Electric Vehicle (EV) Charging Control System Management by Integrating a Microgrid Powered by Renewable Energy Sources
Keywords:
Electric Vehicle, Microgrid, Spatial Bayesian Neural Network (SBNNET), Dollmaker Optimization Algorithm (DMOA), Renewable Energy IntegrationAbstract
This study presents an approach to Electric Vehicle (EV) charging control system management by integrating a microgrid powered by renewable energy sources. The proposed system combines a Spatial Bayesian Neural Network (SBNNET) with the Dollmaker Optimization Algorithm (DMOA). The SBNNET model leverages spatial dependencies, physics-informed training and Bayesian uncertainty modelling to provide precise and reliable forecasts of EV charging demand across diverse geographical regions. Meanwhile, the DMOA efficiently manages the intermittency of renewable energy sources, battery storage utilization and dynamic EV charging schedules to stabilize voltage and frequency fluctuations, reduce energy costs and maximize renewable energy usage. Simulation results of the proposed method demonstrated significant improvements in load prediction accuracy, microgrid stability, and operational efficiency, highlighting the system’s potential to support sustainable and resilient EV charging infrastructure. Finally, this integrated approach offers a promising framework for smarter energy management in electric transportation ecosystems, contributing to the advancement of clean and efficient mobility solutions.
