Table Of Content

sustainability Article Restoration of an Active MV Distribution Grid with a Battery ESS: A Real Case Study MatteoManganelli1,*,MarioNicodemo1,LuigiD’Orazio2,LauraPimpinella2 andMariaCarmenFalvo1 1 DepartmentofAstronautical,ElectricalandEnergyEngineering,SapienzaUniversityofRome, Rome00184,Italy;[email protected](M.N.);[email protected](M.C.F.) 2 e-distribuzione,Rome00198,Italy;luigi.d’[email protected](L.D.); [email protected](L.P.) * Correspondence:[email protected];Tel.:+39-06-4458-5809 (cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1) (cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7) Received:26April2018;Accepted:15June2018;Published:17June2018 Abstract: In order to improve power system operation, Battery Energy Storage Systems (BESSs) have been installed in high voltage/medium voltage stations by Distribution System Operators (DSOs) around the world. Support for restoration of MV distribution networks after a blackout orHVinterruptionisamongthepossiblenewfunctionalitiesofBESSs. Withtheaimtoimprove qualityofservice,thepresentpaperinvestigateswhetheraBESS,installedintheHV/MVsubstation, canimprovetherestorationprocessindicatorsofadistributiongrid. Asacasestudy,anactualactive distributionnetworkofe-distribuzione,themainItalianDSO,hasbeenexplored. Theexistingnetwork islocatedincentralItaly. Itsuppliestwomunicipalitiesofapproximately10,000inhabitantsand includesrenewablegenerationplants. Severalconfigurationsareconsidered,basedon: thestateof thegridatblackouttime;theBESSstateofcharge;andtheinvolvementofDispersedGeneration (DG)intherestorationprocess. Threerestorationplans(RPs)havebeendefined,involvingtheBESS alone,orincoordinationwithDG.AMATLAB®/Simulink® programhasbeendesignedtosimulate therestorationprocessineachconfigurationandrestorationplan. TheresultsshowthattheBESS improvesrestorationprocessqualityindicatorsindifferentsimulatedconfigurations,allowingthe operationincontrolledislandmodeofpartsofdistributiongrids,duringinterruptionsorblackout conditions. The defined restoration plans set the priority and the sequence of controlled island operations of parts of the grid to ensure a safe and better restoration. In conclusion, the results demonstratethataBESScanbeavaluableelementtowardsanimprovedrestorationprocedure. Keywords: battery energy storage system; black start; dispersed generation; medium voltage distributiongrid;smartgrid 1. Introduction TheapplicationofBatteryEnergyStorageSystems(BESSs)inpowersystemshasincreasingly gained attention over the last years. The reasons encompass functions, services and benefits that theycanbringtopowersystemoperation[1–3],e.g.,peakshaving[4–7],voltagestability[8],market functions[9],frequencyresponse[10–13]. BESSsarealsorecognizedasameanstoincreasethelevelof security,reliability,flexibilityandqualityinpowersystems[14,15],especiallyindistributiongrids: atdistributionlevel,thespreadofDispersedGeneration(DG),basedonrenewableenergysources (RES),hasledtochangesthatarestillhardtoaccuratelyforecastandmanage. Inthisframework,theinstallationofBESSsinHighVoltage/MediumVoltage(HV/MV)stations has been recently regarded as a possible solution to many problems of the global power system Sustainability2018,10,2058;doi:10.3390/su10062058 www.mdpi.com/journal/sustainability Sustainability2018,10,2058 2of17 (transmissionanddistribution). ThishasbeenfavoredbymanyDistributionSystemOperators(DSOs), includinge-distribuzione,thelargestItalianDSObyextent[16–18]. An additional driver for this application by e-distribuzione has been financial funding by GovernmentsandtheEuropeanCommunityonpilotprojectsaimedattestingtherealperformanceof BESSsinsmartgrids[19,20]. Atthesametime,inItaly,electricitymarketrulesonthemanagementofthepowersystemsare evolving,accordingtodifferentpolicies[21,22];thisevolutioninvolvesinparticularDSOsandDG dispatching. Inaddition,agreatimpulseisgiventoinvestmentstowardssmartdistributionsystems; i.e., distributiongridswithahigherlevelofautomationandcontrolforguaranteeinghigherlevel of security, reliability, flexibility and resilience of network to improve the quality of service. The ItalianformerregulatoryAuthorityforelectricity,gasandwatersystem(AEEGSI),nowregulatory Authorityforenergy,networksandtheenvironment(ARERA)proposestheseimportantfeaturesfor developmentofasmartdistributionsystem[23]: • ObservabilityofpowerflowsandDG • VoltageregulationatMVlevel • Activepowerregulationatuserlevel • Remotecontrolforthepreventionofthe“unwantedisland”phenomenon • AdvancedoperationofMVgrids • TheuseofESSs Inthesamedocument,AEEG-SIdefinesthepossiblefunctionalitiesforBESSalreadyinstalledin HV/MVstationsforsmartgridpilotprojects: • Energyexchangeprofilelevelling(activepowercontrol)withtheHVnetwork,forthereduction andbettermanagementofthevariabilityoftheDG • Powerfactorcorrectionand/orthevoltagecontrol(reactivepowercontrol) • Back-upfunctionforshortinterruption • Black-startfunctionoflimitedportionsofgrid • Managementandcontrolofelectricvehicleschargingstations(ifownedbytheDSO) Thus,amongallthefunctionalitiesofBESSs,thereisthesupporttopowersystemrestorationafter ablackoutoraninterruptionofaHV/MVstation. Indeed,theBESScharging/dischargingflexibility canhelpkeepthebalancebetweenpowersupplyanddemandduringthesystemrestorationprocess, workingastheslackbusintheabsenceofthemaingrid. Given this background, the application of BESSs in power system restoration is worth investigation. In scientific literature, only few papers (e.g., [24–28]) deal with this specific issue, while a large amount of research work has been focused on BESS charging/discharging control, mitigating the intermittent power generation of renewable energy sources. In [29], the potential rolesofBESSsinpowersystemrestorationaredeeplydiscussedandadetailedcontrolstrategyfor theBESSsduringtherestorationperiodisalsodeveloped, whilementioningprosandconsofthe adoptedapproach. Restorationproblemsaregenerallyaddressedbyadoptingdifferentmathematical techniquesabletodefineareconfigurationoftheelectricgrid,whilesatisfyingitsconstraints. In[30], anoverviewofsuchtechniqueshasbeenexplained: fromdeterministicmathematicalprogramming, toheuristicsandknowledge-basedsystems. However,neitherdistributionsystemoperatorregulatory constraintsnorrealscaleapplicationoftheproposedmathematicaltechniqueshavebeentakeninto account,onthegroundsthatthereisnoexperienceonthematter. The focus of this paper is the study of the use of a BESS in HV/MV station of a real active distribution grid to support the restoration of the system after a blackout on the HV side or an interruption, accordingtotheactualregulationandtechnologiesavailableonthephysicalsystem. Asexplainedinthepaper,theanalyseshavebeencarriedoutwithreferencetoaspecificrealcase study: thepurposeistoverifyifrestorationnetworkplans,developedaccordingtotheactualrulesof Sustainability2018,10,2058 3of17 operationofdistributiongridsandavailablecontrolsresources,canensureamoresecureoperationof aSursetaailndabiislittryi 2b0u18ti, o1n0, xs yFsOtRe mPEaEnR dREtVoIiEmWp rovethequalityofservice(i.e.,reductionoftimeofoutage3a onf d16 numberofcustomersnotsuppliedduringtheHVinterruption). Thevalidationhasbeenperformedin performed in steady state operation, verifying power balance and compliance with voltage and steadystateoperation,verifyingpowerbalanceandcompliancewithvoltageandcurrentconstraints current constraints in all the components of the system (including BESS). Issues related to the inallthecomponentsofthesystem(includingBESS).Issuesrelatedtothetransientsandfrequency transients and frequency and voltage controls are neglected at this stage of the work. andvoltagecontrolsareneglectedatthisstageofthework. 22.. PPrreesseenntt OOppeerraattiioonn PPrroocceedduurree ooff DDiissttrriibbuuttiioonn GGrriiddss dduurriinngg aa BBllaacckkoouutt oorr IInntteerrrruuppttiioonn (w ithout B(wESitSh)out BESS) AAss iiss wweellll kknnoowwnn,, tthhee IIttaalliiaann MMVV ddiissttrriibbuuttiioonn ssyysstteemm iiss ccoonnnneecctteedd ttoo NNaattiioonnaall TTrraannssmmiissssiioonn SSyysstteemmv viiaaH HVV//MMVV ssttaattiioonnss oowwnneedd bbyy DDSSOOss.. TThhee ddiissttrriibbuuttiioonn nneettwwoorrkk ooppeerraatteess wwiitthh aa rraaddiiaall sscchheemmee,, aass sshhoowwnn iinn FFiigguurree 11.. FFiigguurree 11.. TTyyppiiccaall sscchheemmee ooff ccoonnnneeccttiioonn ttoo tthhee HHVV ssyysstteemm ooff aa ddiissttrriibbuuttiioonn ggrriidd iinn IIttaallyy.. During a perturbation or fault, in case of intervention of HV protection systems, it is possible to Duringaperturbationorfault,incaseofinterventionofHVprotectionsystems,itispossibleto have a consequent outage of one or more HV/MV stations. The interruptions or blackout due to an haveaconsequentoutageofoneormoreHV/MVstations. Theinterruptionsorblackoutduetoan event on the transmission system determines a contribution to the Energy Not Supplied (ENS) index, eventonthetransmissionsystemdeterminesacontributiontotheEnergyNotSupplied(ENS)index, defined as: definedas: EENNSS=PiPnt(T(Tiente−TTisnst)) ((11)) int int int whereP isthemeanpowersampledvalueofthelast15minatHVlevelatthetimeofinterruption where 𝑃int is the mean power sampled value of the last 15 min at HV level at the time of interruption (thisvalu𝑖e𝑛𝑡isassumedconstantforthedurationoftheevent);Te istheeventendingtime;Ts isthe (this value is assumed constant for the duration of the event); in𝑇t𝑒 is the event ending timein; t𝑇𝑠 is 𝑖𝑛𝑡 𝑖𝑛𝑡 eventstartingtimeofHVinterruptionorblackout. the event starting time of HV interruption or blackout. ConsideringtheItalianGridCodeonthecontinuityofsupply[31],theshareofENS,definedin Considering the Italian Grid Code on the continuity of supply [31], the share of ENS, defined in (1),contributingtothecontinuityofsupplyindex,hastobeevaluatedasanetvalueofEnergyNot (1), contributing to the continuity of supply index, has to be evaluated as a net value of Energy Not SuppliedENS givenby: Supplied 𝐸𝑁𝑆net