Multi-Objective Dynamic System Model for the Optimal Sizing and Real-World Simulation of Grid-Connected Hybrid Photovoltaic-Hydrogen (PV-H₂) Energy Systems

Hybrid renewable-hydrogen energy systems offer a promising solution for meeting the globe’s energy transition and carbon neutrality goals. This paper presents a new multi-objective dynamic system model for the optimal sizing and simulation of hybrid PV-H₂ energy systems within grid-connected buildings. The model integrates a Particle Swarm Optimisation (PSO) algorithm that enables minimising both the levelised cost of energy (LCOE) and the building carbon footprint with a dynamic model that considers the real-world behaviour of the system components. Previous studies have often overlooked the electrochemical dynamics of electrolysers and fuel cells under transient conditions from intermittent renewables and varying loads, leading to the oversizing of components. The proposed model improves sizing accuracy, avoiding unnecessary costs and space. The multi-objective model is compared to a single-objective PSO-based model that minimises the LCOE solely to assess its effectiveness. Both models were applied to a case study within Robert Gordon University in Aberdeen, UK. Results showed that minimising only the LCOE leads to a system with a 1000 kW PV, 932 kW electrolyser, 22.7 kg H₂ storage tank, and 242 kW fuel cell, with an LCOE of 0.366 £/kWh and 40% grid dependency. The multi-objective model, which minimises both the LCOE and the building carbon footprint, results in a system with a 3187.8 kW PV, 1000 kW electrolyser, 106.1 kg H₂ storage tank, and 250 kW fuel cell, reducing grid dependency to 33.33% with an LCOE of 0.5188 £/kWh.