Smart Charging for Electric Vehicles: Balancing Grid Stability and Driver Satisfaction

Electric vehicles (EVs) are a key component of modern transportation, yet optimizing their charging remains a challenge due to grid constraints and fluctuating renewable energy availability. This study proposes a machine learning-driven framework to enhance EV charging by prioritizing driver satisfaction. By considering factors such as socio-demographic attributes, battery State of Charge (SoC), charging station proximity, and dynamic pricing, the model predicts user satisfaction with high accuracy (87.9%). Trained on data from Hungarian EV users, the proposed optimization algorithm maximizes driver satisfaction while minimizing grid power costs. Simulation results demonstrate that the model achieves an average satisfaction rate of 98.5%, significantly outperforming traditional methods. Additionally, the approach maintains a stable average SoC around 50%, ensuring efficient energy use and grid stability. By dynamically allocating EVs to charging stations and leveraging renewable energy sources, this research presents a novel, user-centric strategy for smart EV charging, contributing to the sustainable evolution of transportation infrastructure.