IRENA-非洲大陸電力系統(tǒng)規(guī)劃進(jìn)展 Advancements in continental power system planning for Africa 2024

Advancementsincontinental

powersystemplanningforAfrica

MethodologicalframeworkoftheAfricanContinentalPowerSystemsMasterplan’sSPLAT-CMPmodel2023

?IRENA2024

Unlessotherwisestated,materialinthispublicationmaybefreelyused,shared,copied,reproduced,printedand/orstored,providedthatappropriateacknowledgementisgivenofIRENAasthesourceandcopyrightholder.Materialinthispublicationthatisattributedtothirdpartiesmaybesubjecttoseparatetermsofuseandrestrictions,andappropriatepermissionsfromthesethirdpartiesmayneedtobesecuredbeforeanyuseofsuchmaterial.

ISBN:978-92-9260-605-3

CITATION

IRENA(2024),AdvancementsincontinentalpowersystemplanningforAfrica,InternationalRenewableEnergyAgency,AbuDhabi.

Availablefordownload:

/publications

.

Forfurtherinformationortoprovidefeedback,pleasecontactIRENAat

info@

.

ABOUTIRENA

TheInternationalRenewableEnergyAgency(IRENA)isanintergovernmentalorganisationthatsupportscountriesintheirtransitiontoasustainableenergyfuture,andservesastheprincipalplatformforinternationalco-operation,acentreofexcellence,andarepositoryofpolicy,technology,resourceandfinancialknowledgeonrenewableenergy.IRENApromotesthewidespreadadoptionandsustainableuseofallformsofrenewableenergy,includingbioenergy,geothermal,hydropower,ocean,solarandwindenergyinthepursuitofsustainabledevelopment,energyaccess,energysecurityandlow-carboneconomicgrowthandprosperity.

ACKNOWLEDGEMENTS

ThisreportwaspreparedbyIRENAincollaborationwiththeInternationalAtomicEnergyAgency(IAEA),undertheguidanceofAsamiMiketaandLarissaPinheiroPupoNogueira(IRENA),bythefollowingauthors:SebastianHendrikSterl(IRENAconsultant),BilalHussain(IRENA),MohamedBassamBenTicha(ex-IRENAconsultant),HimalayaBirShrestha(IRENA),andBrunoMerven(IRENAconsultant).

ThereportbenefitedfromthereviewsandcommentsprovidedbyDanielRusso,NolwaziKhumalo,PaulKomor,AthirNouicerandSeánCollins(IRENA),MarioTot(IAEA),TichakundaSimbiniandTonderayiGumunyu(AUDA-NEPAD),GeorgeGiannakidisandThyrsosHadjicostas(EUGTAF),AlessiaDeVita,GildasSigginiandEneSandraMacharm(GIZ/GET.Transform),MaximilianParzen(PyPSAmeetsEarth),DavideFioriti(PyPSAmeetsEarthandUniversitàdiPisa),MikaelTogebyandcolleagues(DanishEmbassyinEthiopia/EaEnergianalyse),IoannisPappis(SE4All),TomRemyandClaireNicolas(WorldBank),andTrieuMai,AmyRose,PatrickBrownandJarradWright(NREL).

PublicationsandcommunicationssupportwereprovidedbyFrancisField,StephanieClarkeandDariaGazzola.ThereportwaseditedbyJustinFrench-Brooks,withdesignprovidedbyPhoenixDesignAid.

DISCLAIMER

Thispublicationandthematerialhereinareprovided“asis”.AllreasonableprecautionshavebeentakenbyIRENAtoverifythereliabilityofthematerialinthispublication.However,neitherIRENAnoranyofitsofficials,agents,dataorotherthird-partycontentprovidersprovidesawarrantyofanykind,eitherexpressedorimplied,andtheyacceptnoresponsibilityorliabilityforanyconsequenceofuseofthepublicationormaterialherein.

TheinformationcontainedhereindoesnotnecessarilyrepresenttheviewsofallMembersofIRENA.ThementionofspecificcompaniesorcertainprojectsorproductsdoesnotimplythattheyareendorsedorrecommendedbyIRENAinpreferencetoothersofasimilarnaturethatarenotmentioned.ThedesignationsemployedandthepresentationofmaterialhereindonotimplytheexpressionofanyopiniononthepartofIRENAconcerningthelegalstatusofanyregion,country,territory,cityorareaorofitsauthorities,orconcerningthedelimitationoffrontiersorboundaries.

