Artificial Neural Network-Based Control for Enhanced V2V Energy Transfer in Electric Vehicles

Abstract

This project presents an advanced Electric Vehicle-to-Vehicle (V2V) energy transfer system using on-board converters enhanced with an Artificial Neural Network (ANN) controller to achieve faster and more efficient charging. Traditional V2V energy transfer systems rely on off-board interfaces or redundant conversion stages, which increase cost, complexity, and energy losses. The proposed approach eliminates these drawbacks by enabling direct energy exchange between two EV batteries through the use of Type-2 AC charger input ports and minimal switching components. This direct battery connection minimizes conversion losses and enhances overall system efficiency. The ANN controller is introduced to intelligently regulate voltage and current during charging and discharging, ensuring optimal power flow, reduced charging time, and improved system stability under varying load conditions. MATLAB/Simulink simulations demonstrate that the ANN-based system outperforms conventional PI and fuzzy controllers in terms of response speed, steady-state accuracy, and charging efficiency. The results validate the proposed ANN-controlled V2V power transfer strategy as an effective and scalable solution for next-generation EV networks.