Multiobjective optimization design of high frequency transformer based on NSGA-II algorithm

Abstract

Aiming at the optimization design of high frequency transformer in low-voltage high-current power electronic converter, the volume and efficiency of high frequency transformer are taken as the optimization objectives, the calculation methods of magnetic core and winding loss of high frequency transformer are analyzed. On this basis, a two dimensional optimization model is established with saturated magnetic induction intensity and current density as optimization variables. Genetic algorithm has certain advantages in solving nonlinear multiobjective optimization problems. In this paper, NSGA-II algorithm is used to optimize the design of high frequency transformers with ferrite, amorphous alloy and nanocrystalline core materials. By comparing the optimization parameters of the simulation results, the superiority of nanocrystalline materials in the design of high frequency transformer is verified, due to its high saturation magnetic induction strength and high permeability, nanocrystalline materials have better comprehensive performance than ferrite and amorphous alloy materials in high-frequency and high power applications.