    Table 1. Examples of wind power penetration levels, 2005

    Country or region Installed wind

    capacity

    (MW)

    Total installed

    power capacity

    (MW)

    Average

    annual

    penetration

    levela(%)

    Peak

    penetration

    levelb

    (%)

    Western Denmark 3,128 7,488 ~23 >100Germany: 18,428 124,268 ~5 n.a.

    Schleswig-Holstein 2,275 _________c ~28 >100

    Spain 10,028 69,428 ~8 ~25%

    Island systems:

    Swedish island of

    Gotlandd

    90 No local generation

    in normal state

    ~22 >100

    n.a. = Not availableaWind energy production as share of system consumption

    bLevel at high wind production and low energy demand, hence, if peak penetration level is >100%

    excess energy is exported to other regions.cGerman coastal province

    d2002 data. The island of Gotland has a network connection to the Swedish mainland.

    Source: Adapted from Soder, Lennart and Ackerman, Thomas (2005). Wind Power in Power Systems: AnIntroduction, In T. Ackerman (Ed.), Wind Power in Power Systems(pp. 25-51). England: John Wiley andSons, Ltd. Updated and adapted by the author. Reproduced with permission.

    ThemajorityofwinddevelopmentinEuropehastakenplaceinthreecountries:Denmark,

    Germany,andSpain.Together,thosethreecountriesaccountfor50%ofworldwideinstalled

    windcapacity.WinddevelopmentinDenmarkandGermanyhasconsistedofsmall

    installationsofwindturbinesthatarewidelydistributed,takingadvantageofthegeographic

    dispersion

    of

    wind

    resources

    and

    providing

    some

    smoothing

    of

    winds

    variability.

    DenmarkandGermanyalsohavestronginterconnectionswithothercountries,allowingthe

    exportofsurpluswindproductionandtheimportofpowerwhenwindproductionislow.

    Morerecentwinddevelopmentinothercountrieshasoccurredwherethereislittleornogrid

    interconnectionwithothercountries.ExamplesincludeSpain,Ireland,andBritain,where

    internationalgridinterconnectionsaremorelimited.

    AsonshorewinddevelopmentinEuropebecomesmoresaturated,winddevelopmentwill

    likelymoveoffshoreandbemoreconcentratedinsmallergeographicareas.Over54GWof

    offshorewindisinvariousstagesofplanninginEurope(LiebreichandYoung2005).In

    Germanyalone,between25and30GWofoffshorewindcapacityisplannedfortheNorthand

    BalticSeas

    by

    2030

    (Deutsche

    Energie

    Agentur

    2005).

    Not

    only

    will

    wind

    capacity

    be

    more

    concentrated,losingsomeofthesmoothingeffectsforwindfromgeographicdispersion,but

    someoftheproposedoffshorewinddevelopmentisinregionsthatalreadyhavehighwind

    penetration,suchasNorthernGermany,furtheraddingtotheintegrationchallenges.

    AlthoughpresentoperatingpracticeshaveallowedEuropetomanagewindsvariability,there

    issomethoughtthatnewstrategieswillbenecessarytoaccommodatethefuturegrowthof

  • 7/31/2019 AA WindModel

    31/99

    21

    wind.TheUnionfortheCoordinationofTransmissionofElectricity(UCTE),theassociationof

    transmissionsystemoperatorsfrom23Europeancountries,issuedastatementinMay2005

    callingformoregridinfrastructureandotheractionstointegratewindintheEuropeangrid

    (UCTE2005).TheEuropeanWindEnergyAssociationalsoanticipatesthatsomechangesmay

    benecessaryinoperatingthegridathigherlevelsofwindpenetration,andsuggestedthat

    planningbegin

    for

    those

    changes

    (Van

    Hulle

    2005).

    The

    IEA

    is

    sponsoring

    an

    annex,

    Design

    andOperationofPowerSystemswithLargeAmountsofWindPowerProduction,thatbegan

    inmid2006(InternationalEnergyAgency2006).Finally,theEuropeanTransmissionSystem

    Operators(ETSO),theassociationoftransmissionsystemoperatorsinEurope,announcedplans

    toconductaEuropewidewindintegrationstudy.Theplannedstudywillencompass16TSOs

    in14countriesthatrepresentthefourmajorsynchronouselectricitygridsinEurope.Early

    resultsfocusingonwindintegrationsolutionsineachsynchronousgridareexpectedin2008

    (ETSO2006).

