Machine Learning Controller for Optimised Charging in Solar Power Fed EV Battery Swapping Stations

In this era of electric vehicles, battery swapping stations (BSS) have a significant role as they help EV owners and station operators by providing fast, reliable, and convenient solutions to overcome driving range anxiety, high charging time, and high cost of electric vehicles. If these battery swapping stations are fed with renewable energy sources, and an optimized charging infrastructure is integrated to it, this can be made zero-carbon, reliable, convenient, and economical for EV users and station operators. In this work, the authors designed and analyzed a solar power fed battery swapping station, with pattern recognition network-based predictor and controller for charging its batteries. An open dataset collected from the Georgia Tech University campus is selected for the case study. Simulation analysis is carried out using PVSyst, Homer grid, MATLAB/ Simulink, and the results show 83% accuracy in predicting healthy charging rates of the batteries.