An Effective Requests-Based Charging Station Scheduling and Routing Model with HUI-MTA ANN and BD-SSO for EV Charging

Electric Vehicles (EVs) rely on charging stations (CSs), but uncoordinated charging often leads to grid voltage deviations and long waiting times. This study proposes an integrated scheduling and routing framework that combines renewable-powered CSs with intelligent decision-making. A Singleton Membership Modified Fuzzy Logic Controller (SMM-FLC)-based MPPT ensures stable renewable power extraction, while a He-Uniform Initialized ModTanh Activated ANN (HUI-MTA-ANN) selects the optimal CS by considering queue lengths, travel deadlines, and cost. The Bayes Distributed Squirrel Search Optimization (BD-SSO) algorithm then identifies the most efficient route, and Quadratic Programming (QP) schedules charging slots. Results on an EV charging dataset show the framework achieves 97% accuracy with very low false selection rates, while significantly improving renewable energy utilization. The findings highlight a scalable and intelligent solution for reliable EV charging infrastructure.