Machine Learning Based Control of Integrated Drive and Chagre Mode Bi-Modular Power Converter for PV Assisted Light Electric Vehicle

This paper presents the topology and machine learning based power control of unified bi-modular inverter for drive and charge operations of PV assisted battery electric vehicle. A multilevel bi-modular inverter is considered for integrated simultaneous PV and battery power conversion which also configured for charging of battery from PV source with locked rotor acting as interfacing inductor. The proposed pattern recognition type ANN based machine learning control serves for power conversion control for multi-mode operation with maximum power point operation. The proposed system is implemented using MATLAB/SIMULINK. The performance of the proposed system is verified through simulation results for effectiveness of machine learning, accuracy of closed loop power conversion. A comparison on sizing of power converter is presented to verify the structural advantages of the proposed system.