Seite - 201 - in Short-Term Load Forecasting by Artificial Intelligent Technologies

Energies2018,11, 1948 2.4. Performance IndexesHybridPowerModels ThehybridpowermodelofFigure1wasappliedtopredict thesystemresponsesunderdifferent operationconditionsbasedonthe followingmanagementstrategies (seeFigure6): 1. Toavoidwastingrenewableenergy, thewindandsolarpowersubsystemsareoperatedasfollows: whenthebatterySOCisgreater than98%andthe inputrenewablepower, includingsolarand windpower, isgreater thanthe load, redundantrenewableenergyisdumped. Solarenergy is reducedfirstbecause it ismuchmoreabundant thanwindenergy.WhenthebatterySOCis less than95%,all renewableenergy issuppliedto thesystem. 2. ThePEMFCsystemisswitchedonwhenthebatterySOCreachesalowboundof30%.ThePEMFC is thenswitchedoffwhenthebatterySOCrises toahigh limitof40%.ThePEMFCiscontrolled toprovideadefaultcurrent loadof20Awiththehighestenergyefficiency,andit isset toprovide a loadupto50AwhenthebatterySOCcontinuouslydrops to25%[20]. 3. Thechemicalhydrogengeneratorsystemisswitchedonif thestoragehydrogenlevel is lower than a safety limit [25,26]. Wedesigned abatchprocedurewith suitable production rates to satisfy thesystemrequirements. Eachbatchconsumes60gofNaBH4andproducesabout150L ofhydrogen[25]. Thus, thePEMFCcanbecontinuouslyoperated. (a) Management strategy of the renewable energy. (b) Management strategy of the PEMFC. Figure6.Flowchartsof thepowermanagement. 201