    Thegridsituationisdifferentaswinddevelopmentspreadstoothercountriesaroundthe

    world.India,forexample,doesnothaveanationalgridbutinsteadhasfivestateowned

    regionalgrids,

    with

    the

    grids

    in

    rural

    areas

    tending

    to

    be

    weak.

    Periodic

    power

    outages

    in

    India

    arecommonandcauseupto$25billionineconomicdamagesannually,accordingtothe

    governmentofIndia(Sieg2006).Indiahasmovedintofourthplaceamongcountrieswiththe

    mostinstalledwindcapacityandmetits2012targetof5,000MWofwindcapacityin2006

    (RajgorandMathews2006).Similarly,Chinasexplosiveeconomicgrowthhasexceeded

    availableelectricitysuppliesandledtoelectricityshortages,withtwothirdsoftheprovincesin

    Chinaexperiencingblackoutsin2004(Kuetal.undated).Chinahasabout2,600MWofwind

    capacityandhassetagoalof30GWofwindby2020(Jianxiang2006).WindprojectsinChina

    mustmeeta50%localcontentstandardforprojectsapprovedbefore2005,increasingto70%for

    projectsapprovedafter2005.

    Theparticularcircumstancesineachcountry,stateorregionwilldeterminetheeaseof

    integratingvariablerenewableenergygeneration.Amongotherthings,thisincludessuch

    factorsaswhetherthegeneratingmixhasflexibleresourcesornot;whethertherearewell

    functioninganddeephouraheadanddayaheadmarkets;whetherthewindprojectsare

    relativelyspreadoutorconcentrated;whetherthereisavailabletransmission;andwhetherthe

    controlareasarefairlybroadorrelativelysmall.Becausethesecircumstancescanvary

    dramatically,cautionshouldbeusedincomparingcountriesorregionswitheachother.Wind

    integrationwillalmostcertainlybemorechallenginginsmallcontrolareas,inareaswithnot

    muchinterconnections,orinareaswithasmallloadand/orsmallresourcestackascomparedto

    regionswithlargercontrolareas,extensiveinterconnectionsorlargeloadsand/ordeepresource

    stacks.Some

    of

    the

    more

    stringent

    wind

    control

    strategies

    have

    been

    proposed

    in

    countries

    that

    havelittleornogridinterconnections,andtheseparticularcircumstancesneedtobekeptin

    mindwhencomparinginternationalwindintegrationexperiences.

    Thatsaid,theinternationalexperiencewithwindofferssomelessonsforregionsintheUnited

    Statesthathaveorareexpectingsignificantadditionsofwindcapacity.Already,somecountries

    havedevelopedwindforecastingstrategiesandgridcodesaddressingwindpowersystemsthat

    haveformedthebasisforsimilaractionsintheUnitedStates.Thattrendislikelytocontinue.

  • 7/31/2019 AA WindModel

    32/99

    22

    Moreexperiencewithwindintegrationwillbegainedascountriesaddwindtotheirgenerating

    mix.

    Thereportisorganizedasfollows.Theremainderofthischapterprovidesanoverviewof

    worldwidewindandsolarcapacity.Chapter2reviewstheresultsofwindintegrationstudies

    and

    practices

    in

    the

    United

    States

    and

    Europe.

    Chapter

    3

    discusses

    the

    effects

    of

    market

    structureandreviewshowthecapacitycreditofwindisdeterminedinternationallyandinthe

    UnitedStates.Chapter4describesgridoperationissueswithwindtodate.Chapter5reviews

    thesolutionsthatgridoperatorshavedevelopedtohandlethevariabilityofwindgeneration.

    Chapter6presentssomefindingsandimplicationsforCalifornia,whileChapter7provides

    conclusions.Countryspecificprofilesareofferedintheappendixonfourofthefiveleading

    countriesintheworldinregardstoinstalledwindcapacity:Germany,Spain,India,and

    Denmark.(TheUnitedStatesistheotherleadingcountryininstalledwindcapacity.)

    1.1. Worldwide Wind and Solar Capacity

    Windpowergenerationhasbeenrapidlygrowinginpowersystemsthroughouttheworld.

    Table2showsglobalwindenergygeneratingcapacityattheendof2006,aswellaswindcapacityadditionsin2006.AmajorityofthewindpowercapacityhasbeeninstalledinWestern

    Europe,specificallyinDenmark,GermanyandSpain;however,emergingwindenergy

    contributorsincludeIndia,Japan,andChina.Indeed,IndiasurpassedDenmarkin2005asthe

    fourthleadingcountryininstalledwindcapacity(GWEC2006).