IRENA

InternationalRenewableEnergyAgency

Advancementsincontinental

powersystemplanningforAfrica

MethodologicalframeworkoftheAfricanContinentalPowerSystemsMasterplan’sSPLAT-CMPmodel2023

CONTENTS

ABBREVIATIONS 6

ABOUTTHISREPORT 8

1.THESPLAT-CMPMODELFORTHEAFRICANCONTINENTALPOWERSYSTEMS

MASTERPLAN 11

2.ELEMENTSOFSPLAT-CMPMODELDESIGN 13

2.1Powersystemdescription 13

2.2Modellingmethodologyupdates:Constraintsatthesystemandcountrylevel 16

2.3Modellingmethodologyupdates:Powergeneration 20

2.4Modellingmethodologyupdates:Cross-borderpowertransmission 28

2.5Modellingmethodologyupdates:Representationofstorage 31

2.6Summary 32

3.STRATEGIESFORRUNNINGSPLAT-CMPMODELS 34

4.MODELVERSIONCONTROL 39

5.POSSIBLEFUTUREAREASOFWORK/STRATEGIESFORIMPROVEMENT 41

6.CONCLUSION/OUTLOOK 43

DATAAVAILABILITYSTATEMENT 44

ACKNOWLEDGEMENTS 44

REFERENCES 45

APPENDIX 49

4|ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA

FIGURES

Figure1SchematicoverviewoftheRES(ReferenceEnergySystem)ofeachcountrynode

intheSPLAT-CMPmodel 14

Figure2ThetranslationofreservoirhydropowerdatafromtheAfREP-hydrodatabasetothe

SPLAT-CMPmodel 22

Figure3:GeospatialoverviewoftheareasforsolarPV,CSP,onshorewindandoffshore

windpowergenerationincludedintheSPLAT-CMPmodelthroughIRENA’s

MSRapproach 26

Figure4(a)Surgeimpedanceloading(inMW)asafunctionofinterconnectorvoltagelevel,

and(b)lineloadability(asafractionofsurgeimpedanceloading)asafunctionof

interconnectorlength 30

Figure5(a)GenericinterconnectorunitcostsasusedinSPLAT-CMP.(b)Efficiencyofgeneric

interconnectorsasusedinSPLAT-CMP,includinglinelossesandconverterlosses 30

Figure6GraphicalrepresentationofthevariousmodellingelementsoftheSPLAT-CMPmodel 33

Figure7TheCFpertimesliceofSouthAfrica’sonshorewindMSRcluster#2 36

Figure8Effectofspatialspreadonweakwindseasons 37

Figure9ScreenshotfromtheSPLAT-CMPrepositoryon 39

TABLES

Table1Overviewofreservemargincontributionsofthevarioustechnologytypesusedin

theSPLAT-CMPmodelfortheCMPprocess 17

Table2Translationofgenericinterconnectordistancecategoriestovoltagelevelcategories