given by: 𝑛𝑒𝑡 ENS = ENS−E (2) net restDS ENS ENSE (2) whereE istheamountofenergythatcannetberestoredbyrestthDSedistributionsystem.Thisiscalculated restDS aws:here 𝐸 is the amount of energy that can be restored by the distribution system. This is 𝑟𝑒𝑠𝑡 𝐷𝑆 Mcust calculated as: E = Pint ∑TOT[Nclusteri(Tclusteri−Tclusteri)] (3) restDS NTOT cust srest srest cPust i=M1TcOusTt whereNcTuOstT isthetotalnumEbreestrDoSfLVNaTinnOtdT MV[cuNscctuloustsmtereir(sTrscerleusssuttepriplieTdscrldeussutterriin)]g therestorationactivit(i3e)s : cust i1 where 𝑁𝑐𝑇𝑢𝑂𝑠𝑇𝑡 is the total number of LVN aTnOdT M=VM cu∑TclOussTttoerm(Nercslu rsetesrui)pplied during the restoration activiti(e4s): cust cust Mi=TcOl1uTster NTOT   (Nclusteri) (4) cust cust i1 where 𝑀𝑐𝑙𝑢𝑠𝑡𝑒𝑟 is the total number of cluster of customers restored, 𝑁𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑖 is the number of the 𝑇𝑂𝑇 𝑐𝑢𝑠𝑡 customers of i-th cluster of restoration, with a 𝑇𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑖 restoration starting time and 𝑇𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑖 𝑠 𝑟𝑒𝑠𝑡 𝑒 𝑟𝑒𝑠𝑡 restoration ending time; this is equal, for each cluster of customers, to the interruption ending time Sustainability2018,10,2058 4of17 where Mcluster is the total number of cluster of customers restored, Nclusteri is the number of the TOT cust customersofi-thclusterofrestoration,withaTclusteri restorationstartingtimeandTclusterirestoration srest erest enSdusitnaignatbiimlitye 2;0t1h8i,s 10is, xe FqOuRa lP,EfEoRr ReEaVcIhEWcl u sterofcustomers,totheinterruptionendingtimeTe . P4 oifs 16 int int definedin(1). Inthisway, thetermin(3)expressesanequivalenttimeofinterruption. Equations 𝑇𝑒 . 𝑃 is defined in (1). In this way, the term in (3) expresses an equivalent time of interruption. 𝑖𝑛𝑡 𝑖𝑛𝑡 (2)–(4)aregraphicallyillustratedinFigure2. Equations (2)–(4) are graphically illustrated in Figure 2. FiFgiugruer2e. 2P. oPwoweresrvss v.st.i mtimeseosf oifn itnetrerurrputpiotinonan adndre rsetostroartiaotnio:nE: NESNaSn adndE NESNnSetnect aclacluculaltaitoinon[3 [13]1.]. Due to the necessity of reducing the value of ENS, the same Grid Code in [31] foresees actions DuetothenecessityofreducingthevalueofENS,thesameGridCodein[31]foreseesactions based on an interaction and cooperation between TSO and DSO, e.g.: basedonaninteractionandcooperationbetweenTSOandDSO,e.g.:  The evaluation of the entity of system involved by the event (line, power plants, HV substation, • Theevaluationoftheentityofsysteminvolvedbytheevent(line,powerplants,HVsubstation, etc.); this action is a responsibility of the TSO etc.);thisactionisaresponsibilityoftheTSO  In case of fault, the localization of elements of HV network (line, pole, transformer, etc.); this • Incaseoffault,thelocalizationofelementsofHVnetwork(line,pole,transformer,etc.);thisaction action is a responsibility of TSO and DSO isaresponsibilityofTSOandDSO  The evaluation of the possibility of rearranging the HV network scheme, restoring the voltage • TheevaluationofthepossibilityofrearrangingtheHVnetworkscheme,restoringthevoltagefor for all plants and stations; this action is a responsibility of TSO and DSO allplantsandstations;thisactionisaresponsibilityofTSOandDSO If it is not possible to restore shortly (within some minutes) the voltage at the HV station, the TSOIf citanis rneoqtupiroes stihbele DtoSOre sttoo rreessthaorrt ttlhye( wMitVh innestowmoerkm iinnduetepse)ntdheenvtolylt.a Tgheea tmthaeinH aVctsiotantsio enx,etchueteTdS Oby catnher eDqSuOir eatrhe:e DSOtorestarttheMVnetworkindependently. ThemainactionsexecutedbytheDSO are:  Evaluation of network capacity to provide the restoring service, i.e., reserve of power on the • Evsaamluea tigornido,f cnoentwsiodrekricnagp atchiety mtoaprgroinv idoef tchuerrreesntto rrienfgersreerdv ictoe, it.hee., rtehseerrmvealo flipmoiwt eorfo tnhteh efeseadmeers; grfildex,icboinlistyid einr integrmthse omf arregminootef ccuornrternotl rseyfsetrermed otof tthhee tshaemrme aglrliidm, ictoonfstihdeerfienegd ethrse; flpeoxsisbibiliiltiytyi nto tearrmrasnogfer ae mneowte sccohnetmroel osfy dstiesmtriboufttihoen snaemtweogrrkid , consideringthepossibilitytoarrangeanew  scAhepmpleicoaftidoins torifb ruetsitoonrantieotwn oprlakns; this phase is carried out via the remote-control system • AMppolnicitaotrioinngo afnrde sstuopreartvioisniopnla onf sn;etthwisoprkh, ainse caissec aorfr eievdolouutitonv iianttoh ecrrietimcaol tseta-cteosn (torovlerslyosatdems, faults, etc.) • Monitoringandsupervisionofnetwork,incaseofevolutionintocriticalstates(overloads,faults, This operation condition increases the risk for the DSO to have an interruption of service, with etc.) relevant contribution to the value of System Average Interruption Duration Index (SAIDI) and System Average Interruption Frequency Index (SAIFI), the regulated indexes for the assessment of ThisoperationconditionincreasestheriskfortheDSOtohaveaninterruptionofservice,with continuity of the service for the distribution operator [32–34]. relevantcontributiontothevalueofSystemAverageInterruptionDurationIndex(SAIDI)andSystem As well known, the scheme of network operation, during supply restoration, is much different AverageInterruptionFrequencyIndex(SAIFI),theregulatedindexesfortheassessmentofcontinuity from the normal condition. The main factors to consider are: oftheserviceforthedistributionoperator[32–34].  ATshwee lplrkonboawbinli,tyth ethsacth eam faeuolft naertiwseosr, kdoupee rtaot itohne, dinucrrienagsisnugp polfy nreusmtobreart ioonf ,kisilmomuectherdsi ffoefr egnrtid fromthsuebntoernmdeadl ctoon tdhiet isoanm. Te hberemakaeinr factorstoconsiderare:  The level of damage in case of interruption, due to the increasing of number of customers • Theprobabilitythatafaultarises,duetotheincreasingofnumberofkilometersofgridsubtended subtended to the same breaker tothesamebreaker It is obvious that both mentioned parameters are higher with respect to the normal operating condition of the network, increasing the risk of a service interruption. In this framework, a BESS in MV can help the restoration procedure, acting as a slack bus between electricity supply and demand and thus increasing the flexibility of the grid: it can drive the controlled power grid, sustaining the consumption and allowing, whenever possible, the Sustainability2018,10,2058 5of17 • The level of damage in case of interruption, due to the increasing of number of customers subtendedtothesamebreaker Itisobviousthatbothmentionedparametersarehigherwithrespecttothenormaloperating conditionofthenetwork,increasingtheriskofaserviceinterruption. Inthisframework,aBESSinMVcanhelptherestorationprocedure,actingasaslackbusbetween electricitysupplyanddemandandthusincreasingtheflexibilityofthegrid: itcandrivethecontrolled power grid, sustaining the consumption and allowing, whenever possible, the synchronization of DG.Moreover,BESSsandDGs,possiblyoperateincoordinationwitheachother,contributetorisk mitigation. Asamatteroffact,inthetraditionalservicerestorationafaultoccurredonportionofthe networkdeterminesanoutageonboththerestoredandrestoringgrids;ontheotherhand,adoptinga BESSontheMVbus-bar,bothgridsaredecoupledandindependentofeachotherintermsoffault selectionandservicerestoration. Onthesegrounds,anactualsuitablecasestudyofe-distribuzione’sMVactivegridequippedwith a BESS has been selected, in order to quantitatively verify how the involvement of a BESS in the restorationprocedurecanimprovethequalityofservice. Thecasestudyherereportedisjustafirst exampleofapplication,inordertohavenumericalevidenceoftheproposalinarealspecificcase,butit canbegeneralizedinthefuture. Inaddition,itisworthtostressthattheanalysesdealwiththesteady stateoperationofthegrid,disregardinganyissuerelatedtotransients,frequencyandvoltagecontrol. 3. SimulationProcess The simulation of the Restoration Plans (RPs) was carried out via a MATLAB®/Simulink® program, developed for the purpose, consisting in a model of the network in steady-state and in aroutinefortheexecutionofmulti-periodPFA.Inaddition,aMATPOWERprogramwasusedfor comparisonforthevalidationofPFA[35].Thesimulationswerecarriedoutforthepurposeofverifying iftheRPsnetworkproposedareabletoensureasecureoperationofthedistributionsystem. The validation in steady state has been performed, verifying the power balance and compliance with voltage and current constraints in all system components (including BESS). Issues related to the transientsandtofrequencyandvoltagecontrolhavebeendisregardedatthisstage. ThedevelopedMATLAB®/Simulink®programislinkedtoaspreadsheet(e.g. MSExcel®)with networkdataandoperatesasfollows: • ImportsnetworkdatafromspreadsheetinMATLAB® • RunsaPowerFlowAnalysis(PFA)intheSimulink®networkmodelforeachtime-step(15min) • OutputstheresultstoMATLAB®andexportsthemtospreadsheet ForeachofthreeRPs,thevaluesandissuesassessedintheprocedureare: • Thetimeofblackout(t(cid:48)) • TheactivepowerdemandintheHV/MVstationatthetimeoftheblackout(P(cid:48) ) TOT • ThenumberofredLVandMVusers(N (i),N (i))withrespectiveactivepower(P (i), LVcust MVcust LVcust P (i))atstagei MVcust • ThestateofchargeoftheBESSindifferentstages(SOC (i)) BESS • Thepercentageofre-fednetwork,definedastheratioofthenumberofre-fedcustomerstothe totalnumberofcustomerstobere-fed: n(i) = Nc(ui)st ×100= NL(Vi)cust+NM(i)Vcust ×100; (5) % N N cust cust • ThepercentageofLVcustomersre-fedinanystageinrespectofthetotalnumberofLVcustomers tobere-fed: Sustainability2018,10,2058 6of17 (i) n(i) = NLVcust ×100; (6) LV% N LVcust • Thesequence,typeandsizeofDGpowerplantsreconnectedatdifferentstages • The power generated by the MV DG power plants at the stage i of reconnection, that can be (i) injectedinthedistributiongridP MV DG • Thecompliancewithvoltageandfrequencyconstraintsinthedifferentstages(V <V(i)<V , min MAX f <f(i)<f ) min MAX Sustainability 2018, 10, x FOR PEER REVIEW 6 of 16 Itisobviousthattheparameterslisted,especiallythosedefinedin(5)and(6),arestronglyrelatedto It is obvious that the parameters listed, especially those defined in (5) and (6), are strongly related theindicatorsofqualityofservicereportedinSection2andsotheycanbeassumedasindicatorsof to the indicators of quality of service reported in Section 2 and so they can be assumed as indicators theeffectivenessoftheproposedRP.AschematicofthesimulationprocessisreportedinFigure3. of the effectiveness of the proposed RP. A schematic of the simulation process is reported in Figure 3. Figure 3. Schematic of the simulation process. Figure3.Schematicofthesimulationprocess. The charge and discharge process of the BESS has been modeled, in steady state, by considering Tthhee acmhoaurgnte oafn ednedrgisyc hchaarrggeepd roorc deissschoafrtgheed BinE SevSehrya sstbaegee,n pmluos dtheele edn,eirngys tleoassdeys isnta etaec,hb sytacgoen, sviidae ring theamthoe uranttedof cehnaergrginygc/dhiasrcgheadrgoinrgd eifsfcichiaerngceyd. ineverystage,plustheenergylossesineachstage,viathe ratedcharging/dischargingefficiency. 