Enhanced Hybrid Charging Park System Evaluation Using Neural Network Charging Controller

Hybrid charging parks, which combine renewable energy sources with traditional grid systems, have emerged in response to the increasing need for effective electric vehicle (EV) charging infrastructure. Using a Neural Network-based Charging Controller (NNCC), this study suggests an improved assessment methodology for hybrid charging park systems. The controller prioritizes car charging needs, minimizes energy losses, and dynamically optimizes energy distribution by balancing solar, wind, and grid sources. Real-time operational data, such as vehicle wait duration, state of charge (SoC), and fluctuation in renewable energy, were used to train a multi-layer perceptron (MLP) neural network. According to simulation data, the suggested NNCC outperformed traditional rule-based controllers in terms of the charging park's overall power efficiency by 18.7%. Additionally, the method increased the vehicles' final State of Charge (SoC) by an average of 12.5%, guaranteeing quicker and more dependable charging sessions. Additionally, the neural network-based system showed improved flexibility in the face of variable renewable energy circumstances, greatly boosting the resilience and sustainability of smart EV charging ecosystems.