A grid-integrated PV-powered EV charging station featuring artificial neural network-based MPPT and comprehensive power flow control

Bangladesh University of Engineering and Technology, Dhaka, Bangladesh

^* Corresponding author: 1906086@eee.buet.ac.bd

Abstract

This paper presents a grid-connected solar-powered electric vehicle (EV) charging system featuring an artificial neural network (ANN)- based Maximum Power Point Tracking (MPPT) controller. The system integrates a photovoltaic (PV) array, a lithium-ion EV battery, and a single-phase AC grid via a centralized DC bus. The PV array is interfaced via a boost converter governed by the ANN-based MPPT method, which offers superior tracking speed and improved efficiency compared to conventional algorithms. A bidirectional DC-DC converter manages battery charging and discharging, while a full-bridge inverter with an LCL filter allows smooth, bidirectional power exchange with the grid. An adaptive power flow control strategy dynamically switches the system among three operating modes— PV to Vehicle and Grid, Vehicle-to-Grid (V2G), and Grid-to-Vehicle (G2V)—based on solar power availability and battery State of Charge (SOC). Simulation results demonstrate consistent DC bus voltage regulation, precise MPPT performance, and smooth transitions between operating modes, validating the system's adaptability and efficiency under varying conditions. Overall, the proposed framework demonstrates an effective and intelligent EV charging solution that enhances solar energy utilization and contributes to the development of resilient and sustainable smart grid infrastructure.