Comparative Metaheuristic Optimization for Sizing Grid-Connected Hybrid Renewable Systems: A Libya Case Study

Abstract

Hybrid renewable energy system (HRE) sizing is vital for maximizing both economic and environmental performance in grid-connected systems. This study develops a comparative optimization framework for an HRE serving a camp in the southeastern region of Libya, integrating photovoltaic (PV) arrays, wind turbines (WT), battery storage systems (BT), fuel cells (FC), hydrogen tanks (HT) and an Electrolyzer (EL). Three objective functions, Levelized Cost of Energy (LCOE), Grid Reliance Fraction (GRF) and Pollutant Emission Coefficient (PEC), are employed to quantify techno-economic efficiency and environmental impact. To this end, three widely used metaheuristic algorithms; Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Grey Wolf Optimizer (GWO), are employed under identical settings to identify optimal system configurations. Each algorithm is run for a fixed number of iterations, and their diversity and convergence behaviors are monitored to ensure a fair benchmark. The comparative results show that PSO yields the most cost-effective design, achieving an LCOE of $0.1448/kWh, a reduction of 13.5% and 1.3% relative to GA and GWO, respectively and a GRF of 15.73%. These findings demonstrate that PSO strikes the best balance between cost, reliability and computational efficiency for this HRE case study. The framework and insights presented here can guide planners and engineers in selecting the most suitable optimization method for sustainable energy-system design.