A New Multimodal Optimization Algorithm for the Design of In-Wheel Motors

The selection of optimal parameters during the design of an electric motor is a multivariable and multimodal optimization problem that requires a considerable amount of computational calculation time. To solve this type of problem, this paper proposes a novel multimodal optimization algorithm that is assisted by a surrogate model using the newly developed compressed sensing theory. Its effectiveness is confirmed by comparing the optimization results for test functions with the results of conventional optimization methods. These results show that the proposed method has more rapid and accurate convergence characteristics than conventional approaches. To verify the feasibility of its application to electric motors, an in-wheel motor is designed using the proposed algorithm.