Design of magnetic resonant WPT systems <Abstract> This paper proposes a machine learning-based design for a load-independent wireless power transfer (WPT) system with the class-EF inverter topology. The fundamental concept of the proposed design strategy is to solve an optimization problem for achieving high-frequency load-independent operation by full numerical computations. An optimization evaluation function is formulated, considering output voltage and powerdelivery efficiency. The WPT system, designed through the proposed method, achieves a constant output voltage and zerovoltage switching (ZVS) against load variations. The quantitative agreements between experimental and numerical waveforms substantiate the validity of the proposed design. |