Machine learning optimized green hydrogen fuel cell system for sustainable and efficient electric vehicle charging

Abstract

Green hydrogen has emerged as a transformative clean energy carrier with the potential to accelerate the global shift toward sustainable, low-carbon energy systems. It plays a crucial role in supporting the expansion of electric vehicles (EVs) by reducing greenhouse gas emissions and lowering the environmental impact of the transportation sector. In this study, hydrogen is generated through solar-powered electrolysis using a photovoltaic (PV) system integrated with a machine learning-based maximum power point tracking (MPPT) algorithm, which enhances solar energy harvesting and improves overall system efficiency. The research examines global trends in green hydrogen deployment with a particular focus on India, where rising electricity demand aligns with rapid EV infrastructure growth. Two operating configurations are examined in this study, namely a standalone EV charging station without ML optimization and a grid-connected system that incorporates ML-driven MPPT control. Simulations were performed for hydrogen flow rates of 10, 14, 45, and 80 cubic meters per second using real-time solar data from several Indian states to assess system performance under practical environmental conditions. The standalone proton exchange membrane (PEM) fuel cell achieves an efficiency of about 50%. In comparison, the ML-optimized grid-connected system reaches 60.43%, demonstrating the value of intelligent energy management in improving renewable hydrogen utilization. This study introduces a novel ML-enhanced PV hydrogen PEM fuel cell architecture that integrates solar-powered electrolysis, data-driven MPPT control, hydrogen storage, and fuel cell-based power conversion. It also provides a comparative analysis of standalone and grid-connected operations using real multi-state solar data to reveal the influence of hydrogen flow rates on system performance. Overall, the proposed framework offers a sustainable and scalable solution for developing regions facing rising energy demands and highlights the powerful synergy between renewable hydrogen, artificial intelligence, and EV infrastructure in creating clean, resilient, and future-ready energy systems.