Community-Scale Solar Plus Thermal Storage and Demand-Response

A Modeling Study of Local Grid Benefits

Community Solar Value Project September 2017

Santiago Sena, Public Service Company of New Mexico

Jon Hawkins, PNM Resources

Summary

WorkingwiththeteamfromtheCommunitySolarValueProject(CSVP),aprogramco-fundedbytheU.S.DOESunShotprogram,PublicServiceCompanyofNewMexico(PNM)hasexaminedtheideaofcombiningcommunity-scalesolarwithstorageanddemand-response(DR)measures.Thisstrategyhasbeendiscussedintheutilityindustry,asameansofcapturingnumerousbenefits,frommanagingsystempeakstoofferingcustomersmoreclean-energychoices,asrenewable-energypenetrationincreases.Anotherbenefitmightbetheopportunitytositesolarandsolar-plusmeasurestocapturegridbenefits.Thus,thisstudyexaminesanopportunitytomitigatelow-voltageissuesonafeedercomprisedprimarilyofresidentialloads.MightdistributedsolargenerationplusthermalstorageandDRbenefitthiscircuit,ormighttheycreatenewproblems?

PNMmodeledtheimpactonadistributionfeederofmultiplecombinationsofsolararraysofvaryingsizeswithcontrolofresidentialairconditionersandelectricwaterheaters.Ineachscenario,PNMevaluatedtheresultingimprovementinvoltagelevelsoneachdistributiontransformerduringtheperiodoflowvoltage(primarily,lateafternoonandearlyevening).WhileneitherDRalonenorasolararrayalonecouldmitigatealllow-voltageissues,thecombinationofamid-sizedsolararrayandsimpleDRcontrolstrategiesonthermalstorageeliminatedallinstancesoflowvoltage.ThemodelingapproachusedheremaybeextendedandappliedinotherstudiesatPNMorotherutilities,inevaluatingnovelcombinationsofrenewablesandloadflexibilitytoaddressdistributionissues.

1

Introduction

PNMisNewMexico’slargestelectricutility,servingover525,000customersinastatewithrelativelylowpopulationandalargeamountofnaturalresources.Giventhenaturalresourcesinthestate,itisnotsurprisingthatPNM’scustomershaveahighinterestinrenewableenergy.Infact,AlbuquerquewasnotedinarecentU.S.DepartmentofEnergySunShotSolarIndustryUpdateasbeinginthetop5citiesinthenationintermsofpercapitaPVinstalled.1Inaproactiveprojectthatbeganin2010andwascompletedin2011,PNMwasaveryearlypioneerinstudyinghowenergystoragemaybeabletoworktogetherwithdistributedrenewablestohelpimprovetheirvalue.Thateffort,calledtheProsperityEnergyStorageproject,hasbeendevelopedaspartoftheU.S.DepartmentofEnergy(U.S.DOE)SmartGridDemonstrationGrantsintheEnergyStorageProgram.

TheProsperityprojectwasdesignedandbuiltasa500-kW/1MWhutility-scalebatterystoragesystemthatinteracteddirectlywithaco-located500kWPVsolararray.Inthisproject,PNMresearcherslookedatmultiplevaluestreams.Onesuchvaluestreamisrelatedtosmoothingpoweroutput,inresponsetochangesinPVsystemoutpute.g.,fromintermittentcloudcover;ThesystemalsohasthecapabilityofstoringPVsystemoutputathighsolarproductiontimesforuseattimesthataremorealignedwithutilitysystempeakusagetimes.

Whilebatteryenergystorageisanextremelyflexibleresource,inmanymarketsitisstillacostlysolution.Thus,PNM’sexperiencewithProsperitypromptedaninvestigationintootherresourcesthatcouldcombinewithPVtoprovidesimilarbenefitsatalowercost.Inparticular,mighttherebealternativeresourcesforbalancingshortertermfluctuations,allowingthebatterysystemtofocusonitsenergyapplications,andhowsuchalternativeresourcescouldhelpmitigatesomeofthelocationalissuesthatarisewithdistributedgeneration.Aspartofthisinvestigation,PNMmodeledsomescenariosutilizingacommunity-scalesolar(CS)projectcoupledwithdemand-response(DR)resources.Thescenariocouldbeextendedtoimplementationthroughamodelthatpromotescommunitysolarwithcompanionmeasures(utility-sidestorageplusDR);however,thismodelingexercisewasfocusedonlyontechnicalimpacts.