    Worldwidesolarinstallationsarealsosurging,with1,460MWinstalledin2005(seeFigure1).

    Germanyaccountedfor837MWofthistotal,representing57%ofthemarket.Overall,installed

    solargeneratingcapacityexceeds5GWworldwide,andprojectionsarethatannualsolar

    installationswillincreasetobetween3,200MWand3,900MWby2010(Solarbuzz2006).

    Table3presents

    the

    twenty

    largest

    solar

    grid

    connected

    projects

    in

    the

    world.

    Of

    these

    twenty,

    onlyfourwereinstalledbefore2004.Largescalesolarthermalconcentratingprojectsare

    beginningtoappearaswell,withSpainplanning795MWofparabolictroughandpowertower

    projects(WesternGovernorsAssociation2006).

  • 7/31/2019 AA WindModel

    33/99

    23

    Table 2. Global wind energy capacity by country, 2006

    Country

    2006

    Capacity Additions

    (MW)

    2006 Total

    Installed Capacity

    (MW)

    Germany 2,233 20,622

    Spain 1,587 11,615Denmark 12 3,136Italy 417 2,123UK 634 1,963Portugal 694 1,716France 810 1,567Netherlands 356 1,560

    Austria 146 965Greece 173 746Ireland 250 745Sweden 62 572Norway 47 314Belgium 26 193Poland 69 153Other (1) 192 556Europe Total 7,708 48,545

    United States 2,454 11,603Canada 776 1,459North America 3,230 13,062

    India 1,840 6,270China 1,347 2,604Japan 333 1,394Taiwan 84 188

    South Korea 75 173Philippines 0 25Other (2) 0 13Asia 3,679 10,667

    Australia 109 817New Zealand 3 171Pacific Islands 0 12Total Pacific Region 112 1,000

    Brazil 208 237Mexico 85 88Costa Rica 3 74

    Caribbean (w/o Jamaica) 0 35Argentina 0 27Columbia 0 20Jamaica 0 20Other (3) 0 7Latin America 296 508

  • 7/31/2019 AA WindModel

    34/99

    24

    Table 2: Global wind energy capacity by country, 2006 (continued)

    Country

    2006

    Capacity Additions

    (MW)

    2006 Total

    Installed Capacity

    (MW)Egypt 85 230Morocco 60 124Iran 27 48Tunisia 0 20Other (4) 0 11Africa & Middle East 172 433

    World Total 15,197 74,215

    (1) Bulgaria, Croatia, Cyprus, Czech Republic, Estonia, Finland, Faroe Islands, Hungary, Iceland,Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Romania, Slovakia, Slovenia, Switzerland, Turkey,Ukraine.

    (2) Bangladesh, Indonesia, Sri Lanka, Russia;

    (3) Chile, Cuba, Mexico.

    (4) Cape Verde, Israel, Jordan, Nigeria, South Africa

    Source: Global Wind Energy Council Press Release. Global Wind Energy Markets Continue To Boom 2006 Another Record Year. February 2007. Available at http://www.gwec.net/uploads/media/07-02_PR_Global_Statistics_2006.pdf

    Figure 1: Worldwide PV installations in 2005 (MW)Source: 2006 World PV Industry Report Highlights: World Solar Market. Up 34% in2005; 837 MW Installed in Germany. Solarbuzz LLC, March 15, 2006. Available athttp://www.solarbuzz.com/Marketbuzz2006-intro.htm.

  • 7/31/2019 AA WindModel

    35/99

    25

    Table 3. Twenty largest grid-connected photovoltaic systems

    World

    Rank

    Project Location Size

    (MW)

    Date Installed

    1 Solarpark Pocking Pocking, Germany 10 April 2006

    2 Solarpark Muhlhausen Muhlhausen, Germany 6.3 December 2004

    3 Freiland SonnenStrom Miegersbach, Germany 5.27 Part 1, June 2005

    Part 2, December2005

    4 Burstadt Plant Burstadt, Germany 5 February 2005

    5 Solarpark Leipziger Land Espenhain, Germany 5 August 2004

    6 Springerville Generating

    Station

    Tuscon, Arizona, USA 4.59 2001-2004

    7 Solarpark Saarbrucken Saarbrucken, Germany 4 Part 1, June 2004

    Part 2, September

    2005

    Part 3, December

    2005

    8 Solarpark

    Geiseltalsee/Merseburg

    Geiseltalsee/Merseburg,

    Germany

    4 September 2004

    9 Solarpark Zeche Gottelborn

    (Part 1)

