Risk Assessment and Multi-objective Optimization of Current Sharing for Parallel- connected Multiple Modules of DAB Converters

The DAB (Dual - Active Bridge) converter has received significant attention in the fields of automotive electronics and aerospace due to its bidirectional power - flow characteristic. In the context of an increasing demand for higher output currents in power - supply products, the ISOP (Input - Series, Output - Parallel) system offers an effective solution to reduce the input voltage of individual modules and achieve output current sharing. However, multi - module converters are plagued by several drawbacks. These include slow dynamic responses, high sensitivity to parameters, and challenges in accommodating manufacturing tolerances and input disturbances. As a result, ensuring output current sharing in ISOP systems remains a formidable task. This paper focuses on the DAB converter under EPS (Extended Phase - Shift) modulation. An evaluation framework for current - sharing risks in ISOP - DAB systems under EPS modulation is presented. This framework quantifies the impact of inductance value discrepancies and input - voltage dispersions on output current sharing, establishing a parameter - uncertainty propagation model to expose the risk boundaries under the combined influence of multiple factors. Furthermore, a multi - objective optimization genetic algorithm is employed to optimize the inductor current stress and power reflux. Finally, MATLAB simulations are conducted to validate the accuracy of the theoretical derivations presented in this study.