PNMhadidentifiedafeederinitsserviceterritorythatalreadywasdisplayingvoltagesatcertaintimesofthedaythatwerebelowPNM’sdesigncriteria(ANSIRange-AVoltageLimits).Thelowvoltageswereidentifiedduringroutineplanningevaluationandtraditionalsystemimprovementapproacheswereimplementedtomitigatethelowvoltages.However,thegoalofthisanalysiswastoinvestigatethepossibilityofusingDRandCSresources,asopposedtothetraditionalapproachestomitigatelowvoltagesthatoccurasaresultofnormalloadgrowth.2TheanalysispresentedinthisreportassumesallavailableDRresourcesparticipatedintheprogram.TwoDRresourceswereevaluatedin1RobertMargolis,DavidFeldman,DanielBoff,“Q42016/Q12017SolarIndustryUpdate”,page33,U.S.DepartmentofEnergySunShot,April25,20172TheanalysisandconclusionshighlightedinthisreportwerenotfullyvettedthroughPNM’sDistributionPlanningDepartment.ThefindingsandconclusionsinthisreportarethoseoftheauthorsanddonotrepresenttheofficialpositionofPNM.

2

thisanalysis,controlledHVACunits(residentialairconditioning)andcontrolledelectrichotwaterheaters(EHWH).Thecircuitmodeledisprimarilyresidential,buttheseCSandDRmeasuresmayalsobetargetedtoacommercialmarket.

Asindicatedbelow,thescenarioutilizedforthisstudyrepresentsafirst-stepinthetechnicalassessmentofpossibleCSgridbenefits.Theresearchquestionsaresimplythese:

1)Wouldadistributioncircuitthatisexperiencinglowvoltagebeacandidateforasolar-plus-DR(withthermalstorage)strategy?

2)Ifasolar-plus-DRstrategyweresuggestedforacircuitexperiencingperiodiclow-voltagecharacteristics,isthereanidealconfiguration,intermsofsolar-projectscale?

3)Whatarethelikelysynergisticeffectsinthesituationdescribed,ofaddingoneormoreDRmeasures,plusthermalstorage,alongwiththeadditionofdistribution-scalesolargeneration?

Inpreparingthisanalysis,theauthorsworkedwiththeCommunitySolarValueProject(CSVP),whichisaU.S.DOESunShotawardee.CSVPhasworkedwithutilitiesonarangeofDRandstorageoptionsthatcouldbeusedascompanionmeasureswithcommunity-scalesolar,primarilyintherangeof500-kWto10MW.SomeoftheapproachesdiscussedinthisanalysisweredevelopedbasedinpartonCSVP’sworkwithotherutilities,andtheauthorshopetheseresultswillbeofusetoutilitiesconsideringsimilaroptions.Asnotedbelow,CSVPhasidentifiednumerousDRtechnologiesandcustomer-sidestorageoptionsthatarereadilyavailableandhavevarious,flexiblecontrolcharacteristics,aswellascustomer-appeals.CSVPalsohasrecommendedfullyutilizinginvertertechnologyandaddingstrategicsolar-designimprovements.Notalloftheseavailablehigh-valueoptionsaremodeledforthisstudy.Ineffect,thisstudyrepresentsa“firstgate”inthepathtoestablishingacommunity-scalesolar-plusstrategy

Thefollowingcasesweremodeled:

• BaseCaseo Demandresponseandcommunitysolarresourcesnotevaluated

• CommunitySolarOnlyo Onlycommunitysolarresourcesevaluated.o ThesizeofthePVgeneratorwasincreasedfrom1MW-9MWat1MW

increments.o Thelocationofthecommunitysolarsitewasstatic.

• DemandResponseOnlyo Onlydemandresponseresourcesevaluated.o OperationoftheHVACunitswereadvancedordeferredinaneffortto

mitigatevoltageissues.o TheoperationsoftheEHWHunitswereshiftedoutsidepeakloadingperiods.