    Gottelborn, Germany 4 August 2004

    10 Solarpark Hemau Hemau, Germany 4 2003

    11 Fischers Family Warehouse Kronwieden/Dingolfing,

    Germany

    3.7 October 2005

    12 Michelin Reifenwerke KGaA Homburg, Germany 3.5 December 2004,

    expanded June

    2005

    13 Solarpark Penzing Penzing, Germany 3.45 December 2005

    14 Co.Muckenhausen roof

    mounted plant

    Dingolfing, Germany 3.3 October 2004

    15 Centrale di Serre Persano,

    ENEL research center

    Serre, Italy 3.3 1995

    16 Castejon power plant Castejon, Navarre, Spain 2.44 February 2006

    17 Solarpark Hofkirchen, part of

    Solarpark Donau

    Hofkirchen, Germany 2.37 August 2005

    18 Solaranlage Darast Nord Bad Gronenbach/Woringen,

    Germany

    2.3 November 2005

    19 Floriade exhibition hall PVSystem

    Vijfhuizen, Netherlands 2.3 April 2002

    20 Michelin Reifenwerke KGaA Bad Kreuznach, Germany 2.2 2005

    Source: Worlds Largest Photovoltaic Power Plants, pvresources.com. Accessed June 2006.

    Available at http:///www.pvresources.com/en/top50pv.php

  • 7/31/2019 AA WindModel

    36/99

    26

  • 7/31/2019 AA WindModel

    37/99

    27

    2.0 Wind Integration Studies in the United States andWorldwide

    ThischapterwillreviewthewindintegrationstudiesthathavebeenconductedintheUnited

    Statesandinvariouscountriesaroundtheworld.Thesestudiesoftenemphasizetheroleof

    ancillaryservices

    and

    the

    impact

    of

    wind

    power

    on

    the

    need

    for

    and

    availability

    of

    these

    services.WewillbeginbyexamininghowancillaryservicesaredefinedinEuropeandinthe

    UnitedStates.

    Electricpowersystemsneedavarietyofancillaryservicestomaintaingridoperationand

    reliability.Thereisnotgeneralagreementonhowtheseservicesaredefined,andasexplained

    furtherbelow,theUnitedStatesandEuropedefinetheseservicesdifferently.Evenwithinthe

    UnitedStates,theremaybedifferencesinwhatisconsideredancillaryservices.Ingeneral,

    though,thefollowingareconsiderednecessarytomaintainreliablegridoperation:

    RegulationMaintainingsystemfrequencythroughvaryingcertaingeneratingunits,

    typically

    with

    automatic

    generation

    control

    (AGC),

    up

    and

    down

    in

    response

    to

    very

    fast,unexpectedchangesinloadandgeneration.

    LoadFollowingRampinggenerationupordowntoreacttothechangeinexpectedload

    patterns,suchasincreasingloadsinthemorninganddecreasingloadslateintheday.

    SpinningReserveGeneratingcapacity,typicallysynchronizedtothegrid,thatcan

    maintainreliabilityifageneratingunitortransmissionlineistrippedoffline.

    SupplementalreservesThisperformsasimilarfunctiontospinningreserves,i.e.,

    maintainingreliabilityincaseofthelossofamajorgeneratingunitortransmissionline,

    butthegeneratorsprovidingthisservicearenotgenerallysynchronized(nonspinning)

    tothegridandmayneedadditionalstartuptimetocontribute.Insomeinstances,

    supplementalreserves

    may

    also

    replace

    spinning

    reserves

    after

    aperiod

    of

    time

    (Zavadil,etal.2006). Regulationandloadfollowingarereservesusedfornormal

    systemconditions,whilespinningandsupplementalreservesareusedforcontingency

    conditions.

    EuropeandtheUnitedStatesusedifferentterminologyindescribingthesevariousancillary

    services(Table4).InEurope,primaryreservesassistwiththeshortterm,minutetominute

    balancingandcontrolofthepowersystemfrequency,andareequivalentintheUnitedStatesto

    regulation.Primaryreservesmustbeavailablewithinsecondsandistypicallydoneby

    synchronousgeneratorsthatwillautomaticallyincreaseproductionwhenfrequencydropsor

    reduceproductionwhenfrequencyincreases,orfromloadthatcanbedroppedorreduced.

