Energy Efficiency Evaluation and Optimization Methods in Electric Transportation Systems

This study presents a reinforcement learning-based framework for optimizing energy efficiency in electric vehicle (EV) and grid interactions. Managing grid load, especially during peak charging times, becomes critical as EV adoption increases. The framework leverages short-term load forecasting with data from the Electricity Load Forecasting dataset to dynamically balance energy distribution. Experimental results show a 131.5% reduction in Peak-to-Average Load Ratio (PALR) with a GRU model, a 97.53% improvement in Charging Efficiency Index (CEI) with a CNN-LSTM model, and a lower RMSE, demonstrating the framework’s effectiveness in enhancing grid integration and EV charging efficiency.