A comprehensive review of sizing and uncertainty modeling methodologies for the optimal design of hybrid energy systems

ABSTRACTEnergy demand is surging with the rise in population, economic development, and ever-increasing living standards. Due to sustainability and environmental issues, renewable energy sources have emerged as a credible option to meet this increased energy demand. However, it is plagued with the issue of variability and intermittency. Hybrid energy systems are proposed as a possible solution to this problem. The optimal sizing of hybrid energy systems ensures a reliable, efficient, and cost-effective power supply. Therefore, this paper discusses different hybrid energy systems in both on-grid and off-grid configurations, followed by the review of various sizing methodologies. The article also discusses various multi-criteria design indicators acting as decision variables, sensitivity variables, and constraints in different capacities while preparing the mathematical model of hybrid energy systems. As renewable resources and their based systems are inherently uncertain, it becomes imperative to characterize and model the uncertainty associated with such systems. Sincere efforts were made to understand various sources of uncertainty and how to characterize and model these uncertainties using different methodologies. The existing uncertainty modeling approaches were studied, compared, and analyzed. Further, the need for conducting sensitivity analysis and its usage in hybrid energy system design considering different sensitive parameters were also studied.KEYWORDS: Green energy resourceshybrid energy systemssustainable energy technologysize optimizationuncertainty modelingsensitivity analysis Abbreviations COVID=Coronavirus diseaseGHG=Greenhouse gasRES=Renewable energy systemHES=Hybrid energy systemHOMER=Hybrid Optimization of Multiple Electric RenewablesOOP=Object-Oriented ProgrammingMOPSO=Multi-objective particle swarm optimizationLCE/LCOE/COE=Levelized cost of energy/Cost of energyLPSP=Loss of power supply probabilityMES=Modified evolutionary strategyGraSO=Gradient swarm optimizationTNPC/NPC=Total net present cost/Net present costPV=PhotovoltaicWT=Wind turbineDG=Diesel generatorBES=Battery energy storagePBP/SPBP=Payback period/Simple payback periodROI=Rate of investmentIRR=Internal rate of returnNPV=Net present valueRE=Renewable energyES=Energy storageNRE=Non-Renewable energyMATLAB=Matrix LaboratorySC=SupercapacitorMOO=Multi-objective optimizationFC=Fuel cellBG=BiogasBM=BiomassPHS=Pumped hydro storageTAC/TACS:=Total annualized cost of systemHK=HydrokineticLOLP=Loss of load probabilityLOLE=Loss of Load ExpectationSOC=State of chargeLPS=Loss of power supplyASC/ACS=Annualized system cost/Annualized cost of systemNSGA=Non-dominated sorting algorithmPSO=Particle swarm optimizationRF=Renewable fractionRER=Renewable energy resourceArcGIS=Aeronautical reconnaissance coverage geographic information systemGOA=Grasshopper optimization algorithmMILP=Mixed integer linear programmingGA=Genetic AlgorithmICC=Initial capital costTRNSYS=Transient systems simulation programTOPSIS=Technique for order preference by similarity to ideal solutionTOC/OC=Total Operating cost/Operating costEG=Energy generationIR=Inflation rateLeA=Lead acidLI=Lithium-ionSOFC=Solid oxide fuel cellPrEM=Proton exchange membraneEENS=Expected energy not suppliedELF=Equivalent loss factorTEL=Total energy lossLA=Level of autonomyLOEE=Loss of energy expectationEIR=Energy index of reliabilityLUEC/UEC=Levelized unit electricity cost/Unit electricity costEIU=Energy index of unreliabilityLOLR=Loss of load riskTED=Total energy deficitAER=Annual energy recoveryDPSP=Deficiency of power supply probabilityWRE=Wasted renewable energyEE=Excess electricityLCC=Life cycle costLCUC=Life cycle unit costCRF=Capital recovery factorDPBP=Discounted payback periodCIC=Customer interruption costsCF/CFOE=Carbon footprint of energyCE=Carbon emissionsHDI=Human development indexEC=Employment creationSCC=Social cost of carbonSLLPR=Seasonal loss of load probability ratioDC=Direct currentAI=Artificial IntelligenceMPSO=Modified Particle swarm optimizationPSO-RF=PSO based on repulsion factorPSO-CF=PSO with constriction factorPSO-W=PSO with adaptive inertia weightACO=Ant colony optimizationSA=Simulated annealingACOR=Ant colony optimization for continuous domainsABC=Artificial bee colonyHS=Harmony searchMOCSA=Multi-objective crow search algorithmDHS/DHS=Discrete harmony searchCS=Cuckoo searchGPAP=Grid power absorption probabilityALO=Ant lion optimizationGWO=Grey wolf algorithmMDO-MOPSO=Multiple design option-Multi-objective particle swarm optimizationHCHSA=Hybrid chaotic search/harmony search/simulated annealingTLBO=Teaching – learning-based optimizationMOL=Many optimizing liaisonsTS=Tabu searchGUI=Graphic user interfaceAML=Algebraic modeling languageHSU=Hydrogen storage unitBCA=Branch and cut algorithmSSR=Self-sufficiency ratioTCS/TSC/TC=Total cost of systemPOO=Pareto optimal optimizationTDC=Total discounted costAOC=Average outage costLOLH=Loss of load hoursREGP=Renewable energy generation penetrationHPBS=Hybrid pumped battery storagePOE=Price of electricityTIC=Total investment costSSO=Social spider algorithmPDF=Probability density functionLOLF=Loss of load frequencyFOSMM=First order second-moment methodPEM=Point estimate methodMAED=Multi-area economic dispatchUT=Unscented transformationDER=Distributed energy resourcesIGDT=Information gap decision theoryRO=Robust optimizationAMFA=Adaptive modified firefly algorithmOATSA=One-at-a-time sensitivity analysisLSA=Local sensitivity analysisGSA=Global sensitivity analysisDisclosure statementThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.