2-Step Monte Carlo Tree Search for Optimal Design of High-Frequency Toroidal Inductors in Power Electronics Circuits

This paper proposes an optimization technique for toroidal inductors utilizing a 2-step Monte Carlo tree search. The approach combines a global search utilizing analytical formulas with a local search involving three-dimensional finite element analysis. Through the proposed method, optimal core materials, sizes, number of turns, and windings are determined to maximize the efficiency and minimize the size of the inductors. The proposed method is applied to the optimization of the inductors in a 6.78MHz electric-coupling power transfer system. The proposed method found the optimal inductor with low computational cost and the system measurement confirmed that the optimized inductor improves over 1.8% of the system efficiency compared to the previously designed inductor.