4. Case Study 4. CaseStudy 4.1. MV Grid 4.1. MVGTrhide reference case study is an active MV grid of e-distribuzione, connected to the HV system and to the BESS, with the layout shown in Figure 4. ThereferencecasestudyisanactiveMVgridofe-distribuzione,connectedtotheHVsystemand totheBESS,withthelayoutshowninFigure4. Figure 4. Case study MV grid: HV/MV stations and MV lines. The HV/MV station includes one 16 MVA YYn transformer, with the neutral conductor on the MV side grounded via mobile Petersen coil. The MV bus-bar feeds 8 MV lines operated radially, in accordance with Italian distribution grid practices. Details on the composition of the MV network are reported in Tables 1 and 2. The 4 MV DG power plants are characterized in Table 3. Sustainability 2018, 10, x FOR PEER REVIEW 6 of 16 It is obvious that the parameters listed, especially those defined in (5) and (6), are strongly related to the indicators of quality of service reported in Section 2 and so they can be assumed as indicators of the effectiveness of the proposed RP. A schematic of the simulation process is reported in Figure 3. Figure 3. Schematic of the simulation process. The charge and discharge process of the BESS has been modeled, in steady state, by considering the amount of energy charged or discharged in every stage, plus the energy losses in each stage, via the rated charging/discharging efficiency. 4. Case Study 4.1. MV Grid Sustainability2018T,1h0e, r2e0f5e8rence case study is an active MV grid of e-distribuzione, connected to the HV system and 7of17 to the BESS, with the layout shown in Figure 4. Figure 4. Case study MV grid: HV/MV stations and MV lines. Figure4.CasestudyMVgrid:HV/MVstationsandMVlines. The HV/MV station includes one 16 MVA YYn transformer, with the neutral conductor on the TheHMVV/ sMideV gsrotauntidoend ivniac mluodbeilse Poenteers1e6n McoiVl. AThYe YMnV tbruasn-bsafro fremedes r8, MwVit hlintehs eopneeruatterda rlacdoianlldy,u inc toronthe accordance with Italian distribution grid practices. Details on the composition of the MV network are MVsidegroundedviamobilePetersencoil. TheMVbus-barfeeds8MVlinesoperatedradially,in reported in Tables 1 and 2. The 4 MV DG power plants are characterized in Table 3. accordancewithItaliandistributiongridpractices. DetailsonthecompositionoftheMVnetworkare reportedinTables1and2. The4MVDGpowerplantsarecharacterizedinTable3. Table1.MVNetworkComposition(lines). Line TotalLength NakedOverheadLine UndergroundCable OverheadCable (km) (km) (%) (km) (%) (km) (%) 1 0.09 0.00 0.0% 0.00 0.0% 0.09 100.0% 2 10.48 0.01 0.1% 10.15 96.9% 0.23 2.2% 3 17.88 1.27 7.1% 10.66 59.6% 5.95 33.3% 4 19.13 13.64 71.3% 2.87 15.0% 2.62 13.7% 5 25.21 19.51 77.4% 1.03 4.1% 4.67 18.5% 6 7.68 0.00 0.0% 0.00 0.0% 7.68 100.0% 7 8.43 0.00 0.0% 8.43 100.0% 0.00 0.0% 8 0.08 0.00 0.0% 0.00 0.0% 0.075 100.0% Total 88.98 34.43 33.14 21.32 Table2.MVNetworkComposition(CustomersandMV/LVStations). Line MV/LVStations LVCustomers MVCustomers 1 0 0 0 2 11 663 0 3 20 1228 0 4 13 1104 1 5 18 1312 2 6 9 1220 0 7 12 1139 0 8 0 0 0 Total 83 6666 3 Table3.DGPowerPlants. Line Type RatedPower(kVA) 4 PV 690 4 Hydro 3,000 5 PV 970 8(dedicated) Wind 4,000 Althoughthelinesareradiallyoperated,therearecommonnodesbetweenthedifferentlines. Thisisusefulformodifyingnetworkconfigurationincriticaloperatingconditions,e.g.,restoration afterablackout. Concerningnetworkautomation,theremote-controlsystemallowsthesupervision oftheMVnetworkandthe28nodesRemoteControlNodes(RCNs),inadditiontoeachlinecircuit Sustainability 2018, 10, x FOR PEER REVIEW 7 of 16 Table 1. MV Network Composition (lines). Line Total Length Naked Overhead Line Underground Cable Overhead Cable (km) (km) (%) (km) (%) (km) (%) 1 0.09 0.00 0.0% 0.00 0.0% 0.09 100.0% 2 10.48 0.01 0.1% 10.15 96.9% 0.23 2.2% 3 17.88 1.27 7.1% 10.66 59.6% 5.95 33.3% 4 19.13 13.64 71.3% 2.87 15.0% 2.62 13.7% 5 25.21 19.51 77.4% 1.03 4.1% 4.67 18.5% 6 7.68 0.00 0.0% 0.00 0.0% 7.68 100.0% 7 8.43 0.00 0.0% 8.43 100.0% 0.00 0.0% 8 0.08 0.00 0.0% 0.00 0.0% 0.075 100.0% Total 88.98 34.43 33.14 21.32 Table 2. MV Network Composition (Customers and MV/LV Stations). Line MV/LV Stations LV Customers MV Customers 1 0 0 0 2 11 663 0 3 20 1228 0 4 13 1104 1 5 18 1312 2 6 9 1220 0 7 12 1139 0 8 0 0 0 Total 83 6666 3 Table 3. DG Power Plants. Line Type Rated Power (kVA) 4 PV 690 4 Hydro 3,000 5 PV 970 8 (dedicated) Wind 4,000 Although the lines are radially operated, there are common nodes between the different lines. Sustainability2018,10,2058 8of17 This is useful for modifying network configuration in critical operating conditions, e.g., restoration after a blackout. Concerning network automation, the remote-control system allows the supervision of the MV network and the 28 nodes Remote Control Nodes (RCNs), in addition to each line circuit breakerontheMVbus-bar. Figure5showstheRCNineachMVlineofthecasestudy;0denotesthe breaker on the MV bus-bar. Figure 5 shows the RCN in each MV line of the case study; 0 denotes the RCNcorrespondingtothelinecircuitbreakersattheheadoflineonMVbus-barside. RCN corresponding to the line circuit breakers at the head of line on MV bus-bar side. Sustainability 2018, 10, x FOR PEER REVIEW 8 of 16 Figure 5. Remote control nodes for each MV line of the case study. Figure5.RemotecontrolnodesforeachMVlineofthecasestudy. The definition of the RP is based on the location of RCNs for each line, the number of MV/LV Thestdateiofinnsi tainodn MoVf tchuestoRmPeirss, bthaesier davoenragthe eacltoivcea atinodn reoafctRivCe Nposwfeorr deeamcahnldi.n e,thenumberofMV/LV stationsandMVcustomers,theiraverageactiveandreactivepowerdemand. 4.2. BESS and Control System 4.2. BESSandA CLoitnhtiruoml-SIoynst BemESS is installed in the HV/MV station and connected to the MV bus-bar via a dedicated line (Line 1). A schematic of the BESS control system and an image of the installation are ALpitrhesieunmted-I oinn FBigEuSreS 6i.s Tihnes stpaelcleifdicaitniotnhse ofH thVe/ EMSSV arset: ationandconnectedtotheMVbus-barviaa dedicatedline(Line1). AschematicoftheBESScontrolsystemandanimageoftheinstallationare  BESS rated capacity: 2 MWh presentedinFigure6. ThespecificationsoftheESSare:  BESS rated power: 2 MVA  Battery module rated voltage: 15.2 V • BESSratedcapacity: 2MWh  Battery module rated capacity: 0.625 kWh • BESS raTteodtalp eoffwicieern:c2y:M 80V%A • Batte ryImnsotadlluatlieonra: tCeodntvaoinletarsg (e1:0 1c5o.n2taViners, 8 power converter systems, 3264 batteries), 400 m² • Batte ryPmroovdiduelres:r Saytestdemca inptaecgirtayto:r0 a.6n2d5 bkatWterhies: NEC  Technology: Li-Ion (Lithium metal oxide) • Totalefficiency: 80%  Life: 4,000 cycles, 10 years • Insta llatMioanxi:mCuomn taauixnileirasry( 1co0ncsounmtpatiinoenr: s7,887.7p5o kwWehr/dcoayn vertersystems,3264batteries),400m2 • Providers: Systemintegratorandbatteries:NEC • Technology: Li-Ion(Lithiummetaloxide) • Life: 4,000cycles,10years • Maximumauxiliaryconsumption: 787.75kWh/day Figure 6. Schematic of the BESS, based on [36] (left panel), and image of the installation (right panel). To clarify how the proposed procedure can be then implemented in the physical system, a general description of the control architecture available in the e-distribuzione MV grid (including the BESS in HV/MV stations), is detailed. Sustainability 2018, 10, x FOR PEER REVIEW 8 of 16 Figure 5. Remote control nodes for each MV line of the case study. The definition of the RP is based on the location of RCNs for each line, the number of MV/LV stations and MV customers, their average active and reactive power demand. 