Energy Management in Photovoltaic-Based Electric Vehicle Charging Systems With Battery Storage and Vehicle-to-Grid Support

The rapid growth of Electric Vehicles (EVs) and the increasing reliance on renewable energy sources (RESs) have highlighted the need for intelligent, storage-optimized charging infrastructure. However, conventional photovoltaic (PV)-based EV charging systems often suffer from intermittency, storage inefficiencies, and limited integration with the power grid, leading to increased operational costs and reduced sustainability. To address these challenges, this paper proposes a hybrid energy management (EM) framework that integrates a Pelican Optimization Algorithm (POA) and a Triple-Memristor Hopfield Neural Network (TMHNN) for optimizing Battery Energy Storage Systems (BESS) and Vehicle-to-Grid (V2G) operations in PV-powered EV charging systems. POA is employed to optimize power flow (PF) and storage scheduling, while TMHNN accurately forecasts energy demand, enabling dynamic coordination between PV generation, energy storage, and EV charging. Simulation results demonstrate that the suggested POA-TMHNN approach significantly reduces the cost of energy (COE) to $0.0841/kWh and carbon emissions to 173 956 kg, outperforming benchmark methods such as Particle Swarm Optimization with Adaptive Neuro-Fuzzy Inference System (PSO-ANFIS), Chicken Search Optimization with Spike Neural Network (CSA-SNN), Multiobjective Gray Wolf Optimization (MOGWO), Giant Trevally Tunicate Swarm Optimizer (GTTSO), and Multi-Objective Optimization (MOO) approaches. The findings confirm that the proposed method enhances storage utilization, operational efficiency, and environmental sustainability. This study contributes to the development of intelligent storage-centric EM systems suitable for next-generation EV infrastructure.