Pareto Optimum Design of the Magnetic Components in DAB Converters Based on Nondominated Sorting Genetic Algorithms-II
Zhiqiang Guo, Zi'ang Chen, Zhongyuan Chen
Abstract
This article presents a Pareto optimum design of the magnetic components in dual active bridge (DAB) converters based on nondominated sorting genetic algorithms-II (NSGA-II). A low switching frequency can reduce the switching loss, but the volume of the transformer will be increased, which may not meet the volume requirement. A high switching frequency can reduce the volume of the transformer, but it will decrease the efficiency. The multiobjective optimum design complicates the converter design. Although the triple phase-shift control can improve the efficiency of the DAB converter, how to increase the efficiency further by optimizing the magnetic component is still a difficult concern. The NSGA-II is used in this article to optimize the efficiency of the DAB converter and the volume of the magnetic components. Through crossover and mutation of the population in the genetic algorithm, the feasible solution is selected by the nondominated sorting. With the increase of the iterations, the population of the candidate solutions converges to a set of Pareto optimal solutions instead of a single optimal solution. The method can make a tradeoff between efficiency and volume. Finally, the prototypes based on Pareto optimal solutions are built to verify the accuracy of the proposed method.