• CommunitySolarplusDemandResponse

3

o Demandresponseresourcesevaluatedinconjunctionwithcommunitysolarresources

o TheoperationoftheHVACunitswereadvancedordeferredinanefforttomitigatelowvoltageissues.

o TheoperationsoftheEHWHweredeferredbetweenpeakloadingperiods.

DemandResponseResources

Company-sponsoredthirdpartyloadresearchidentifiedHVACsaturationat32.3%andEHWHsaturationat9.1%.ForthepurposesofthisanalysisitwasassumedallavailableHVACandEHWHunitsparticipatedintheDRprogram.Thisresultedin763customerswithHVACunitsand212customerswithEHWHunits.HVACandEHWHunitswererandomlyassignedtodistributiontransformersaccordingtothepreviouslyidentifiedsaturationlevels.HypotheticalloadprofilesfortheDRresourcewererandomlygeneratedand24-hourloadprofilesforresidentialloadsweredeveloped.ForthepurposeofthisanalysisDRloadswereshiftedfromon-peaktooff-peakperiods.ThetotalbaselineDRloadprofiledevelopedforthisanalysisishighlightedinFigure1.

Figure1-TotalDRBaselineLoadProfile

HVACUnits

MeteredcustomerHVACloaddatawasnotavailableforuseinthisanalysis.Therefore,baselinehypotheticalHVACloadprofilesweregenerated.AnexampleofageneratedHVACloadprofileishighlightedinError!Referencesourcenotfound..AllHVACunitsweresimulatedusinga5.6kWdemand.TheONdurationtimeforHVACunitswassetfor15minutesandtheOFFdurationtimewasrandomlygeneratedbetween30-90minutes.

4

Figure2-GeneratedHVACBaseline15MinuteLoadProfile

HVACloadswerecontrolledinoneoftwoways:(1)forthetimeintervalbetween12:00PM-6:30PMindividualHVACloadswereadvanced15minutescomparedtothebaselineprofile,ensuringdistributionvoltagesremainwithintheplanningcriteria,(2)fortimeintervalsbetween6:30PM-12AMindividualHVACloadsweredelayed15minutes,ensuringdistributionvoltagesremainwithintheplanningcriteria.Forexample,forthebaselineHVACloaddepictedinFigure2thecorrespondingcontrolledHVACloadprofileisidentifiedinFigure3.

Figure3-ControlledHVACLoadProfileExample

EHWHUnits

MeteredcustomerEHWHloaddatawerenotavailableforuseinthisanalysis.Therefore,baselinehypotheticalEHWHloadprofilesweregenerated.AnexampleofageneratedEHWHloadprofileishighlightedinFigure4.TheEHWHloadsweregeneratedusingademandof3.9kW.TheONdurationtimefortheEHWHwassetfor15minutesandtheOFFdurationtimewassetfor75minutes.TheinitialONoperationfortheEHWHwererandomlygenerated.FortheDRcontrolledcasesevaluatedinthisanalysisEHWHloads

5

werecompletelycurtailedbetweenthehoursof5:00PM–10PM.Forexample,forthebaselineEHWEloaddepictedinFigure4thecorrespondingcontrolledEHWHloadprofileisidentifiedinFigure5.

Notethattoday’sgrid-interactiveelectricwaterheatertechnologiescansupportamoreflexible,load-responsivecontrolscenariothanthosemodeledhere.Asafirst-stepanalysis,thisstudyreliedonbasictime-of-daytypecontrolstrategiesandassumedonlyexistingwaterheaters,retrofittedwithcontrolsthatdonotrequirecustomerstohaveadvancedelectronicmeters.Similarly,theHVACcontrolstrategyselectedforthismodelingeffortwasintendedtobereadilyachievable,usingexistingtechnologiesandsimplechangestoexistingprotocols.(SeeadditionalCSVPresourcesformoreonDRandcustomer-sidestorageoptions.)

Figure4-BaselineEHWHLoadProfile

Figure5-ControlledEHWHLoadProfileExample

FeederDistributionVoltage

Thecompany-acceptedcriteriaforappropriatedistributionvoltageistheAmericanNationalStandardInstitute(ANSI)StandardC84.1.Thestandardspecifiesvoltageranges

6

fornormal(Range-A)andemergency(RangeB)conditions.Range-Adefinestheservicevoltageasaminimumof114voltsandamaximumof126volts.

