Abstract
To achieve carbon neutrality in the built environment, designers worldwide must prioritize the development of more sustainable buildings, particularly those integrating hybrid solar configurations, to reduce reliance on conventional energy sources. In this context, energy storage systems are quintessential. Although numerous studies have examined the techno-economic performance of hybrid solar Photovoltaic (PV)/battery/hydrogen systems, the substantial potential of hydrogen under different climate zones, load profiles, and parameter variations in the off-grid building stocks remains underexplored. This study seeks to address this gap by comprehensively investigating PV/battery/hydrogen systems through a tailored optimization workflow while accounting for the dynamic interrelation between hourly spatiotemporal variations in solar energy output and energy demand across six different climate zones. In addition to optimizing the number and tilt angle of photovoltaic panels, the battery capacity, and storage modulation coefficient, the workflow optimally sizes the key hydrogen system components, including the electrolyzer, storage tank, and fuel cell. The main objective is to minimize the Levelized Cost of Energy (LCOE) via Particle Swarm Optimization (PSO) and compare it against two other metaheuristic algorithms: Genetic Algorithm (GA) and Ant Colony Optimization (ACO). PSO proved its superiority by providing relatively lower LCOE values of 0.44-0.7 $/kWh, while ensuring the targeted Self- Sufficiency Ratio (SSR) of 100 %. Moreover, incorporating the hydrogen system reduces the LCOE by up to 0.12 $/kWh (15.4 %) in Ifrane, compared to battery-only systems. These findings highlight the importance of climate-sensitive design and hybrid storage strategies in enhancing the economic and energy resilience of off-grid net-zero energy buildings. At an SSR of 90 %, LCOE ranges between 0.28 and 0.37 $/kWh, reflecting a 36.4-54.3 % cost reduction. This shows that fully renewable systems, while feasible, may incur substantially higher costs of electricity. Interestingly, the sensitivity analysis reveals that the optimal configurations are highly sensitive to the discount rate, as a reduction from 5 % to 2 % leads to an LCOE decrease of up to 0.097 $/kWh, followed by the battery's techno-economic parameters. Lastly, the Monte Carlo-based uncertainty analysis further shows that the optimized LCOE values consistently fall within the interquartile range of the probability distributions, confirming the consistency of the optimal solutions.
NETL
NSTL
Elsevier