Enhanced Power Quality Improvement in a PV-Based EV Charging Station Interfaced with a Three-Phase Grid Using ANN-Based Control

Abstract

This paper presents an enhanced power quality improvement technique for a solar photovoltaic (PV)-based electric vehicle (EV) charging station interfaced with a three-phase grid. The proposed system integrates an artificial neural network (ANN)-based controller to improve system performance by reducing settling time and minimizing ripple content. The charging station operates in both standalone and grid-connected modes, ensuring efficient EV battery charging while maintaining grid stability. Additionally, the station provides reactive power compensation and harmonics mitigation, reducing total harmonic distortion (THD) in grid currents below 5% as per IEEE-519 standards. The ANN controller optimizes the control of the voltage source converter (VSC) and the bidirectional DC-DC converter, enhancing dynamic response under grid disturbances and unbalanced conditions. The system effectively performs (i) harmonics current compensation, (ii) EV battery charging/discharging control, (iii) simultaneous EV charging and harmonics mitigation, and (iv) simultaneous EV discharging and harmonics compensation. A robust synchronization mechanism ensures seamless transition between grid-connected and standalone modes. Simulation results demonstrate that the ANN-based controller significantly outperforms conventional methods in reducing ripples and achieving faster stabilization, making it a reliable solution for power quality improvement in PV-based EV charging stations.