Passive variable switching frequency sinusoidal pulse width modulation (PVSFSPWM) can effectively reduce the loss and improve the current ripple by reasonably selecting injection function and depth. Aiming at the problem that the injection function and depth are limited in improvements in the system performance under the PVSFSPWM method due to artificial experience

this article presents an optimized PVSFSPWM based on the nondominated sorting genetic algorithm-Ⅱ (NSGA-Ⅱ)

abbreviated as NSGA-Ⅱ-PVSFSPWM. Switching loss and current ripple are used as the optimization objectives

and each switching cycle of the PVSFSPWM is optimized by the NSGA-Ⅱ algorithm to select a reasonable injection function and depth. A 1 kW full-bridge inverter is built. Experimental results show that

compared with the constant switching frequency sinusoidal pulse width modulation (CSFSPWM)

the peak harmonic of the output bridge arm voltage is decreased from 185.77 V to 78.35 V under the proposed method. The current ripple is decreased by 3.92%. The efficiency is improved by 1.32%. Compared to the PVSFSPWM method before optimization

the current ripple is reduced by 2%. The efficiency is improved by 1.04%. The results show that the proposed method can be adopted to effectively suppress the peak value of harmonic and to improve the system performance

which is valuable for engineering applications.