ANN BASED THREE-PHASE ISOLATED MULTILEVEL ACDC CONVERTER FOR DUAL ELECTRIC VEHICLE BATTERY CHARGING

Abstract

The rapid growth of electric vehicles (EVs) has intensified the need for efficient, reliable, and intelligent charging infrastructures capable of simultaneously charging multiple vehicles. This paper presents an Artificial Neural Network (ANN)-based control strategy for a threephase isolated multilevel AC–DC converter designed for dual electric vehicle battery charging applications. The proposed topology employs a multilevel converter integrated with a high-frequency isolation transformer to achieve galvanic isolation, enhanced power quality, and reduced voltage stress across switching devices. The ANN controller is implemented to regulate the DC output voltage and charging current under varying grid conditions and battery state-of-charge levels. The intelligent controller provides superior dynamic response, reduced steady-state error, and improved robustness compared with conventional proportional–integral (PI) controllers. Furthermore, the dual-output architecture enables independent charging of two EV batteries with effective power sharing and minimal circulating current. Simulation studies carried out in MATLAB/Simulink demonstrate that the proposed ANNcontrolled system achieves low total harmonic distortion (THD), near-unity power factor, fast transient performance, and improved charging efficiency. The results validate the suitability of the proposed converter for future smart EV charging stations.