asusedintheSPLAT-CMPmodelsetup 29

Table3Summaryofhowthenumberofvariables,constraintsandnon-zeromatrixelements

changewiththenumberoftimeslicesintheSPLAT-CMPmodelusedfortheCMP 34

BOXES

Box1AnalysesandtoolsinvolvedinthefirstCMPexercise 9

Box2ToolselectionconsiderationsoftheCMPstakeholders(AUDA-NEPAD,2023c) 10

Box3PotentialmethodstoinvestigatewindvariabilityaspectsintheSPLAT-CMPmodel

forfutureassessments 36

ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA|5

ABBREVIATIONS

AC

alternatingcurrent

ACEC

AfricanCleanEnergyCorridor

AfREP

AfricanRenewableElectricityProfiles

AfSEM

AfricanSingleElectricityMarket

AU

AfricanUnion

AUDA-NEPAD

AfricanUnionDevelopmentAgency–NewPartnershipforDevelopment

CAPEX

capitalexpenditure

CAPP

CentralAfricanPowerPool

CCGT

closed-cyclegasturbine

CF

capacityfactor

CMP

ContinentalMasterplan

COMELEC

ComitéMaghrébind’électricité

CSP

concentratedsolarpower

DNI

directnormalirradiation

EAPP

EasternAfricaPowerPool

EEZ

ExclusiveEconomicZone

EU

EuropeanUnion

EUGTAF

EuropeanUnionGlobalTechnicalAssistanceFacility

GHI

globalhorizontalirradiation

HFO

heavyfueloil

HVAC

high-voltagealternatingcurrent

HVDC

high-voltagedirectcurrent

IAEA

InternationalAtomicEnergyAgency

IIASA

InternationalInstituteforAppliedSystemsAnalysis

IRENA

InternationalRenewableEnergyAgency

kW

kilowatt

LCOE

levelisedcostofelectricity

LL

lineloadability

MESSAGE

ModelforEnergySupplyStrategyAlternativesandtheirGeneralEnvironmentalImpact

MSR

ModelSupplyRegion

6|ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA

MW

megawatt

MWh

megawatthour

NREL

NationalRenewableEnergyLaboratory(US)

OCGT

open-cyclegasturbine

OSeMOSYS

OpenSourceEnergyModellingSystem

PV

photovoltaics

RCP

RepresentativeConcentrationPathway

RoR

run-of-river

SAPP

SouthernAfricaPowerPool

SIL

surgeimpedanceloading

SM

solarmultiple

SNSP

systemnon-synchronouspenetration

SPLAT

SystemPlanningTest

SSP

SharedSocioeconomicPathway

TW

terawatt

UN

UnitedNations

USD

UnitedStatesdollar

VRE

variablerenewableenergy

WAPP

WestAfricanPowerPool

WETO

WorldEnergyTransitionsOutlook

WACC

weightedaveragecostofcapital

ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA|7

ABOUTTHISREPORT

ThisreportprovidesdetailedinsightsintothedesignandsetupoftheSystemPlanningTest-ContinentalMasterplan(SPLAT-CMP)model,theunderlyingoptimisationmodeloftheAfricaContinentalMasterPlan(CMP)forelectricitygenerationandtransmission.CMPisaninitiativetaskedtotheAfricanUnionDevelopmentAgency–NewPartnershipforDevelopment(AUDA-NEPAD)aspertherecommendationsadoptedbytheAfricanheadsofstateatthe34thOrdinarysessionoftheAssemblyofHeadsofStateandGovernmentoftheAfricanUnionin2021.Subsequently,inthe37thOrdinarysessionoftheAssembly,theCMPwasformallyadoptedasanAfricanUnionAgenda2063flagshipproject.

TheCMPinitiativewasundertakentosupporttheAUAfricanSingleElectricityMarket(AfSEM)initiativelaunchedinJune2021.TheAfricanpowerpoolsareactivelyinvolvedintheCMPprocesswithaviewtoaligningmodellingandplanningprocesseswithregionalpoliciesandinitiatives.AlongsidethedesignatedstaffofAUDA-NEPAD,theofficialCMPmodellingteamincludesrepresentativesfromWAPP(WestAfricanPowerPool),SAPP(SouthernAfricanPowerPool),EAPP(EasternAfricaPowerPool),CAPP(CentralAfricanPowerPool),COMELEC(ComitéMaghrébind’électricité)membercountries,andEU-GlobalTechnicalAssistanceFacility(EU-GTAF).

IRENAandtheIAEAsupporttheCMPinitiativeasofficiallyendorsedmodellingpartners.

ThefirstCMPexerciseentailedthreeplanninganalyses:(i)demandprojections,(ii)generationandcross-bordertransmissionexpansionoptimisation,and(iii)networkstudies(furtherexplainedinBox1).BasedonagreedtoolselectioncriteriafinalisedduringtheconsultationsthatformedpartoftheCMPprocess(Box2),thestakeholdersselectedIRENAandIAEA’sSPLAT-MESSAGEframeworkforthegenerationandcross-bordertransmissionexpansionanalysis.

