Techno-economic multi-objective planning of hydrogen-enabled hybrid microgrids: A realistic case study

• Multi-objective framework developed for microgrid planning • Four models evaluated for LCOE, HES hosting, and LPSP • Ten population-based algorithms tested for performance • Techno-economic assessment of microgrid cost and renewable trade-offs Microgrids (MGs) play an important role in enhancing energy reliability, integrating renewable resources, and supporting decentralized power systems. This paper proposes a comprehensive techno-economic, multi-objective planning framework for hybrid MGs. Four complementary optimization models are introduced to minimize the levelized cost of energy (LCOE), maximize the hosting capacity of renewable energy and hydrogen-based energy storage (HES), and reduce the loss of power supply probability (LPSP), both individually and in combined formulations. The framework incorporates HES components, including PEM electrolyzers, hydrogen tanks, and PEM fuel cells, together with photovoltaic systems, wind turbines, battery storage, electric vehicles, and grid connectivity. A comparative metaheuristic optimization platform, comprising ten recent population-based algorithms, is employed to evaluate convergence behavior, robustness, and solution quality. A realistic MG case study in Dammam, Saudi Arabia, is used to validate the proposed framework. The results quantify the trade-offs among cost, renewable energy penetration, and environmental performance, and provide practical insights for optimal sizing and operation of hybrid MGs. The first model achieves the lowest LCOE, with a 41.3 % reduction relative to the multi-objective model, but shows reductions in sustainability and reliability. The second and third models increase renewable and HES penetration by approximately 6.1 %. These gains are accompanied by cost increases exceeding 230 %. Sensitivity analysis indicates that HES integration increases the LCOE, while reducing carbon emissions by 17.3 % and improving MG reliability by about 6.47%. The algorithmic comparison shows differences in robustness across the tested models, while the Starfish Optimization Algorithm provides consistently high-quality and robust solutions.