4.2. BESS and Control System A Lithium-Ion BESS is installed in the HV/MV station and connected to the MV bus-bar via a dedicated line (Line 1). A schematic of the BESS control system and an image of the installation are presented in Figure 6. The specifications of the ESS are:  BESS rated capacity: 2 MWh  BESS rated power: 2 MVA  Battery module rated voltage: 15.2 V  Battery module rated capacity: 0.625 kWh  Total efficiency: 80%  Installation: Containers (10 containers, 8 power converter systems, 3264 batteries), 400 m²  Providers: System integrator and batteries: NEC Sustainability2018, 10,T20ec5h8nology: Li-Ion (Lithium metal oxide) 9of17  Life: 4,000 cycles, 10 years  Maximum auxiliary consumption: 787.75 kWh/day Figure 6. Schematic of the BESS, based on [36] (left panel), and image of the installation (right panel). Figure6.SchematicoftheBESS,basedon[36](leftpanel),andimageoftheinstallation(rightpanel). To clarify how the proposed procedure can be then implemented in the physical system, a general description of the control architecture available in the e-distribuzione MV grid (including the ToclarifyhBEoSwS int hHeV/pMrVo sptaotisoends), pisr doectaeildedu. r ecanbethenimplementedinthephysicalsystem,ageneral descriptionofthecontrolarchitectureavailableinthee-distribuzioneMVgrid(includingtheBESSin HV/MVstations),isdetailed. e-distribuzione MV grid management and control is in charge of 28 control centers; each one is responsibleforseveralactivities,e.g.,:qualityofservicecertification,managementofcrewsinthefield, network and equipment predictive maintenance, and cooperation with the TSO in order to prevent blackoutsandemergencies.ThesystemisequippedwithaSupervisoryControlandDataAcquisition (SCADA), located in the control centers, which communicates with Remote Terminal Units (RTUs), Sustainability 2018, 10, x FOR PEER REVIEW 9 of 16 locatedintheHV/MVandMV/LVstations.IncaseoffaultontheHVgrid,whentheTSOrequiresthe DSOtorestoresee-rdvistirciebuozinonet hMeVM grVid mneantwagoemrkenitn adnde pcoenntrdole ins tilny ,chtahregeS oCf 2A8 DcoAntrsoyl csetnetmers;i seaachb loenet ois identifyand responsible for several activities, e.g.,: quality of service certification, management of crews in the isolatetheMVnetworkportiontobere-supplied.Consequently,theSCADAcanalsosendtheBESSa field, network and equipment predictive maintenance, and cooperation with the TSO in order to black-startremproevteenct obmlacmkoauntsd a,nmd eamkeinrggenscuierse. Tthhea tsyasltlemth ies esqeuciuprpietdy wchithe cak Ssuhpearvveisobrey eCnonpterorlf oanrdm Deadtac orrectly.The communicatiAoncquairscithiointe (cStCuArDeAis), bloacsaetedd oinn ththe ecoCntlrioeln cte/nSteersr,v werhimch ocdomelm.uTnhiceatBesE wSiSthl oRceamlocteo nTetrrmolilnearl assumesthe Units (RTUs), located in the HV/MV and MV/LV stations. In case of fault on the HV grid, when the serverrolewhile,ontheDSOside,thereisanRTU,thatassumestheclientroleandisremotelyconnected TSO requires the DSO to restore service on the MV network independently, the SCADA system is withthecontraobllec teo nidteenrtitfhy aatndh iossotlastet hthee DMSVO neStwCoArkD pAort.ioTnh teo bine froe-rsmupaptliieodn. Ceoxncsheqaunegnetldy, btheet SwCeAeDnAc claienn tandserver iscompliantawlsoit shen[d3 t7h]e. BAESsS as hbloacwk-nstairnt rFemigouter ceom7,mianndo, rmdaekrintgo suerxe ethcaut taell taheb sleaccukri-tys tcahretckos phaevrea btieoenn andsupply performed correctly. The communication architecture is based on the Client/Server model. The BESS powertoatargetload,theBESSlocalcontrollermustreceivevoltagecommands(V )foreachstagefrom local controller assumes the server role while, on the DSO side, there is an RTU, that assestumes the theSCADAtcoliecnot nrotrleo lanthd eisp roeimnotteolyf dcoenlnivecetreyd vwoitlht athgee sc.onUtrpolo ncenrteecr etihvaitn hgosVtss etthce oDmSOm SaCnAdDsAf.r oTmhe theSCADA, theBESSdeteinrfmorminaetisont hexechtaanrggeedt bpeotwineetno cflidenetl iavnde rsyervveor litsa cgoembplyiacnat lwciuthl a[3ti7n]. gAtsh sheodwinf fienr Feingucree b7,e itnw eentheV set order to execute a black-start operation and supply power to a target load, the BESS local controller values from the SCADA and the voltage values measured at point of delivery under the conditions must receive voltage commands (Vset) for each stage from the SCADA to control the point of delivery thatfrequencvyolitsagaesc. oUnpsotna rnectevivainlug eVsoet fco5m0mHanzdas nfrdomz ethreo S-sCeAqDuAe, nthcee BvEoSSlt adegteermisinaelsw thaey tsarsgeett ptooinzte orfo .Inorderto avoidinrushdceulirvreeryn tvofllotawge ibnyt ocatlchuelaltoinagd t,heth deifBfeEreSnSce pbreotwveiedne sthae sVosetf tvasltuaerst ffruomn ctthieo nSC,AwDhAic ahndin tchree asesvoltage voltage values measured at point of delivery under the conditions that frequency is a constant value graduallyaccordingtoasoftstartperiod(e.g.,from1sto20s). Inthiswork,thesystemisstudiedin of 50 Hz and zero-sequence voltage is always set to zero. In order to avoid inrush current flow into stationarystatthee; ltohade,r tehfeo BreE,SSit pirsovaisdseus ma seodft stthaartt ftuhnectivono,l twahgicehi isnccroenassetsa vnotl,taigme pgroadseudallyb yactchorediBngE StoS a, operatingas thevoltageresgofut slatatrot rp.eriod (e.g., from 1 s to 20 s). In this work, the system is studied in stationary state; therefore, it is assumed that the voltage is constant, imposed by the BESS, operating as the voltage regulator. Figure 7. Black-start control scheme for the BESS. Figure7.Black-startcontrolschemefortheBESS. Hence, it is clear that the proposed procedure can be implemented in the physical system, both fully automated and manually guided by the system operator. 4.3. Simulation Scenarios and Hypotheses For a given grid layout and control system, a RP of the MV grid is a function of the initial operation scenario, in terms of:  The state of the network in the time of blackout, i.e., amount of active power demanded by the loads and amount of power supplied by DG, which define the aggregate equivalent load at the MV bus-bar Sustainability2018,10,2058 10of17 Hence,itisclearthattheproposedprocedurecanbeimplementedinthephysicalsystem,both fullyautomatedandmanuallyguidedbythesystemoperator. 4.3. SimulationScenariosandHypotheses Foragivengridlayoutandcontrolsystem,aRPoftheMVgridisafunctionoftheinitialoperation scenario,intermsof: • Thestateofthenetworkinthetimeofblackout,i.e.,amountofactivepowerdemandedbythe loadsandamountofpowersuppliedbyDG,whichdefinetheaggregateequivalentloadatthe MVbus-bar • TheStateofCharge(SOC)oftheBESSatthetimeoftheblackout Sustainability 2018, 10, x FOR PEER REVIEW 10 of 16 Thus,t heTfihres Sttsattee pof tCohabrugeil (dSOaCR) Pof itshet oBEaSsSs uatm thee ttihmee bolfa tchke obluactktoimut eandtheBESSSOC. For the caTsheuss, tthued fyir,sto spteepr taot ibounildd aa RtPa ifso tor atshsuremee rtheef ebrlaecnkcoeut dtiamyes an(wd teheek BdESaSy ,SOSCat. urday and Sunday) inJanuaryweFreor athvea cilaaseb sletu.dFy,o orpeeraactihono dfattha efomr ,thtrheee raefgegrernecge adtaeyse (qwueievkadlaeyn, Staltouarddaya tantdh eSuMndVayb) iuns -barwas calculatedJvaniuaaaryP wFeAre; tahveailraebsleu. lFtsora eraechre opf othrteemd, tihne Faigggurergeat8e, ewquhivearleentth leoasdig ant tohfe aMctVi vbeusp-boawr werasi spositive calculated via a PFA; the results are reported in Figure 8, where the sign of active power is positive whenpowerflowsfromtheHVsidetotheMVside(passivedistributionnetwork). when power flows from the HV side to the MV side (passive distribution network). Figure 8. Daily net active power profile at the MV bus-bar for three reference days. Figure8.DailynetactivepowerprofileattheMVbus-barforthreereferencedays. Comparing the profile of the three days, it is evident that the distribution grid is passive only in some hours of the weekday, due to the combination of a greater load and a minor DG production. Comparingtheprofileofthethreedays,itisevidentthatthedistributiongridispassiveonly During the two weekend days, the grid behaves mostly as an equivalent generator in respect of the insomehoursoftheweekday,duetothecombinationofagreaterloadandaminorDGproduction. HV system. DuringthetwOonw theeeske egnrodunddasy, tsh,et hcheoigcre ihdasb beeheanv toe sbumilod sutply RaPs aassnumeqinugi tvhae lweneetkgdeany ears athtoe rreifnerreenscpe ectofthe HVsystemc.ase, and check its feasibility in the other two days, when the net load at the MV bus-bar is mostly negative, due to a lower passive load and a higher generation. Onthesegrounds,thechoicehasbeentobuildupRPassumingtheweekdayasthereference With reference to the weekday, two times of blackout have been considered: maximum positive case,andcheckitsfeasibilityintheothertwodays,whenthenetloadattheMVbus-barismostly net load time (2:30 p.m.) and the maximum negative net load time (6:30 p.m.), as shown in Figure 7. negative,duetAonaothloerw heyrpoptahsessiisv deealolsa wditahn thde aSOhCig ohf eBrEgSSe nate trhaet tiiomne. of blackout. Two level of SOC have Withrbeeefenr ceonncsiedetoredth: 1e00w%e aenkdd 5a0y%,.t wotimesofblackouthavebeenconsidered: maximumpositive Based on the location of RCNs, different RPs were developed, divided in 10 stages. For each stage: netloadtime(2:30p.m.) andthemaximumnegativenetloadtime(6:30p.m.),asshowninFigure7. Anoth erhTyhep loistth oef soipsednienagl sanwd ictlhositnhge oSpOeraCtioonfs BwEasS Sseta wttithhe retsipmecet otof tbhela vcakrioouust .liTnews olevelofSOChave  The number of LV and MV customers and DG power plants to be reconnected were defined, beenconsidered: 100%and50%. assuming a time interval of 30 s and 300 s, respectively BasedonthelocationofRCNs,differentRPsweredeveloped,dividedin10stages. Foreachstage: The RPs were firstly simulated, assuming the absence of MV DG, in order to verify restoration • Thelicsatpoafbioliptieesn oinf tghea snydstecmlo ussiningg oonpley rtahtei oinntesrnwala DsSsOe tsowuirtches,r ei.se.p, ewcitthtooutt hloecavl aDrGio. uThsel isnamese RPs were then checked, involving the MV DG plants, in reference to a smart management of the grid. The • The number of LV and MV customers and DG power plants to be reconnected were defined, rationale is then to shift the connection of distributed generation as much as possible, to take assumaidnvganatatgiem oef tihne tsetorrvaagle osyfs3te0ms, eavnedntu3a0l0ly sc,ornesisdpeericntgi vneo ldyistributed generation contribution (only energy storage contribution) in the restoration process. TheRPswTheer reesfiurltsst olybtsaiinmedu floart tehde ,tharseseu RmPsi,n bguiltth uep ainb rseefenrcenecoe fofM thVe wDeeGkd,ainy, oforrd theer ttwoov sceerniafyriorse storation capabilitieosf opfotshitieves ymsatxeimmuums ianngd noenglaytivthe emianxitmerunma lneDt SloOads, owuitrhcoeust, ain.ed. ,wwithit hDoGu, tanlodc ianl rDefeGre.nTceh eof sameRPs different initial SOC of BESS, are reported. werethenchecked,involvingtheMVDGplants,inreferencetoasmartmanagementofthegrid. The Table 4 summarizes the list of RCN operation for each stage in each line for the RP1. The table is based on RP1. RPs differ according to the stage of reconnection of MV DG, i.e., RCNs 3 and 6 of line 4 are closed during Stage 3, 4 or 5, in RP1, RP2 or RP3, respectively.