Amaximumvoltagedropof4%isthecompanydesignlimitfromthedistributiontransformerprimarytothecustomerservicemeter.Furthermore,PNMtypicallysetssubstationtransformersecondaryvoltagesat2.0%-2.5%above120voltsnominal.ApplyingthecompanyvoltagedropdesignlimitstotheANSIC84.1Range-Alimitsresultsinaminimumvoltagerequirementof118.7voltsattheprimarydistributiontransformerhighsidetoensureautilizationvoltageof114voltsatthecustomermeter.Furthermore,PNMgenerallydoesnotmodeldistributiontransformersandcustomerservicedrops.Forthepurposeofthisanalysis,distributiontransformersprimaryvoltagesbelow118.7voltsandabove126voltsareconsideredoutsideRange-A.

FeederModel

ThefeederusedinthisanalysisisidentifiedasHAMI13.Duringroutineplanningevaluation,theSynergifeedermodelidentifiedlowvoltagesbelowtheANSIC84.1Range-Alowerlimitduringpeakloadingconditions.ThelowvoltageswereconcentratedtoA-phaseandB-phaselateralslocatedattheendofthefeeder.TheSynergivoltageheatmapisidentifiedinFigure6.Mitigatingthelowvoltagesrequiredthefollowingsystemimprovements:

• Relocatinganoverhead1200kVARfixedcapacitor• Installinganewpad-mounted1800kVARswitchedcapacitorbank• TransferringasinglephaseloopfromtheB-phasetotheA-phase.

7

Figure6-SynergiHAMI13FeederModelVoltageProfile

OpenDSSModelDevelopment

TheSynergifeedermodelwasconvertedtoOpenDSSformatusingscriptsdevelopedinVisualBasicforApplication(VBA).Aone-to-oneconversionwasperformedbetweentheSynergifeedermodelandtheOpenDSSequivalentformat.AvisualcomparisonbetweenofthecircuitplotmodelsisshowninFigure7.PowerflowanalysiswasdoneusingOpenDSS,drivenbyVBAthroughtheCOMinterface.

8

Figure7-OpenDSSandSynergiFeederCircuitPlots

OpenDSSBaseModel

SupervisoryControlandDataAcquisition(SCADA)feederloaddatacorrespondingtothepeakloadingdaywasusedtodevelopresidentialloadprofiles.TheresidentialtransformerloadprofilesweredevelopedbytreatingtheEHWHandHVACunitsasspotloadsandallocatingSCADA15minutefeederloaddataaccordingtoconnectedKVA.ModelingtheDRloadasspotloadsensuredthatonlynon-deferrableloadswerecapturedintheallocationfactors.TheresultingOpenDSSmodelfeederloadprofile,withDRresourcesandnon-deferrableloadsidentifiedisshowninFigure8.

Figure8-OpenDSSFeederLoadProfile

ThebasecaseOpenDSSfeedermodeldevelopedforthisanalysisidentifiedmultiplesingle-phaselateralswithprimarydistributionvoltagesbelowtheANSIC84.1Range-Alimit.TheSynergiandOpenDSSdistributionvoltageprofileresultsareidentifiedinFigure9.Both

9

modelsidentifysystemiclowvoltage,clusteredonsinglephaselateralsattheendofthefeeder.

Figure9-Synergi&OpenDSSVoltageCircuitPlots

TheBaseCaseOpenDSSminimumtransformerprimaryvoltageandfeederloadprofilearegraphedinFigure10.Thefeederpeakof9.4MWoccurredat6:30PM.OpenDSSanalysisresultsidentifytransformerprimaryvoltagesbelowtheANSIC84.1Range-Alowerlimitbetweenthehoursor4:00PMand8:00PM.TheidentifiedtimeintervalalsocorrespondstothepeakfeederloadingintervalasidentifiedinFigure10.