    Usually,the

    amount

    of

    primary

    reserve

    is

    defined

    by

    the

    largest

    power

    plant

    that

    can

    be

    lost

    whilemaintaininggridreliability.SecondaryreservesinEuropetakeoverforprimaryreserves

    10to30minuteslater,freeingupcapacitytobeusedasprimaryreserves.Sourcesforsecondary

    reservesincludequickstartgasturbines,pumpedstoragehydroprojectsandloadreductionor

    shedding.Likeprimaryreserves,secondaryreservesmayequalthelargestgeneratingunit,

    althoughafactormaybeaddedtoaccountforloadforecasterrors(HolttinenandHirvonen

  • 7/31/2019 AA WindModel

    38/99

    28

    2005).TheclosestterminologyintheUnitedStatesforsecondaryreservesiseitheroperating

    reservesorloadfollowingreserves,whichmayincludebothspinningandnonspinning

    components.LongertermreservesinEuropearecalledtertiaryreservesandareavailableinthe

    periodsaftersecondaryreserves.Tertiaryreservesareclosesttosupplementalreservesinthe

    UnitedStates,althoughthetimescalesmaybedifferentbetweenEuropeandtheUnitedStates.

    Theterms

    primary

    and

    secondary

    reserves

    will

    be

    used

    when

    describing

    the

    international

    experiencewithintegratingvariablerenewableenergygeneration.

    Inadditiontousingdifferentterminology,EuropeandtheUnitedStatesusedifferent

    frequenciesfortheelectricgrid.Europeoperatesat50HzandtheUnitedStatesoperatesat60

    HZ.

    Table 4. Reserve definitions in Germany, Ireland, and the United States

    Short-term

    reserves

    Medium-term

    Reserves

    Long-term

    reserves

    Germany Primary reserve:

    available within 30seconds, released by

    transmission system

    operator

    Secondary reserve:

    available within 5minutes, released

    by transmission

    system operator

    Minute reserve:

    available within 15minutes, called by

    transmission

    system operator

    from supplier

    n/a

    Ireland Primary operating

    reserve: available

    within 15 seconds

    (inertial response/

    fast response)

    Secondary

    operating reserve:

    operates over

    timeframe of 15-90

    seconds

    Tertiary response:

    from 90 seconds

    onwards (dynamic

    or static reserve)

    n/a

    United States Regulation horizon: 1

    minute to 1 hour with1- to 5-second

    Load-following horizons: 1 hour within

    increments 5- to 10 -minute increments(intra-hour) and several hours (inter-hour)

    Unit-

    commitmenthorizon: 1 day

    to 1 week with

    1-hour time

    increments

    Source: Gul, T. and Stenzel, T. 2005. Variability of Wind Power and Other Renewables: Management Optionsand Strategies. Paris: International Energy Agency.

    FourelectricallysynchronouszonesarepresentinEurope:theNordiccountries,theUCTE

    countries,GreatBritain,andIreland.

    TheNordic

    synchronous

    zone

    serves

    Finland,

    Sweden,

    Norway,

    and

    Eastern

    Denmark.

    Overall,25millionpeopleareserved,andabout90GWofgeneratingcapacityislocated

    inthiszone.Thetransmissionsystemoperatorshaveorganizedacooperativebody

    knownasNordelforadministeringtheNordicelectricitymarket.Totalprimarycontrol

    reserveis1,600MW,consistingofoperatingreservesof600MWandadisturbance

    reserveof1,000MW.

  • 7/31/2019 AA WindModel

    39/99

    29

    TheUCTEzoneservesabout500millionpeoplein23countries,withabout603GWof

    generatingcapacitylocatedinUCTE.ForUCTE,primaryreservesmustbeactivated

    within30secondsandcoverthelossofupto3,000MWofproduction.

    TheNationalGridCompanyisthegridoperatoroftheelectricitygridinEngland,Wales

    andScotland.About81GWofgeneratingcapacityislocatedinGreatBritain,with

    interconnectionsto

    France

    (2,000

    MW)

    and

    Northern

    Ireland

    (450

    MW)

    and

    requires

    reservestocoverthelossof1,320MW.

    TwoTSOs,theEirGridandtheSystemOperatorsNorthernIreland,administerthegrid

    inIreland,withageneratingcapacityof7,600MWa