Description:
Akhil, A.A.; Huff, G.; Currier, A.B.; Kaun, B.C.; Rastler, D.M.; Chen, S.B.; Cotter, A.L.; Bradshaw, D.T.;. Gauntlett . Licensee MDPI, Basel, Switzerland.
ebook img

Restoration of an Active MV Distribution Grid with a Battery ESS PDF

17 Pages·2017·5 MB·English
by  
Save to my drive
Quick download
Download
Upgrade Premium
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Restoration of an Active MV Distribution Grid with a Battery ESS

sustainability Article Restoration of an Active MV Distribution Grid with a Battery ESS: A Real Case Study MatteoManganelli1,*,MarioNicodemo1,LuigiD’Orazio2,LauraPimpinella2 andMariaCarmenFalvo1 1 DepartmentofAstronautical,ElectricalandEnergyEngineering,SapienzaUniversityofRome, Rome00184,Italy;[email protected](M.N.);[email protected](M.C.F.) 2 e-distribuzione,Rome00198,Italy;luigi.d’[email protected](L.D.); [email protected](L.P.) * Correspondence:[email protected];Tel.:+39-06-4458-5809 (cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1) (cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7) Received:26April2018;Accepted:15June2018;Published:17June2018 Abstract: In order to improve power system operation, Battery Energy Storage Systems (BESSs) have been installed in high voltage/medium voltage stations by Distribution System Operators (DSOs) around the world. Support for restoration of MV distribution networks after a blackout orHVinterruptionisamongthepossiblenewfunctionalitiesofBESSs. Withtheaimtoimprove qualityofservice,thepresentpaperinvestigateswhetheraBESS,installedintheHV/MVsubstation, canimprovetherestorationprocessindicatorsofadistributiongrid. Asacasestudy,anactualactive distributionnetworkofe-distribuzione,themainItalianDSO,hasbeenexplored. Theexistingnetwork islocatedincentralItaly. Itsuppliestwomunicipalitiesofapproximately10,000inhabitantsand includesrenewablegenerationplants. Severalconfigurationsareconsidered,basedon: thestateof thegridatblackouttime;theBESSstateofcharge;andtheinvolvementofDispersedGeneration (DG)intherestorationprocess. Threerestorationplans(RPs)havebeendefined,involvingtheBESS alone,orincoordinationwithDG.AMATLAB®/Simulink® programhasbeendesignedtosimulate therestorationprocessineachconfigurationandrestorationplan. TheresultsshowthattheBESS improvesrestorationprocessqualityindicatorsindifferentsimulatedconfigurations,allowingthe operationincontrolledislandmodeofpartsofdistributiongrids,duringinterruptionsorblackout conditions. The defined restoration plans set the priority and the sequence of controlled island operations of parts of the grid to ensure a safe and better restoration. In conclusion, the results demonstratethataBESScanbeavaluableelementtowardsanimprovedrestorationprocedure. Keywords: battery energy storage system; black start; dispersed generation; medium voltage distributiongrid;smartgrid 1. Introduction TheapplicationofBatteryEnergyStorageSystems(BESSs)inpowersystemshasincreasingly gained attention over the last years. The reasons encompass functions, services and benefits that theycanbringtopowersystemoperation[1–3],e.g.,peakshaving[4–7],voltagestability[8],market functions[9],frequencyresponse[10–13]. BESSsarealsorecognizedasameanstoincreasethelevelof security,reliability,flexibilityandqualityinpowersystems[14,15],especiallyindistributiongrids: atdistributionlevel,thespreadofDispersedGeneration(DG),basedonrenewableenergysources (RES),hasledtochangesthatarestillhardtoaccuratelyforecastandmanage. Inthisframework,theinstallationofBESSsinHighVoltage/MediumVoltage(HV/MV)stations has been recently regarded as a possible solution to many problems of the global power system Sustainability2018,10,2058;doi:10.3390/su10062058 www.mdpi.com/journal/sustainability Sustainability2018,10,2058 2of17 (transmissionanddistribution). ThishasbeenfavoredbymanyDistributionSystemOperators(DSOs), includinge-distribuzione,thelargestItalianDSObyextent[16–18]. An additional driver for this application by e-distribuzione has been financial funding by GovernmentsandtheEuropeanCommunityonpilotprojectsaimedattestingtherealperformanceof BESSsinsmartgrids[19,20]. Atthesametime,inItaly,electricitymarketrulesonthemanagementofthepowersystemsare evolving,accordingtodifferentpolicies[21,22];thisevolutioninvolvesinparticularDSOsandDG dispatching. Inaddition,agreatimpulseisgiventoinvestmentstowardssmartdistributionsystems; i.e., distributiongridswithahigherlevelofautomationandcontrolforguaranteeinghigherlevel of security, reliability, flexibility and resilience of network to improve the quality of service. The ItalianformerregulatoryAuthorityforelectricity,gasandwatersystem(AEEGSI),nowregulatory Authorityforenergy,networksandtheenvironment(ARERA)proposestheseimportantfeaturesfor developmentofasmartdistributionsystem[23]: • ObservabilityofpowerflowsandDG • VoltageregulationatMVlevel • Activepowerregulationatuserlevel • Remotecontrolforthepreventionofthe“unwantedisland”phenomenon • AdvancedoperationofMVgrids • TheuseofESSs Inthesamedocument,AEEG-SIdefinesthepossiblefunctionalitiesforBESSalreadyinstalledin HV/MVstationsforsmartgridpilotprojects: • Energyexchangeprofilelevelling(activepowercontrol)withtheHVnetwork,forthereduction andbettermanagementofthevariabilityoftheDG • Powerfactorcorrectionand/orthevoltagecontrol(reactivepowercontrol) • Back-upfunctionforshortinterruption • Black-startfunctionoflimitedportionsofgrid • Managementandcontrolofelectricvehicleschargingstations(ifownedbytheDSO) Thus,amongallthefunctionalitiesofBESSs,thereisthesupporttopowersystemrestorationafter ablackoutoraninterruptionofaHV/MVstation. Indeed,theBESScharging/dischargingflexibility canhelpkeepthebalancebetweenpowersupplyanddemandduringthesystemrestorationprocess, workingastheslackbusintheabsenceofthemaingrid. Given this background, the application of BESSs in power system restoration is worth investigation. In scientific literature, only few papers (e.g., [24–28]) deal with this specific issue, while a large amount of research work has been focused on BESS charging/discharging control, mitigating the intermittent power generation of renewable energy sources. In [29], the potential rolesofBESSsinpowersystemrestorationaredeeplydiscussedandadetailedcontrolstrategyfor theBESSsduringtherestorationperiodisalsodeveloped, whilementioningprosandconsofthe adoptedapproach. Restorationproblemsaregenerallyaddressedbyadoptingdifferentmathematical techniquesabletodefineareconfigurationoftheelectricgrid,whilesatisfyingitsconstraints. In[30], anoverviewofsuchtechniqueshasbeenexplained: fromdeterministicmathematicalprogramming, toheuristicsandknowledge-basedsystems. However,neitherdistributionsystemoperatorregulatory constraintsnorrealscaleapplicationoftheproposedmathematicaltechniqueshavebeentakeninto account,onthegroundsthatthereisnoexperienceonthematter. The focus of this paper is the study of the use of a BESS in HV/MV station of a real active distribution grid to support the restoration of the system after a blackout on the HV side or an interruption, accordingtotheactualregulationandtechnologiesavailableonthephysicalsystem. Asexplainedinthepaper,theanalyseshavebeencarriedoutwithreferencetoaspecificrealcase study: thepurposeistoverifyifrestorationnetworkplans,developedaccordingtotheactualrulesof Sustainability2018,10,2058 3of17 operationofdistributiongridsandavailablecontrolsresources,canensureamoresecureoperationof aSursetaailndabiislittryi 2b0u18ti, o1n0, xs yFsOtRe mPEaEnR dREtVoIiEmWp rovethequalityofservice(i.e.,reductionoftimeofoutage3a onf d16 numberofcustomersnotsuppliedduringtheHVinterruption). Thevalidationhasbeenperformedin performed in steady state operation, verifying power balance and compliance with voltage and steadystateoperation,verifyingpowerbalanceandcompliancewithvoltageandcurrentconstraints current constraints in all the components of the system (including BESS). Issues related to the inallthecomponentsofthesystem(includingBESS).Issuesrelatedtothetransientsandfrequency transients and frequency and voltage controls are neglected at this stage of the work. andvoltagecontrolsareneglectedatthisstageofthework. 22.. PPrreesseenntt OOppeerraattiioonn PPrroocceedduurree ooff DDiissttrriibbuuttiioonn GGrriiddss dduurriinngg aa BBllaacckkoouutt oorr IInntteerrrruuppttiioonn (w ithout B(wESitSh)out BESS) AAss iiss wweellll kknnoowwnn,, tthhee IIttaalliiaann MMVV ddiissttrriibbuuttiioonn ssyysstteemm iiss ccoonnnneecctteedd ttoo NNaattiioonnaall TTrraannssmmiissssiioonn SSyysstteemmv viiaaH HVV//MMVV ssttaattiioonnss oowwnneedd bbyy DDSSOOss.. TThhee ddiissttrriibbuuttiioonn nneettwwoorrkk ooppeerraatteess wwiitthh aa rraaddiiaall sscchheemmee,, aass sshhoowwnn iinn FFiigguurree 11.. FFiigguurree 11.. TTyyppiiccaall sscchheemmee ooff ccoonnnneeccttiioonn ttoo tthhee HHVV ssyysstteemm ooff aa ddiissttrriibbuuttiioonn ggrriidd iinn IIttaallyy.. During a perturbation or fault, in case of intervention of HV protection systems, it is possible to Duringaperturbationorfault,incaseofinterventionofHVprotectionsystems,itispossibleto have a consequent outage of one or more HV/MV stations. The interruptions or blackout due to an haveaconsequentoutageofoneormoreHV/MVstations. Theinterruptionsorblackoutduetoan event on the transmission system determines a contribution to the Energy Not Supplied (ENS) index, eventonthetransmissionsystemdeterminesacontributiontotheEnergyNotSupplied(ENS)index, defined as: definedas: EENNSS=PiPnt(T(Tiente−TTisnst)) ((11)) int int int whereP isthemeanpowersampledvalueofthelast15minatHVlevelatthetimeofinterruption where 𝑃int is the mean power sampled value of the last 15 min at HV level at the time of interruption (thisvalu𝑖e𝑛𝑡isassumedconstantforthedurationoftheevent);Te istheeventendingtime;Ts isthe (this value is assumed constant for the duration of the event); in𝑇t𝑒 is the event ending timein; t𝑇𝑠 is 𝑖𝑛𝑡 𝑖𝑛𝑡 eventstartingtimeofHVinterruptionorblackout. the event starting time of HV interruption or blackout. ConsideringtheItalianGridCodeonthecontinuityofsupply[31],theshareofENS,definedin Considering the Italian Grid Code on the continuity of supply [31], the share of ENS, defined in (1),contributingtothecontinuityofsupplyindex,hastobeevaluatedasanetvalueofEnergyNot (1), contributing to the continuity of supply index, has to be evaluated as a net value of Energy Not SuppliedENS givenby: Supplied 𝐸𝑁𝑆net given by: 𝑛𝑒𝑡 ENS = ENS−E (2) net restDS ENS ENSE (2) whereE istheamountofenergythatcannetberestoredbyrestthDSedistributionsystem.Thisiscalculated restDS aws:here 𝐸 is the amount of energy that can be restored by the distribution system. This is 𝑟𝑒𝑠𝑡 𝐷𝑆 Mcust calculated as: E = Pint ∑TOT[Nclusteri(Tclusteri−Tclusteri)] (3) restDS NTOT cust srest srest cPust i=M1TcOusTt whereNcTuOstT isthetotalnumEbreestrDoSfLVNaTinnOtdT MV[cuNscctuloustsmtereir(sTrscerleusssuttepriplieTdscrldeussutterriin)]g therestorationactivit(i3e)s : cust i1 where 𝑁𝑐𝑇𝑢𝑂𝑠𝑇𝑡 is the total number of LVN aTnOdT M=VM cu∑TclOussTttoerm(Nercslu rsetesrui)pplied during the restoration activiti(e4s): cust cust Mi=TcOl1uTster NTOT   (Nclusteri) (4) cust cust i1 where 𝑀𝑐𝑙𝑢𝑠𝑡𝑒𝑟 is the total number of cluster of customers restored, 𝑁𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑖 is the number of the 𝑇𝑂𝑇 𝑐𝑢𝑠𝑡 customers of i-th cluster of restoration, with a 𝑇𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑖 restoration starting time and 𝑇𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑖 𝑠 𝑟𝑒𝑠𝑡 𝑒 𝑟𝑒𝑠𝑡 restoration ending time; this is equal, for each cluster of customers, to the interruption ending time Sustainability2018,10,2058 4of17 where Mcluster is the total number of cluster of customers restored, Nclusteri is the number of the TOT cust customersofi-thclusterofrestoration,withaTclusteri restorationstartingtimeandTclusterirestoration srest erest enSdusitnaignatbiimlitye 2;0t1h8i,s 10is, xe FqOuRa lP,EfEoRr ReEaVcIhEWcl u sterofcustomers,totheinterruptionendingtimeTe . P4 oifs 16 int int definedin(1). Inthisway, thetermin(3)expressesanequivalenttimeofinterruption. Equations 𝑇𝑒 . 𝑃 is defined in (1). In this way, the term in (3) expresses an equivalent time of interruption. 𝑖𝑛𝑡 𝑖𝑛𝑡 (2)–(4)aregraphicallyillustratedinFigure2. Equations (2)–(4) are graphically illustrated in Figure 2. FiFgiugruer2e. 2P. oPwoweresrvss v.st.i mtimeseosf oifn itnetrerurrputpiotinonan adndre rsetostroartiaotnio:nE: NESNaSn adndE NESNnSetnect aclacluculaltaitoinon[3 [13]1.]. Due to the necessity of reducing the value of ENS, the same Grid Code in [31] foresees actions DuetothenecessityofreducingthevalueofENS,thesameGridCodein[31]foreseesactions based on an interaction and cooperation between TSO and DSO, e.g.: basedonaninteractionandcooperationbetweenTSOandDSO,e.g.:  The evaluation of the entity of system involved by the event (line, power plants, HV substation, • Theevaluationoftheentityofsysteminvolvedbytheevent(line,powerplants,HVsubstation, etc.); this action is a responsibility of the TSO etc.);thisactionisaresponsibilityoftheTSO  In case of fault, the localization of elements of HV network (line, pole, transformer, etc.); this • Incaseoffault,thelocalizationofelementsofHVnetwork(line,pole,transformer,etc.);thisaction action is a responsibility of TSO and DSO isaresponsibilityofTSOandDSO  The evaluation of the possibility of rearranging the HV network scheme, restoring the voltage • TheevaluationofthepossibilityofrearrangingtheHVnetworkscheme,restoringthevoltagefor for all plants and stations; this action is a responsibility of TSO and DSO allplantsandstations;thisactionisaresponsibilityofTSOandDSO If it is not possible to restore shortly (within some minutes) the voltage at the HV station, the TSOIf citanis rneoqtupiroes stihbele DtoSOre sttoo rreessthaorrt ttlhye( wMitVh innestowmoerkm iinnduetepse)ntdheenvtolylt.a Tgheea tmthaeinH aVctsiotantsio enx,etchueteTdS Oby catnher eDqSuOir eatrhe:e DSOtorestarttheMVnetworkindependently. ThemainactionsexecutedbytheDSO are:  Evaluation of network capacity to provide the restoring service, i.e., reserve of power on the • Evsaamluea tigornido,f cnoentwsiodrekricnagp atchiety mtoaprgroinv idoef tchuerrreesntto rrienfgersreerdv ictoe, it.hee., rtehseerrmvealo flipmoiwt eorfo tnhteh efeseadmeers; grfildex,icboinlistyid einr integrmthse omf arregminootef ccuornrternotl rseyfsetrermed otof tthhee tshaemrme aglrliidm, ictoonfstihdeerfienegd ethrse; flpeoxsisbibiliiltiytyi nto tearrmrasnogfer ae mneowte sccohnetmroel osfy dstiesmtriboufttihoen snaemtweogrrkid , consideringthepossibilitytoarrangeanew  scAhepmpleicoaftidoins torifb ruetsitoonrantieotwn oprlakns; this phase is carried out via the remote-control system • AMppolnicitaotrioinngo afnrde sstuopreartvioisniopnla onf sn;etthwisoprkh, ainse caissec aorfr eievdolouutitonv iianttoh ecrrietimcaol tseta-cteosn (torovlerslyosatdems, faults, etc.) • Monitoringandsupervisionofnetwork,incaseofevolutionintocriticalstates(overloads,faults, This operation condition increases the risk for the DSO to have an interruption of service, with etc.) relevant contribution to the value of System Average Interruption Duration Index (SAIDI) and System Average Interruption Frequency Index (SAIFI), the regulated indexes for the assessment of ThisoperationconditionincreasestheriskfortheDSOtohaveaninterruptionofservice,with continuity of the service for the distribution operator [32–34]. relevantcontributiontothevalueofSystemAverageInterruptionDurationIndex(SAIDI)andSystem As well known, the scheme of network operation, during supply restoration, is much different AverageInterruptionFrequencyIndex(SAIFI),theregulatedindexesfortheassessmentofcontinuity from the normal condition. The main factors to consider are: oftheserviceforthedistributionoperator[32–34].  ATshwee lplrkonboawbinli,tyth ethsacth eam faeuolft naertiwseosr, kdoupee rtaot itohne, dinucrrienagsisnugp polfy nreusmtobreart ioonf ,kisilmomuectherdsi ffoefr egnrtid fromthsuebntoernmdeadl ctoon tdhiet isoanm. Te hberemakaeinr factorstoconsiderare:  The level of damage in case of interruption, due to the increasing of number of customers • Theprobabilitythatafaultarises,duetotheincreasingofnumberofkilometersofgridsubtended subtended to the same breaker tothesamebreaker It is obvious that both mentioned parameters are higher with respect to the normal operating condition of the network, increasing the risk of a service interruption. In this framework, a BESS in MV can help the restoration procedure, acting as a slack bus between electricity supply and demand and thus increasing the flexibility of the grid: it can drive the controlled power grid, sustaining the consumption and allowing, whenever possible, the Sustainability2018,10,2058 5of17 • The level of damage in case of interruption, due to the increasing of number of customers subtendedtothesamebreaker Itisobviousthatbothmentionedparametersarehigherwithrespecttothenormaloperating conditionofthenetwork,increasingtheriskofaserviceinterruption. Inthisframework,aBESSinMVcanhelptherestorationprocedure,actingasaslackbusbetween electricitysupplyanddemandandthusincreasingtheflexibilityofthegrid: itcandrivethecontrolled power grid, sustaining the consumption and allowing, whenever possible, the synchronization of DG.Moreover,BESSsandDGs,possiblyoperateincoordinationwitheachother,contributetorisk mitigation. Asamatteroffact,inthetraditionalservicerestorationafaultoccurredonportionofthe networkdeterminesanoutageonboththerestoredandrestoringgrids;ontheotherhand,adoptinga BESSontheMVbus-bar,bothgridsaredecoupledandindependentofeachotherintermsoffault selectionandservicerestoration. Onthesegrounds,anactualsuitablecasestudyofe-distribuzione’sMVactivegridequippedwith a BESS has been selected, in order to quantitatively verify how the involvement of a BESS in the restorationprocedurecanimprovethequalityofservice. Thecasestudyherereportedisjustafirst exampleofapplication,inordertohavenumericalevidenceoftheproposalinarealspecificcase,butit canbegeneralizedinthefuture. Inaddition,itisworthtostressthattheanalysesdealwiththesteady stateoperationofthegrid,disregardinganyissuerelatedtotransients,frequencyandvoltagecontrol. 3. SimulationProcess The simulation of the Restoration Plans (RPs) was carried out via a MATLAB®/Simulink® program, developed for the purpose, consisting in a model of the network in steady-state and in aroutinefortheexecutionofmulti-periodPFA.Inaddition,aMATPOWERprogramwasusedfor comparisonforthevalidationofPFA[35].Thesimulationswerecarriedoutforthepurposeofverifying iftheRPsnetworkproposedareabletoensureasecureoperationofthedistributionsystem. The validation in steady state has been performed, verifying the power balance and compliance with voltage and current constraints in all system components (including BESS). Issues related to the transientsandtofrequencyandvoltagecontrolhavebeendisregardedatthisstage. ThedevelopedMATLAB®/Simulink®programislinkedtoaspreadsheet(e.g. MSExcel®)with networkdataandoperatesasfollows: • ImportsnetworkdatafromspreadsheetinMATLAB® • RunsaPowerFlowAnalysis(PFA)intheSimulink®networkmodelforeachtime-step(15min) • OutputstheresultstoMATLAB®andexportsthemtospreadsheet ForeachofthreeRPs,thevaluesandissuesassessedintheprocedureare: • Thetimeofblackout(t(cid:48)) • TheactivepowerdemandintheHV/MVstationatthetimeoftheblackout(P(cid:48) ) TOT • ThenumberofredLVandMVusers(N (i),N (i))withrespectiveactivepower(P (i), LVcust MVcust LVcust P (i))atstagei MVcust • ThestateofchargeoftheBESSindifferentstages(SOC (i)) BESS • Thepercentageofre-fednetwork,definedastheratioofthenumberofre-fedcustomerstothe totalnumberofcustomerstobere-fed: n(i) = Nc(ui)st ×100= NL(Vi)cust+NM(i)Vcust ×100; (5) % N N cust cust • ThepercentageofLVcustomersre-fedinanystageinrespectofthetotalnumberofLVcustomers tobere-fed: Sustainability2018,10,2058 6of17 (i) n(i) = NLVcust ×100; (6) LV% N LVcust • Thesequence,typeandsizeofDGpowerplantsreconnectedatdifferentstages • The power generated by the MV DG power plants at the stage i of reconnection, that can be (i) injectedinthedistributiongridP MV DG • Thecompliancewithvoltageandfrequencyconstraintsinthedifferentstages(V <V(i)<V , min MAX f <f(i)<f ) min MAX Sustainability 2018, 10, x FOR PEER REVIEW 6 of 16 Itisobviousthattheparameterslisted,especiallythosedefinedin(5)and(6),arestronglyrelatedto It is obvious that the parameters listed, especially those defined in (5) and (6), are strongly related theindicatorsofqualityofservicereportedinSection2andsotheycanbeassumedasindicatorsof to the indicators of quality of service reported in Section 2 and so they can be assumed as indicators theeffectivenessoftheproposedRP.AschematicofthesimulationprocessisreportedinFigure3. of the effectiveness of the proposed RP. A schematic of the simulation process is reported in Figure 3. Figure 3. Schematic of the simulation process. Figure3.Schematicofthesimulationprocess. The charge and discharge process of the BESS has been modeled, in steady state, by considering Tthhee acmhoaurgnte oafn ednedrgisyc hchaarrggeepd roorc deissschoafrtgheed BinE SevSehrya sstbaegee,n pmluos dtheele edn,eirngys tleoassdeys isnta etaec,hb sytacgoen, sviidae ring theamthoe uranttedof cehnaergrginygc/dhiasrcgheadrgoinrgd eifsfcichiaerngceyd. ineverystage,plustheenergylossesineachstage,viathe ratedcharging/dischargingefficiency. 