10

Figure10-OpenDSSFeedLoadingandMinimumTransformerPrimaryVoltages

ThetotalnumberoftransformerswithprimaryvoltagesbelowtheANSIC84.1Range-Alowerlimitof118.7voltsishighlightedinFigure11.ThenumberoftransformerswithprimaryvoltagesbelowtheANSIC84.1Range-Alowerlimitspeaksat6:45PMwith58transformers.Themajorityoflowvoltagesoccurbetween5:15PM-7:30PM.ThelocationoftransformerswithvoltagesbelowRange-AareidentifiedinFigure12.Thetransformersarelocatedattheendofthefeederonsinglephaselaterals.

Figure11-BaseCase:TransformerswithPrimaryVoltagebelow118.7volts

11

Figure12-BaseCase:LocationofTransformerswithLowVoltages

Analysis

DemandResponseOnly

ControllingDRresourcesmarginallydecreasesthenumberoftransformerswithprimaryvoltagesbelowtheANSIC84.1Range-Alowerlimitof118.7volts.AcomparisonbetweenthenumbersoftransformerswithvoltagesbelowthedesigncriteriaforboththeDRcontrolledcaseandtheBaseCasesarehighlightedinFigure13.ControllingDRresourcesresultsinvoltagesabovethedesigncriteriaforinstancesbefore4:00PMandafter7:15PM.

Figure13-DRControlledCase:TransformerswithVoltagesbelow118.7Volts

12

TheminimumtransformervoltagesforthecontrolledanduncontrolledcasesarehighlightedinFigure14.ThelocationsoftransformerswithprimaryvoltagesbelowRange-AlowerlimitsareidentifiedinFigure15.ComparedtotheBaseCaseanalysisresults,controllingDRresourcesresultsinmitigatingalltransformerprimaryvoltageslocatedontheB-phaselaterattheendofthefeeder.

Figure14-MinimumTransformerPrimaryVoltage:ControlledDRvs.BaseCase

Figure15-ControlledDRCase:LocationofTransformerswithLowVoltage

CommunitySolarOnly

InthisanalysistheCSresourcewasmodeledusingaunitypowerfactorandassuminganidealPVgenerationoutputforboththeCSfacilityandexistingdistributivegeneration.Furthermore,theCSresourcewasmodeledusingastaticfacilitylocation.TheCSfacilitycapacitywasincrementedfrom1MWto9MWat1MWintervals.Considerable

13

improvementintransformerprimaryvoltagesresultinCSoutputsof3-6MW.However,transformerprimaryvoltagesremainbelowtheANSIC84.1Range-Alowerlimit.TheminimumtransformerprimaryvoltagesresultingfromvariousCSfacilityoutputsaregraphedinFigure16.

Figure16-CSOnlyCase:MinimumTransformerPrimaryVoltage

PeakCSoutputandpeakfeederloadingareoffsetby5hourswithpeakCSoutputoccurringat1:30PMandpeakloadingoccurringat6:30PM.ACSoutputof9MWresultsinthemajorityoftransformerprimaryvoltageswithintheANSIC84.1Range-Alowerlimits.However,theoutputofa9MWCSfacilityprovidesinsufficientvoltagesupporttomitigatelowvoltagesbetween7:00PM-8:00PM.Forvisualizationthe9MWCSoutput,feederloadprofileandremainingnumberoftransformerswithvoltagesbelowtheANSIC84.1Range-Avoltagelimitof118.7voltsaregraphedinFigure17andthelocationofthosetransformersareidentifiedonFigure18.Theremaininglowvoltagesareclusteredattheendofthefeeder.

Figure17-BaseCaseFeederProfilewith9MWCSOutput&AssociatedXFMRSbelowRange-A

14

Figure18-9MWCSCase:LocationofTransformerswithVoltagesbelow118.7volts

CommunitySolar&DemandResponse

AcombinationofCSandcontrolledDRresourcessignificantlyimprovedprimarytransformervoltages.TransformerprimaryvoltagesforvariouscasesofCSoutputsareshowninFigure19.UtilizingaCSoutputof4.5MWincombinationwithcontrollingDRresourcesensuredalltransformerprimaryvoltagesarewithintheANSIC84.1Range-Avoltagelimitrequirements.