Tothisend,IRENAcombineditspre-existingpowersystemcapacityexpansionmodelsfor50Africancountries,1theso-calledSystemPlanningTest(SPLAT)models,intooneSPLAT-AfricamodeltoserveasastartingpointforCMPmodelling.Overthecourseoftwoyears,theCMPmodellingteamreceivedhands-ontrainingfromthemodellingpartnersonthedevelopmentanduseoftheSPLAT-Africamodel.TheCMPprocessentailedmanyvirtualandin-personstakeholderengagementeventsandday-to-dayteamworkamongthestaffofAUDA-NEPAD,thepowerpoolsandNorthAfricancountryrepresentatives.ThisenabledtheCMPmodellingteamtoupdatetheSPLAT-Africamodelintotheso-calledSPLAT-CMPversion2023.Duringtheprocess,themodelwasequippedwithup-to-dateinformationcollectedon48Africancountries.2

ByprovidingadetaileddescriptionofthemethodologybehindtheSPLAT-CMPmodel,thisreportaimstoenhancethetransparencyandthereproducibilityofthemodellingresults.ItwillalsoenabletheCMPstakeholderstobetterunderstandthemodelresultsandinspireitsfutureuseinthenextCMPexercises.The

1AllAfricaexceptMadagascar,Mauritius,theComorosandtheSeychelles.

2AllSPLATcountriesexceptCaboVerde,andS?oToméandPríncipe.

8|ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA

reportalsocomplementstheeffortsbehindtherecentlydevelopedweb-baseduserguidefortheSPLATinterface,availableat

https://splat-tutorial.readthedocs.io/en/latest/.

TheSPLATinterfaceisanExcelbasedinterfacerequiredtoimplementtheSPLATmodellingframework,aswellastocontrolandvisualisethemodelcontentsandresultsrespectively.

Thefinalmodellingresults–aswellasthemodelinputsandtheirbasis,asfinalisedbytheCMPmodellingteamandagreedwiththebroadstakeholders–aredocumentedinaseparateseriesofreportspreparedwiththefinancialassistanceoftheEuropeanCommission(AUDA-NEPAD,2023a,2023b,2023c).Thereportseriesishostedonlineatthecentraliseddatabase“Mwanga”ofAUDA-NEPAD,developedspecificallytofacilitatedataexchangeamongCMPstakeholdersinthefuture(

/

).

Thisreportisstructuredasfollows:Section1providesanoverallintroductiontotheSPLATmodelframeworkandtherationaleforselectingitforCMPpurposes;Section2describesindetailallnewmodelelementsthatwereintroducedintheSPLAT-CMPmodel;Section3goesontodescribethepracticalitiesofandstrategiesforsettingtimeslicesforrunningtheSPLAT-CMPmodel,allowingforpracticalruntimeswhileretainingadequatespatio-temporalresolution;Section4describesmodelversioncontrol;Section5summarisespossiblestrategiesforfurthermodelimprovement;andSection6concludeswithafutureoutlook.

Box1AnalysesandtoolsinvolvedinthefirstCMPexercise

InthetechnicalplanningforavisionofacontinentallyintegratedAfricanpowergrid,variouselementsneededtobeconsideredseparatelywiththeirownsuitablemodelapproaches.Consequently,inthefirstCMPexercise,AUDA-NEPADwaschargedwithdevelopingthreedistinctpillarsofplanning:

(1)electricitydemandprojectionsforeverycontinentalAfricancountryupto2040

(2)long-termcapacityexpansionplanningscenariosforpowergenerationandcross-bordertransmissioninfrastructureupto2040tomeettheprojecteddemand

(3)networkstudiestoconfirmthetechnicalfeasibilityofthecross-borderinterconnectorsselectedinthecapacityexpansionoptimisationscenarios.

ThedemandassessmentanalysisperformedusingEViewsdefinedthefutureelectricitydemandprojectionsinvariousscenarios.Theseincludedabaselinetrajectoryasreferencecase(AUDA-NEPAD,2023d)andthreealternativescenarios(AUDA-NEPAD,2023e)reflectinglow,mediumandhighaspirationaltargetsonvariousdevelopmentaspects,suchaselectricityaccessratesandper-capitaincomegrowth.

