Magnetic Gear Design Optimization by Genetic Algorithm with ANN Controlled Crossover and Mutation

This paper presents a design optimization by Genetic Algorithm (GA) with controlled crossover and mutation by Artificial Neural Network (ANN). Co-axial Magnetic Gear (MG) design has been put under consideration for that research. Optimization process aims the improvement of MGs torques interaction by design sizes variation. Six different optimization parameters are defined which represents the MG design radiuses. ANN is trained for the optimization problem parameters preselection as a part of the GA optimization algorithm. Solution genomes crossover and mutation are controlled during new generation creation. The results obtained for MG design optimization shows the effectiveness of the proposed method, which, we believe, is suitable for wide variety of design optimization problems.