Multi-objective optimization design for output characteristics of LCC resonant converter

摘要

The equivalent circuit and the mathematical model are obtained for the capacitive output filter series-parallel resonant converter by a fundamental model approximation (FMA) method in this paper. Based on them, a Pareto multi-objective optimization with genetic algorithm is applied to improve the output characteristics of the converter. With finegrained fitness assignment strategy and density estimation, the algorithm can achieve massive and well-distributed Pareto optimal solutions by the minimization of two duality functions. These optimal solutions correspond to the resonant circuit parameters and the control parameters with respect to the different corresponding performance targets. The Pareto multi-objective optimization algorithm can get some good results, which is proved by simulation results. The algorithm can also save the time of circuit debugging.