A stacking machine‐learning based method for accelerating magnetic coupler design with ferrite cores in inductive power transfer applications

Abstract This paper presents a magnetic coupler design method for WPT systems based on stacking machine‐learning algorithms. A synthetic dataset generated by ANSYS Maxwell is used for training and evaluating machine‐learning models. Stacking technology effectively combines the results from multiple models and make best predictions. The designed model allows us to obtain the optimal values of the coil inner radius and number of turns , when other coil design parameters such as the coil outer radius , the wire diameter , and the coil inductance , are given based on the application environment. The proposed method provides practical solutions meeting design requirements quickly, easily accommodating magnetic couplers with ferrite cores, and showing advantages in designing magnetic couplers with optimal power transfer efficiency.