Figure19-DR&CSCase:MinimumPrimaryTransformerVoltageProfile

15

AllDRloads,roughly1.2MW,areshiftedoffpeakwithaCSoutputof4.5MW.

Figure20–DR&CSCase:ResuktingDRLoadProfile

Conclusion

TheanalysisidentifiedthatutilizingacombinationofCSandDRresourcesimprovedvoltagessignificantly.ACSoutputof4.5MWandleveragingDRresourcesensuredallprimarytransformervoltageswerewithinplanningcriterialowerlimits.UtilizingCSresourcesexclusivelydidprovidemarginalimprovementsinfeedervoltageathighCSpenetrationlevelsbutdidnotensureallprimarytransformervoltageswerewithinplanningcriterialowerlimits.Furthermore,controllingDRresourcesexclusivelyresultedinmarginalimprovementinprimarytransformervoltagesbutdidnotcompletelyensureallprimarytransformersvoltageswerewithinplanningcriterialowerlimits.

ThisanalysisutilizedverysimpleDRcontrolalgorithmstoshiftHVACandEHWHloadsoutsideofpeakloadingperiods.InthisanalysisitwasassumedthatallavailableHVACandEHWHunitsparticipatedinthedemandresponseprogram.Furthermore,CSresourcesanddistributivegenerationweremodeledusingidealPVgenerationoutputs.

Analysisidentifiedthefollowingresults:

• AcombinationofDRandCSresourcessignificantlyimprovestransformerprimaryvoltagescomparedtothebasecase.

• UtilizingDRresourcesandaCSfacilityof4.5MWensuredalltransformerprimaryvoltageswereaboveRange-Alowerlimits.

• ACSfacilitywithanoutputof9MWmarginallydecreasesthenumberoftransformerswithlowvoltagesbelowRange-Alimitsbefore7:00PM.However,theoutputof9MWCSfacilityprovidesinsufficientvoltagesupporttomitigatelowvoltagesafter7:00PM.

16

• ControllingDRresourcesmarginallydecreasesthenumberoftransformerswithprimaryvoltagesbelowtheRange-Alowerlimit.ControllingDRresourcesresultsinvoltagesabovethedesigncriteriaforinstancesbefore4:00PMandafter7:15PM.

Thesimplicityofthecontrolalgorithmsusedinthemodelingisencouragingandmayindicatethatmoreautomatedandcoordinatedcontrolstrategiesusingintra-dayfeedbackcouldyieldimprovedresults,comparedtothedeterministictime-of-daycontrolstrategiesmodeledinthispaper.Thistypeofscenariosuggestsananalysisforfutureinvestigation.

WhiletheanalysisdidprovidesomeveryinterestingresultsregardingthecombinationofcommunitysolarplusDR,andtheircapabilitytoworkinconcerttomitigatefeederissues,thereisstilladoseofrealitythatmustbeconsidered.SomehypothesesaroundPVinstallationsbelievethatsolaralonecouldbeapossibleresourcetopossiblydeferfeederupgrades.Inthismodel,thesolarinstallationdidmitigatesomelowvoltageissues,butfellshortasthesunwassettingwhenwebegantoagainseethevoltageissuesweweretryingtomitigate.Alsodemandresponse,traditionallyusedasapeakingresourceforthebulkelectricsystem,onlydidamarginaljobonmitigatingvoltageissuesonthedistributionfeederitself.Additionalengineeringandsolar-plusmeasurescouldachieveagriddeferralinsomecases,butasthisanalysissuggests,thatwouldnotbeasimpleundertaking,atbest.

ThepurposeofthisstudywastoinvestigatethefeasibilityofutilizingCSandDRresourcestomitigatelowvoltagesresultingfromnormaldistributionfeederloadgrowth.Forthecircuitthatprovidedthebasisforthisstudy,actualmeasuretakentomitigatevoltageissuesincluded:

• Relocatinganexistingoverhead1200kVARcapacitorbank• Installinganewpad-mounted1800kVARcapacitorbank• Transferringasingle-phasefromtheB-phasetotheC-phase.