ThepowersystemcapacityexpansionanalysisperformedusingtheSPLAT-MESSAGEframeworkidentifiedcost-optimalprospectsfornewgenerationcapacity,cross-borderinterconnectionandstorageassetsupto2040undervariousscenarios(AUDA-NEPAD,2023a,2023b).Asacapacityexpansionmodel,theSPLAT-CMPmodelmainlyassessedthecostimplicationsassociatedwiththebuildoutofthescenariocapacitieswithoutdeepdivingintotheengineeringaspectsofgridoperation.Nevertheless,certainimportantoperationalandengineeringaspectswereaccountedforinasimplifiedway,i.e.theplanningreserveconstraint(Section2.2.1),theupperboundoninstantpenetrationofvariablerenewables(Section2.2.2),andothers.

ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA|9

Box1AnalysesandtoolsinvolvedinthefirstCMPexercise(continued)

ThenetworkanalysisperformedusingthePSS?EtoolaimedtorefineandvalidatetheSPLAT-CMPresults.Itcarriedoutanengineeringanalysiscoveringsteady-statesystemstudiestoassesspotentialgridreliabilitygaps,suchasthevariouscriticalcontingenciesinvolvinglossofmajorgeneratorsortransmissionlinesacrossdifferentpartsofthecontinentalgridsystem(AUDA-NEPAD,2023f).Advancedstudies,suchastransientandfrequencystabilitystudies,wereoutsidethescopeoftheanalysis.

Inadditiontotheabovethreeanalyses,theCMPexercisealsodevelopedamethodologyforcost–benefit-basedrankingofthecandidategenerationandinterconnectionprojectsofkeyrelevanceinthemediumterm,being2023-2032(AUDA-NEPAD,2023g).Thisanalysiswasusedtoevaluatetheinternalrateofreturn(IRR),netpresentvalue(NPV),andotherfinancialmetrics/indicatorsoftheprojects.ThisanalysisdifferedfromtheSPLAT-CMPanalysisinthesensethatitevaluatedthepriorityoftheprojectsfromaninvestor’sperspectiveinsteadofthesystemvalueperspective.Thecost–benefitstudyaimedtoinformthenextstepsintheCMPprocess,whichwouldfocusonamulti-criteriarankingoftheprojectsfortheirfurtherintegrationintothevariousinfrastructurefacilitationmechanismsinprogressinAfrica.

Box2ToolselectionconsiderationsoftheCMPstakeholders(AUDA-NEPAD,2023c)

WhileselectingthevariousmodelstocarryouttheCMPexercise,theCMPstakeholderspaidparticularattentiontomaintaininguniformitywiththetoolsandmodelsalreadyutilisedand/orselectedforupdatingtheindividualpowerpoolmasterplans.Thedetailedcriteriaincludedthefollowingconsiderations:

i.Functionality:Isthemodelfitforpurposeforthejobtobeundertakenandcanthemodelhandleallthemajorissuesthatneedtobemodelled?

ii.Costoflicences:Isthecostofpotentiallicencesprohibitive,giventheneedfortrainingforagivennumberofpersonnelandtheneedforthemaintenanceofdatabases?

iii.Transparencyofthemodel:Isthemodellingmethodologysufficientlydocumented?Canitbeusedtounderstandtheexpectedoutcomesofthemodel?

iv.Reliabilityofthemodel:Canthemodelbeusedtoconfirm/validateknownoutcomesfrompastmodellingapproaches?Canitrunproperlywithoutnon-convergenceissues,andwouldtheresultsremaininsensitivetodifferentmodelversions?

v.User-friendlinessofthemodel:

–Candataeasilybeenteredinthemodel,includingbulkchanges,andcantheresultsbemadeavailableinanaccessibleformatthatcanbedirectlyusedinreportsorpresentations?

–Doesthemodelhavearobuststandardsetofplotsandtablestoallowtheresultstobereviewedinanaccessiblemanner?

–Doesthemodelproducereportsthatcanbeeasilycustomisedforvariousapplications?

–Doesthemodelhavea(graphical)interface?

