Multi-Objective-Optimized Parameter Design Method for High-frequency Resonant Converter

In recent years, the high-frequency resonant converter has become a research hotspot for soft switching and high-power density. However, the increasing number of resonant components adds difficulties to the resonant parameter design. For this reason, this paper proposes an optimized parameter design method for high-frequency resonant converters, which utilizes the back propagation neural network (BPNN) prediction model combined with a multi-objective genetic algorithm (MOGA). This method comprehensively considers the main parameters that can optimize efficiency. A prototype with 18V input, 5V/10W output and 10MHz switching frequency was built in the laboratory. Experimental results verify the effectiveness of the proposed parameter optimization method.