Asidefromtheinstallationofthe1800kVARvoltagecontrolledcapacitor,theidentifiedsystemimprovementsrepresentmarginable,inexpensivesolutions.Itwouldbehighlyunrealistictousethisproblemaloneasanargumenttoinstallamulti-milliondollarcommunitysolararrayandimplementacustomer-facingDRprogramwithtechnologyandprogramadministrationassociatedcosts.Additionally,whilethe4.5-MWarrayseemsidealforthisapplication,thereareseveralotherconsiderationsprerequisitetoanyconstructiondecision.Forexample,thesizeofthearraymight,fromatechnicalstandpoint,alsobedependentonthefeedertowhichitisinterconnected,andpossiblyevenonthehostingcapacityofthatfeeder.Landuseandbusinessfactors,marketing,andprogramdesignconsiderationsalsowouldcomeintoplaypriortoanysolardevelopmentdecision.

WouldaddingalargesolararrayandDRprogrambeacosteffectivesolutionforthevoltageimbalancesonthiscircuit?Inaword,no.However,theevidenceisclearthatasolar-plusstrategywouldnot,inthiscase,beatechnicalimpediment.Rather,itcouldsupporttheutility’scommitmenttosystemreliability.Itisimportanttolookatprogramssuchascommunitysolar,DRandenergyefficiencyholistically,andoutsidethetraditionaloperationalandplanningsilosthatexistatmanyutilities.

17

Relevantquestionsforfurtherconsiderationinclude:

! Doestheutilitydesiremorecommunity-scalesolar,inthe500-kW-to10-MWrange?

! Wouldacommunitysolarimplementationmodelhelpmovemorecustomersolarfromtherooftop(whereitisgenerallyinvisibleandnon-strategic)totheutility’spreferredscaleandlocations?

! Doestheutilityhaveenergyefficiencyorloadmanagementdesiresorrequirements?

! ArethereprogramobjectivesthatneedtobemetforStaterequirements?

! Doestheutilityhaveacleanelectrificationinitiative?TheintroductionofDRandstoragemeasures,suchasgrid-interactiveelectricwaterheaters,controlledHVAC,icestorage,andstrategicelectricvehicle(EV)programscouldsupportgreateruseofelectricity,inanageofrisingrenewable-energypenetration.

! Doestheutilityforeseehavingagreenhousegasmitigationplan?

! DoestheutilityIntegratedResourcePlan(IRP)callformoreDRorrenewableresources?

Alltheseareimportantconsiderationsinlookingforopportunitiestoimplementnewdistributedresourceprojectsandprograms,whilemaintainingreliabilityandoptimizationoftheutility’sdistributionsystem,toensurequalityservicetoallcustomers.

18

About the Authors Santiago Sena is a distribution system engineer with PNM in Albuquerque, NM. Jon Hawkins is manager of advanced technology and strategy for PNM Resources, and has participated with the Community Solar Value Project (CSVP) as a member of its Utility Forum, leading the exploration of opportunities to create greater net benefits from solar-plus storage and demand response measures. Jill Cliburn is the Principal Investigator of the CSVP. John Powers is the Project Officer of the CSVP. About the Community Solar Value Project The Community Solar Value Project aims to increase the scale, reach, and value of utility-based community solar programs by using strategic solar technologies, siting, and design, improving procurement strategies, target marketing, and integrating suitable companion measures, such as demand-response and storage into broad program designs. The project is led by the San Francisco-area energy consulting and analytics firm Extensible Energy, LLC, with support from Cliburn and Associates, LLC, Olivine, Inc., and Navigant Consulting. Utility participants include the Sacramento (California) Municipal Utility District (SMUD), Public Service of New Mexico, and other utilities nationwide. The project is powered by SunShot, under the Solar Market Pathways program of the U.S. Department of Energy. Contact: http://www.communitysolarvalueproject.com; info@communitysolarvalueproject.com Acknowledgment The information, data, or work presented herein was funded in part by the Office of Energy Efficiency and Renewable Energy (EERE), U.S. Department of Energy, under Award Number DE-EE0006905. Disclaimer This modeling project report a product of the Community Solar Value Project. As such, it is intended for review by project stakeholders and as a means to further the development of the technical and market strategies proposed herein. This work is not expressly endorsed or accepted by PNM. The information, data, or work presented herein was funded in part by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.