An Improved Genetic Algorithm for Hybrid Magnetic Core Design in Electric Vehicle Wireless Charging Applications

This paper proposes an improved nondominated sorting genetic algorithm (NSGA-II) to optimize the distribution of hybrid magnetic cores in electric vehicle wireless charging applications. By considering the physical behaviors of different magnetic materials, the superior advantages of the ferrite core and the nanocrystalline core can be utilized and the computational cost can be largely reduced. The optimized hybrid core structure can achieve a 7% weight reduction and 16% cost saving compared to the traditional ferrite core design without sacrificing the overall performance. The experimental results show that under a coupling coefficient k of 0.27, the transfer efficiency of the WPT system with the proposed hybrid magnetic core can reach 95.88% at a transfer power of 6.6kW, which is nearly identical to the traditional full ferrite structure.