4. Case Study 4. CaseStudy 4.1. MV Grid 4.1. MVGTrhide reference case study is an active MV grid of e-distribuzione, connected to the HV system and to the BESS, with the layout shown in Figure 4. ThereferencecasestudyisanactiveMVgridofe-distribuzione,connectedtotheHVsystemand totheBESS,withthelayoutshowninFigure4. Figure 4. Case study MV grid: HV/MV stations and MV lines. The HV/MV station includes one 16 MVA YYn transformer, with the neutral conductor on the MV side grounded via mobile Petersen coil. The MV bus-bar feeds 8 MV lines operated radially, in accordance with Italian distribution grid practices. Details on the composition of the MV network are reported in Tables 1 and 2. The 4 MV DG power plants are characterized in Table 3. Sustainability 2018, 10, x FOR PEER REVIEW 6 of 16 It is obvious that the parameters listed, especially those defined in (5) and (6), are strongly related to the indicators of quality of service reported in Section 2 and so they can be assumed as indicators of the effectiveness of the proposed RP. A schematic of the simulation process is reported in Figure 3. Figure 3. Schematic of the simulation process. The charge and discharge process of the BESS has been modeled, in steady state, by considering the amount of energy charged or discharged in every stage, plus the energy losses in each stage, via the rated charging/discharging efficiency. 4. Case Study 4.1. MV Grid Sustainability2018T,1h0e, r2e0f5e8rence case study is an active MV grid of e-distribuzione, connected to the HV system and 7of17 to the BESS, with the layout shown in Figure 4. Figure 4. Case study MV grid: HV/MV stations and MV lines. Figure4.CasestudyMVgrid:HV/MVstationsandMVlines. The HV/MV station includes one 16 MVA YYn transformer, with the neutral conductor on the TheHMVV/ sMideV gsrotauntidoend ivniac mluodbeilse Poenteers1e6n McoiVl. AThYe YMnV tbruasn-bsafro fremedes r8, MwVit hlintehs eopneeruatterda rlacdoianlldy,u inc toronthe accordance with Italian distribution grid practices. Details on the composition of the MV network are MVsidegroundedviamobilePetersencoil. TheMVbus-barfeeds8MVlinesoperatedradially,in reported in Tables 1 and 2. The 4 MV DG power plants are characterized in Table 3. accordancewithItaliandistributiongridpractices. DetailsonthecompositionoftheMVnetworkare reportedinTables1and2. The4MVDGpowerplantsarecharacterizedinTable3. Table1.MVNetworkComposition(lines). Line TotalLength NakedOverheadLine UndergroundCable OverheadCable (km) (km) (%) (km) (%) (km) (%) 1 0.09 0.00 0.0% 0.00 0.0% 0.09 100.0% 2 10.48 0.01 0.1% 10.15 96.9% 0.23 2.2% 3 17.88 1.27 7.1% 10.66 59.6% 5.95 33.3% 4 19.13 13.64 71.3% 2.87 15.0% 2.62 13.7% 5 25.21 19.51 77.4% 1.03 4.1% 4.67 18.5% 6 7.68 0.00 0.0% 0.00 0.0% 7.68 100.0% 7 8.43 0.00 0.0% 8.43 100.0% 0.00 0.0% 8 0.08 0.00 0.0% 0.00 0.0% 0.075 100.0% Total 88.98 34.43 33.14 21.32 Table2.MVNetworkComposition(CustomersandMV/LVStations). Line MV/LVStations LVCustomers MVCustomers 1 0 0 0 2 11 663 0 3 20 1228 0 4 13 1104 1 5 18 1312 2 6 9 1220 0 7 12 1139 0 8 0 0 0 Total 83 6666 3 Table3.DGPowerPlants. Line Type RatedPower(kVA) 4 PV 690 4 Hydro 3,000 5 PV 970 8(dedicated) Wind 4,000 Althoughthelinesareradiallyoperated,therearecommonnodesbetweenthedifferentlines. Thisisusefulformodifyingnetworkconfigurationincriticaloperatingconditions,e.g.,restoration afterablackout. Concerningnetworkautomation,theremote-controlsystemallowsthesupervision oftheMVnetworkandthe28nodesRemoteControlNodes(RCNs),inadditiontoeachlinecircuit Sustainability 2018, 10, x FOR PEER REVIEW 7 of 16 Table 1. MV Network Composition (lines). Line Total Length Naked Overhead Line Underground Cable Overhead Cable (km) (km) (%) (km) (%) (km) (%) 1 0.09 0.00 0.0% 0.00 0.0% 0.09 100.0% 2 10.48 0.01 0.1% 10.15 96.9% 0.23 2.2% 3 17.88 1.27 7.1% 10.66 59.6% 5.95 33.3% 4 19.13 13.64 71.3% 2.87 15.0% 2.62 13.7% 5 25.21 19.51 77.4% 1.03 4.1% 4.67 18.5% 6 7.68 0.00 0.0% 0.00 0.0% 7.68 100.0% 7 8.43 0.00 0.0% 8.43 100.0% 0.00 0.0% 8 0.08 0.00 0.0% 0.00 0.0% 0.075 100.0% Total 88.98 34.43 33.14 21.32 Table 2. MV Network Composition (Customers and MV/LV Stations). Line MV/LV Stations LV Customers MV Customers 1 0 0 0 2 11 663 0 3 20 1228 0 4 13 1104 1 5 18 1312 2 6 9 1220 0 7 12 1139 0 8 0 0 0 Total 83 6666 3 Table 3. DG Power Plants. Line Type Rated Power (kVA) 4 PV 690 4 Hydro 3,000 5 PV 970 8 (dedicated) Wind 4,000 Although the lines are radially operated, there are common nodes between the different lines. Sustainability2018,10,2058 8of17 This is useful for modifying network configuration in critical operating conditions, e.g., restoration after a blackout. Concerning network automation, the remote-control system allows the supervision of the MV network and the 28 nodes Remote Control Nodes (RCNs), in addition to each line circuit breakerontheMVbus-bar. Figure5showstheRCNineachMVlineofthecasestudy;0denotesthe breaker on the MV bus-bar. Figure 5 shows the RCN in each MV line of the case study; 0 denotes the RCNcorrespondingtothelinecircuitbreakersattheheadoflineonMVbus-barside. RCN corresponding to the line circuit breakers at the head of line on MV bus-bar side. Sustainability 2018, 10, x FOR PEER REVIEW 8 of 16 Figure 5. Remote control nodes for each MV line of the case study. Figure5.RemotecontrolnodesforeachMVlineofthecasestudy. The definition of the RP is based on the location of RCNs for each line, the number of MV/LV Thestdateiofinnsi tainodn MoVf tchuestoRmPeirss, bthaesier davoenragthe eacltoivcea atinodn reoafctRivCe Nposwfeorr deeamcahnldi.n e,thenumberofMV/LV stationsandMVcustomers,theiraverageactiveandreactivepowerdemand. 4.2. BESS and Control System 4.2. BESSandA CLoitnhtiruoml-SIoynst BemESS is installed in the HV/MV station and connected to the MV bus-bar via a dedicated line (Line 1). A schematic of the BESS control system and an image of the installation are ALpitrhesieunmted-I oinn FBigEuSreS 6i.s Tihnes stpaelcleifdicaitniotnhse ofH thVe/ EMSSV arset: ationandconnectedtotheMVbus-barviaa dedicatedline(Line1). AschematicoftheBESScontrolsystemandanimageoftheinstallationare  BESS rated capacity: 2 MWh presentedinFigure6. ThespecificationsoftheESSare:  BESS rated power: 2 MVA  Battery module rated voltage: 15.2 V • BESSratedcapacity: 2MWh  Battery module rated capacity: 0.625 kWh • BESS raTteodtalp eoffwicieern:c2y:M 80V%A • Batte ryImnsotadlluatlieonra: tCeodntvaoinletarsg (e1:0 1c5o.n2taViners, 8 power converter systems, 3264 batteries), 400 m² • Batte ryPmroovdiduelres:r Saytestdemca inptaecgirtayto:r0 a.6n2d5 bkatWterhies: NEC  Technology: Li-Ion (Lithium metal oxide) • Totalefficiency: 80%  Life: 4,000 cycles, 10 years • Insta llatMioanxi:mCuomn taauixnileirasry( 1co0ncsounmtpatiinoenr: s7,887.7p5o kwWehr/dcoayn vertersystems,3264batteries),400m2 • Providers: Systemintegratorandbatteries:NEC • Technology: Li-Ion(Lithiummetaloxide) • Life: 4,000cycles,10years • Maximumauxiliaryconsumption: 787.75kWh/day Figure 6. Schematic of the BESS, based on [36] (left panel), and image of the installation (right panel). To clarify how the proposed procedure can be then implemented in the physical system, a general description of the control architecture available in the e-distribuzione MV grid (including the BESS in HV/MV stations), is detailed. Sustainability 2018, 10, x FOR PEER REVIEW 8 of 16 Figure 5. Remote control nodes for each MV line of the case study. The definition of the RP is based on the location of RCNs for each line, the number of MV/LV stations and MV customers, their average active and reactive power demand. 4.2. BESS and Control System A Lithium-Ion BESS is installed in the HV/MV station and connected to the MV bus-bar via a dedicated line (Line 1). A schematic of the BESS control system and an image of the installation are presented in Figure 6. The specifications of the ESS are:  BESS rated capacity: 2 MWh  BESS rated power: 2 MVA  Battery module rated voltage: 15.2 V  Battery module rated capacity: 0.625 kWh  Total efficiency: 80%  Installation: Containers (10 containers, 8 power converter systems, 3264 batteries), 400 m²  Providers: System integrator and batteries: NEC Sustainability2018, 10,T20ec5h8nology: Li-Ion (Lithium metal oxide) 9of17  Life: 4,000 cycles, 10 years  Maximum auxiliary consumption: 787.75 kWh/day Figure 6. Schematic of the BESS, based on [36] (left panel), and image of the installation (right panel). Figure6.SchematicoftheBESS,basedon[36](leftpanel),andimageoftheinstallation(rightpanel). To clarify how the proposed procedure can be then implemented in the physical system, a general description of the control architecture available in the e-distribuzione MV grid (including the ToclarifyhBEoSwS int hHeV/pMrVo sptaotisoends), pisr doectaeildedu. r ecanbethenimplementedinthephysicalsystem,ageneral descriptionofthecontrolarchitectureavailableinthee-distribuzioneMVgrid(includingtheBESSin HV/MVstations),isdetailed. e-distribuzione MV grid management and control is in charge of 28 control centers; each one is responsibleforseveralactivities,e.g.,:qualityofservicecertification,managementofcrewsinthefield, network and equipment predictive maintenance, and cooperation with the TSO in order