Multiobjective Optimization Approach for Common-Mode Voltage Suppression Considering Drive and Parasitic Parameters Based on NSGA-II

This article investigates the impact of SiC MOSFET module’s drive and parasitic parameters on common-mode voltage (CMV). A time-domain model of the switching noise source is derived, considering drive and parasitic parameters, and its frequency spectrum is obtained through the Fourier transform. The propagation path of CM noise based on a double-pulse test circuit is, furthermore, analyzed. Second, an ANSYS/Simplorer-based simulation test platform is established to evaluate the comprehensive sensitivity analysis of the impact of various parameters on electrical stress, loss, and CMV. Based on this, this article proposes a multiobjective optimization algorithm where the simulation solution is used as the initial solution set. The optimization of the parameters with high comprehensive sensitivity is performed using a multiobjective genetic optimization algorithm, and the optimal design was selected from the Pareto front generated by Nondominated Sorting Genetic Algorithm (NSGA-II). The importance of these proper drive parameters is verified by a 1500 V/150 kW Boost converter. The results show that the optimized selected drive parameters operating at rated conditions, provide attenuation of up to 5.2 dB in the 0.15–30 MHz frequency bands, and reduce the voltage stress and the current stress by 33.8 V and 31 A.