System Sizing of Hybrid Renewable Systems Under Inverter and Contracted Grid Power Constraints with Flexible Load Integration

Abstract

Island microgrids face significant challenges, including seasonal load variations, weak interconnections, and high demand charges leading to oversized energy systems. This paper introduces a novel bi-level optimization approach, combining a genetic algorithm for capacity sizing and a rolling-horizon mixed-integer linear programming controller for daily dispatch scheduling. The method simultaneously optimizes photovoltaic arrays, battery storage, hydrogen tanks, inverter ratings, and contracted grid-import limits to minimize the net present cost. The approach was applied to a municipal energy community in Formentera, Spain, with an annual demand of approximately 203 megawatt-hours. Compared to the baseline scenario (661,677 euros), the proposed framework reduced the net present cost to 612,945 euros without load flexibility. Introducing load flexibility further decreased costs: 606,879 euros at 6 percent and 599,134 euros at 8 percent flexibility. Increased flexibility resulted in modest reductions in photovoltaic capacity from 155 to 150 kilowatt-peak, inverter size from 77 to 72 kilowatts, and contracted grid-import limits from 41 to 37 kilowatts. The findings underscore the significant economic and operational advantages of integrating demand-side flexibility into the co-optimization of component sizing, enhancing both the resilience and autonomy of islanded hybrid renewable energy systems.