to prevent blackoutsandemergencies.ThesystemisequippedwithaSupervisoryControlandDataAcquisition (SCADA), located in the control centers, which communicates with Remote Terminal Units (RTUs), Sustainability 2018, 10, x FOR PEER REVIEW 9 of 16 locatedintheHV/MVandMV/LVstations.IncaseoffaultontheHVgrid,whentheTSOrequiresthe DSOtorestoresee-rdvistirciebuozinonet hMeVM grVid mneantwagoemrkenitn adnde pcoenntrdole ins tilny ,chtahregeS oCf 2A8 DcoAntrsoyl csetnetmers;i seaachb loenet ois identifyand responsible for several activities, e.g.,: quality of service certification, management of crews in the isolatetheMVnetworkportiontobere-supplied.Consequently,theSCADAcanalsosendtheBESSa field, network and equipment predictive maintenance, and cooperation with the TSO in order to black-startremproevteenct obmlacmkoauntsd a,nmd eamkeinrggenscuierse. Tthhea tsyasltlemth ies esqeuciuprpietdy wchithe cak Ssuhpearvveisobrey eCnonpterorlf oanrdm Deadtac orrectly.The communicatiAoncquairscithiointe (cStCuArDeAis), bloacsaetedd oinn ththe ecoCntlrioeln cte/nSteersr,v werhimch ocdomelm.uTnhiceatBesE wSiSthl oRceamlocteo nTetrrmolilnearl assumesthe Units (RTUs), located in the HV/MV and MV/LV stations. In case of fault on the HV grid, when the serverrolewhile,ontheDSOside,thereisanRTU,thatassumestheclientroleandisremotelyconnected TSO requires the DSO to restore service on the MV network independently, the SCADA system is withthecontraobllec teo nidteenrtitfhy aatndh iossotlastet hthee DMSVO neStwCoArkD pAort.ioTnh teo bine froe-rsmupaptliieodn. Ceoxncsheqaunegnetldy, btheet SwCeAeDnAc claienn tandserver iscompliantawlsoit shen[d3 t7h]e. BAESsS as hbloacwk-nstairnt rFemigouter ceom7,mianndo, rmdaekrintgo suerxe ethcaut taell taheb sleaccukri-tys tcahretckos phaevrea btieoenn andsupply performed correctly. The communication architecture is based on the Client/Server model. The BESS powertoatargetload,theBESSlocalcontrollermustreceivevoltagecommands(V )foreachstagefrom local controller assumes the server role while, on the DSO side, there is an RTU, that assestumes the theSCADAtcoliecnot nrotrleo lanthd eisp roeimnotteolyf dcoenlnivecetreyd vwoitlht athgee sc.onUtrpolo ncenrteecr etihvaitn hgosVtss etthce oDmSOm SaCnAdDsAf.r oTmhe theSCADA, theBESSdeteinrfmorminaetisont hexechtaanrggeedt bpeotwineetno cflidenetl iavnde rsyervveor litsa cgoembplyiacnat lwciuthl a[3ti7n]. gAtsh sheodwinf fienr Feingucree b7,e itnw eentheV set order to execute a black-start operation and supply power to a target load, the BESS local controller values from the SCADA and the voltage values measured at point of delivery under the conditions must receive voltage commands (Vset) for each stage from the SCADA to control the point of delivery thatfrequencvyolitsagaesc. oUnpsotna rnectevivainlug eVsoet fco5m0mHanzdas nfrdomz ethreo S-sCeAqDuAe, nthcee BvEoSSlt adegteermisinaelsw thaey tsarsgeett ptooinzte orfo .Inorderto avoidinrushdceulirvreeryn tvofllotawge ibnyt ocatlchuelaltoinagd t,heth deifBfeEreSnSce pbreotwveiedne sthae sVosetf tvasltuaerst ffruomn ctthieo nSC,AwDhAic ahndin tchree asesvoltage voltage values measured at point of delivery under the conditions that frequency is a constant value graduallyaccordingtoasoftstartperiod(e.g.,from1sto20s). Inthiswork,thesystemisstudiedin of 50 Hz and zero-sequence voltage is always set to zero. In order to avoid inrush current flow into stationarystatthee; ltohade,r tehfeo BreE,SSit pirsovaisdseus ma seodft stthaartt ftuhnectivono,l twahgicehi isnccroenassetsa vnotl,taigme pgroadseudallyb yactchorediBngE StoS a, operatingas thevoltageresgofut slatatrot rp.eriod (e.g., from 1 s to 20 s). In this work, the system is studied in stationary state; therefore, it is assumed that the voltage is constant, imposed by the BESS, operating as the voltage regulator. Figure 7. Black-start control scheme for the BESS. Figure7.Black-startcontrolschemefortheBESS. Hence, it is clear that the proposed procedure can be implemented in the physical system, both fully automated and manually guided by the system operator. 4.3. Simulation Scenarios and Hypotheses For a given grid layout and control system, a RP of the MV grid is a function of the initial operation scenario, in terms of:  The state of the network in the time of blackout, i.e., amount of active power demanded by the loads and amount of power supplied by DG, which define the aggregate equivalent load at the MV bus-bar Sustainability2018,10,2058 10of17 Hence,itisclearthattheproposedprocedurecanbeimplementedinthephysicalsystem,both fullyautomatedandmanuallyguidedbythesystemoperator. 4.3. SimulationScenariosandHypotheses Foragivengridlayoutandcontrolsystem,aRPoftheMVgridisafunctionoftheinitialoperation scenario,intermsof: • Thestateofthenetworkinthetimeofblackout,i.e.,amountofactivepowerdemandedbythe loadsandamountofpowersuppliedbyDG,whichdefinetheaggregateequivalentloadatthe MVbus-bar • TheStateofCharge(SOC)oftheBESSatthetimeoftheblackout Sustainability 2018, 10, x FOR PEER REVIEW 10 of 16 Thus,t heTfihres Sttsattee pof tCohabrugeil (dSOaCR) Pof itshet oBEaSsSs uatm thee ttihmee bolfa tchke obluactktoimut eandtheBESSSOC. For the caTsheuss, tthued fyir,sto spteepr taot ibounildd aa RtPa ifso tor atshsuremee rtheef ebrlaecnkcoeut dtiamyes an(wd teheek BdESaSy ,SOSCat. urday and Sunday) inJanuaryweFreor athvea cilaaseb sletu.dFy,o orpeeraactihono dfattha efomr ,thtrheee raefgegrernecge adtaeyse (qwueievkadlaeyn, Staltouarddaya tantdh eSuMndVayb) iuns -barwas calculatedJvaniuaaaryP wFeAre; tahveailraebsleu. lFtsora eraechre opf othrteemd, tihne Faigggurergeat8e, ewquhivearleentth leoasdig ant tohfe aMctVi vbeusp-boawr werasi spositive calculated via a PFA; the results are reported in Figure 8, where the sign of active power is positive whenpowerflowsfromtheHVsidetotheMVside(passivedistributionnetwork). when power flows from the HV side to the MV side (passive distribution network). Figure 8. Daily net active power profile at the MV bus-bar for three reference days. Figure8.DailynetactivepowerprofileattheMVbus-barforthreereferencedays. Comparing the profile of the three days, it is evident that the distribution grid is passive only in some hours of the weekday, due to the combination of a greater load and a minor DG production. Comparingtheprofileofthethreedays,itisevidentthatthedistributiongridispassiveonly During the two weekend days, the grid behaves mostly as an equivalent generator in respect of the insomehoursoftheweekday,duetothecombinationofagreaterloadandaminorDGproduction. HV system. DuringthetwOonw theeeske egnrodunddasy, tsh,et hcheoigcre ihdasb beeheanv toe sbumilod sutply RaPs aassnumeqinugi tvhae lweneetkgdeany ears athtoe rreifnerreenscpe ectofthe HVsystemc.ase, and check its feasibility in the other two days, when the net load at the MV bus-bar is mostly negative, due to a lower passive load and a higher generation. Onthesegrounds,thechoicehasbeentobuildupRPassumingtheweekdayasthereference With reference to the weekday, two times of blackout have been considered: maximum positive case,andcheckitsfeasibilityintheothertwodays,whenthenetloadattheMVbus-barismostly net load time (2:30 p.m.) and the maximum negative net load time (6:30 p.m.), as shown in Figure 7. negative,duetAonaothloerw heyrpoptahsessiisv deealolsa wditahn thde aSOhCig ohf eBrEgSSe nate trhaet tiiomne. of blackout. Two level of SOC have Withrbeeefenr ceonncsiedetoredth: 1e00w%e aenkdd 5a0y%,.t wotimesofblackouthavebeenconsidered: maximumpositive Based on the location of RCNs, different RPs were developed, divided in 10 stages. For each stage: netloadtime(2:30p.m.) andthemaximumnegativenetloadtime(6:30p.m.),asshowninFigure7. Anoth erhTyhep loistth oef soipsednienagl sanwd ictlhositnhge oSpOeraCtioonfs BwEasS Sseta wttithhe retsipmecet otof tbhela vcakrioouust .liTnews olevelofSOChave  The number of LV and MV customers and DG power plants to be reconnected were defined, beenconsidered: 100%and50%. assuming a time interval of 30 s and 300 s, respectively BasedonthelocationofRCNs,differentRPsweredeveloped,dividedin10stages. Foreachstage: The RPs were firstly simulated, assuming the absence of MV DG, in order to verify restoration • Thelicsatpoafbioliptieesn oinf tghea snydstecmlo ussiningg oonpley rtahtei oinntesrnwala DsSsOe tsowuirtches,r ei.se.p, ewcitthtooutt hloecavl aDrGio. uThsel isnamese RPs were then checked, involving the MV DG plants, in reference to a smart management of the grid. The • The number of LV and MV customers and DG power plants to be reconnected were defined, rationale is then to shift the connection of distributed generation as much as possible, to take assumaidnvganatatgiem oef tihne tsetorrvaagle osyfs3te0ms, eavnedntu3a0l0ly sc,ornesisdpeericntgi vneo ldyistributed generation contribution (only energy storage contribution) in the restoration process. TheRPswTheer reesfiurltsst olybtsaiinmedu floart tehde ,tharseseu RmPsi,n bguiltth uep ainb rseefenrcenecoe fofM thVe wDeeGkd,ainy, oforrd theer ttwoov sceerniafyriorse storation capabilitieosf opfotshitieves ymsatxeimmuums ianngd noenglaytivthe emianxitmerunma lneDt SloOads, owuitrhcoeust, ain.ed. ,wwithit hDoGu, tanlodc ianl rDefeGre.nTceh eof sameRPs different initial SOC of BESS, are reported. werethenchecked,involvingtheMVDGplants,inreferencetoasmartmanagementofthegrid. The Table 4 summarizes the list of RCN operation for each stage in each line for the RP1. The table is based on RP1. RPs differ according to the stage of reconnection of MV DG, i.e., RCNs 3 and 6 of line 4 are closed during Stage 3, 4 or 5, in RP1, RP2 or RP3, respectively.

Description:
Akhil, A.A.; Huff, G.; Currier, A.B.; Kaun, B.C.; Rastler, D.M.; Chen, S.B.; Cotter, A.L.; Bradshaw, D.T.;. Gauntlett . Licensee MDPI, Basel, Switzerland.
See more

The list of books you might like

Upgrade Premium
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.
DMCA & Copyright: Dear all, most of the website is community built, users are uploading hundred of books everyday, which makes really hard for us to identify copyrighted material, please contact us if you want any material removed.
Copyright @ PDFdrive.to, 2022 | We love our users