10|ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA

1.THESPLAT-CMPMODELFOR

THEAFRICANCONTINENTAL

POWERSYSTEMSMASTERPLAN

TheSPLATmodelsarecountry-specificgenerationcapacityexpansionmodelscoveringtheentireAfricancontinentandareroutinelyusedbyIRENAfornationalandregionalcapacity-buildingprogrammesinAfrica

(IRENA,n.d.a),andthePlanningandProspectsforRenewablePowerinAfricaanalyses(IRENA,n.d.b).TheSPLATmodelswerecreatedutilisingamodelgeneratorcalledtheModelforEnergySupplyStrategyAlternativesandtheirGeneralEnvironmentalImpact(MESSAGE).MESSAGEisaversatile,adaptable,bottom-up,multi-yearenergysystemmodelthatutiliseslinearandmixed-integeroptimisationtechniques.ItwasfirstestablishedbytheInternationalInstituteofAppliedSystemAnalysis(IIASA),buthasbeensubsequentlyimprovedbytheInternationalAtomicEnergyAgency(IAEA).

Formodellingpowersystems,theMESSAGEsoftwareneedsaprojectionofdemand,adatabaseofcurrentcapitalassetsforpowergenerationandtransmission,technicalandeconomicspecificationsofpowersupplytechnologies,andalist(alongwithtechnicalandeconomicspecifications)offutureinvestmentoptionsforpowergenerationandtransmissionforthemodeltoinvestin.MESSAGEaimstominimisethenetpresentvalue(subjecttoauser-definedoveralldiscountrate)oftotalsystemcostsincurredtomeetauser-definedlevelofdemandoveragivenplanninghorizonandunderaspecificsetofassumptions.AMESSAGEuserguidecourtesyoftheIAEAispubliclyavailable(IAEA,2016).MESSAGEisafree-of-costtoolforIAEAmemberstatesandsupportsworkingwithfreeprogrammingsolvers.

Startingfromthepowerinfrastructureinachosenbaseyearinagivenregion,aMESSAGE-basedmodelcomputesaprogressionofpossibletechnologycombinationsperperiod(user-defined;typicallyyearly)thatcanachievethelowest-costobjectiveacrosstheplanningphasetomeetuser-defineddemandandconstraints.Theconstraintscanrepresenttechnicalconstraintslinkedtopowersystemoperation,butmayalsopolicyassumptionsthatreflectscenariostorylines.Whilethemodelcantechnicallyrunatfullhourlyresolution,itistypicallyrunusingtimeslice-basedapproachestoavoidunrealisticcomputationalrequirements(seeSection3formoredetails).Thesetimeslicesusuallyrepresent“common”dailyandseasonalvariabilitiesinpowergenerationanddemand.

Themodelproducesthelowesttotaldiscountedcost3ofthesystem,whichincludesinvestment,operationandmaintenance,fuel,andanyothercostsdefinedbytheuser(e.g.penaltiesongreenhousegasemissions;costsofunmetdemand).Additionally,itensuresthatallsystemrequirementsaremet,suchassufficientresourcesandcapacitytoachievedesiredproductionlevels,meetingauser-definedreservemargin,orstayingwithinuser-definedtechnologydeploymentlimits,aswellasanypolicyobjectivesthattheusermaydefine(e.g.renewableenergytargets,CO2emissionlimits,speedoftechnologyrollout).

3Discountingisaprocessofconvertingacostvalueprojectedtoincurinthefutureintoanequivalentcurrentvalue.IntheCMPmodel,thebaseyearof2019reflectsthe“current”,andthevariouscostsanticipatedtoincurindifferentyearsacrossthestudyperiodarebroughtto2019equivalentsassumingafixeddiscountratedescribedinSection2.3.4

ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA|11

TheSPLAT-CMPmodelwasdesignedinaversatileandflexiblewaytocontrolmodelinputsandoutputs,andallowconfigurationofrunswithanycombinationofAfricancountries.Theultimategoalwastorunscenariosatthecontinentallevelandexaminesuggestedinvestmentstrategiesinnewgenerationandcross-bordertransmissionprojectsundervariousscenarios.ThemodelwasestablishedbymergingpreviousversionsoftheSPLATmodelforindividualAfricanpowerpools,documentedinIRENA’sreportseriesonplanningandprospectsforrenewablepowerinAfrica:SPLAT-WfortheWestAfricanPowerPool(IRENA,2018),SPLAT-ACECfortheAfricanCleanEnergyCorridor,consistingoftheEasternandSouthernAfricanpowerpools

(IRENA,2021a),SPLAT-CfortheCentralAfricanPowerPool(IRENA,2021b),andSPLAT-NforNorthAfricaincludingCOMELECcountries(IRENA,2023b).

