A Bidirectional Fast EV Charger for Wide Voltage Range Using Three-Level DAB Based on Current and Voltage Stress Optimization

Converter cost is the crucial concern while designing off-board electric vehicle (EV) chargers for high-voltage charging applications using advanced high breakover voltage power semiconductor switches. To solve this problem, multilevel topology is used with the typical dual active bridge (DAB) for the development of high-voltage and high-power EV charging. It offers low cost, compact, less complex control, highly efficient, and reliable development. This article presents the design and development of bidirectional off-board fast EV charger with voltage and current stress optimization for wide voltage range. It is designed with two-stage power conversion. Here, active front-end converter is connected to three phase grid and controlled through synchronous reference frame (SRF) d-q control approach for active rectification. The large range of power and voltage variation in off-board EV charger design makes it challenging to develop a controller that allows the converter to operate inside the zero voltage switching (ZVS) operating region. In this context, a detailed circuit analysis of dc–dc stage with dual phase shift (DPS) control under different operating conditions is presented to identify the coordinates of safe operating region with reduced voltage stress. To select the converter control parameters with optimum current stress in conjunction with reduced voltage stress, particle-swarm-optimization technique is applied to the identified operating regions. This provides a unique selection criterion for both control parameters over wide voltage conversion ratio. A coordinated control is designed based on the identified boundaries for easy implementation and less computation burden on microcontroller units (MCUs). A 7.2-kW system is designed and simulated on MATLAB/Simulink. A 3.3-kW laboratory prototype is designed to validate the identified operating points with wide battery voltage range under varying source and loading scenarios.