TheSPLAT-CMPmodelincludesanumberofupdatesascomparedtopreviousSPLATmodels,andsomestate-of-the-artmodellingfeaturesthatmaybenefittheenergymodellingcommunity.

12|ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA

2.ELEMENTSOFSPLAT-CMPMODELDESIGN

2.1POWERSYSTEMDESCRIPTION

InthespiritofpreviousSPLATmodels,theSPLAT-CMPmodelinitssimplestformcanberepresentedasfollows.Eachcountryisrepresentedasasinglenodecontainingdifferentenergy“l(fā)evels”,whichrepresentenergyattheresourcelevel(primary),energyconvertedintoelectricitybypowerplants(secondary),electricityatthetransmissionlevel(tertiary),andelectricityatthedistributionlevel(final).Theseenergylevels,alongwithvariousothermodelelements(seebelow),areschematicallyshowninFigure1.

Theprimarylevelrepresentstheavailabilityofresources(measuredintermsofprimaryenergysupply)ateachcountrynodelevel.Theavailabilityoftheseresourcesisdeterminedbothbydomesticresourcesaswellasacountry’scapacityforimports,i.e.byexogenousconstraints.Fuelpricesareexogenouslysetatthislevel.IntheSPLATmodels,asadefault,fuelpricesaredifferentiatedbyfueltypeandbylocalproductionversusimports,butnotdifferentiatedonacountrybasis(i.e.country-levelsubsidiesorotherpricedistortionsarenotreflectedinthemodel).

Thesecondarylevelisthemostimportantinthemodel,asitcontainsallthepowergenerationassetswithinacountrythatconvertaresource(primarylevel)intoelectricityreadytofeedintothetransmissiongrid(tertiarylevel).Itnotonlycontainsexistingpowergenerationassets,butalsocommittedassets(specificplantsthatarenotonlineyet,butwhichareconsideredascomingonlineinthefuturewithcompletecertaintyinaknownyear;thedefinitionofcommittedaccordingtotheCMPplanningprocessisaplantthathasreachedfinancialclose)aswellasallcandidatetechnologies,amongwhichthemodelcanchooseaselectiontocoverthefuturesupply–demandgap.Inadditiontouser-definedcandidatecapacity,SPLAT-CMPmodelentailspre-loadedgenericcandidatesupplyoptionsformostgenerationtechnologiestoallowtheusertodesigntheexpansionoptimisationaccordingtospecificscenariodefinitions.

Foralltechnologiesthatareduetocomeonlineinthefuture,aconstructiontime(numberofyears)isassumed,andtheoptimisationthusaccountsfortheconstructionstartingbeforetheyearofdeployment.Theinterestexpenseduringconstructioniscomputedbythemodelassumingthattheovernightcapitalexpenseisdistributedequallyduringtheconstructionperiod(includingtheyearinwhichthetechnologycomesonline).Theinterestexpensetogetherwiththeovernightcapitalexpenseofthetechnologyiseventuallyannualisedandtheresultingannualisedcostsincurringeachyearduringtheprojectlifeperiodarebroughttoacommonbaseyearbydiscounting(seeSection2.3.4formoredetailsondiscountrateparameter)forthepurposeofmakingoptimizationdecisions.Theconstructiontimesdifferbytechnology,representingthecomplexityoftheengineeringinvolvedindifferenttypesofpowerplants.Wenotethatonlyengineeringconstructiontimeswereconsidered,whereasinreality,planningandregulatoryapprovalstendtotake(substantially)moretime,especiallyfornuclearpowerplantsandlargehydropowerplants.

ADVANCEMENTSINCONTINENTALPOWERSYSTEMPLANNINGFORAFRICA|13

Figure1SchematicoverviewoftheRES(ReferenceEnergySystem)ofeachcountrynodein

theSPLAT-CMPmodel

Secondary

User-de?nedexisting,committedandcandidatesandpre-loadedgenericcandidates

Gas-?redplantsOCGT/CCGT**

Coal-?redplants**

HFO-?redplants*