Multi-Objective Optimization of a Transverse Flux Machine With Claw-Pole and Flux-Concentrating Structure

This paper describes a new method aiming at the multi-objective optimization of a transverse flux machine with a claw-pole and flux-concentrating structure (CPTFM). Considering the 3-D main magnetic flux in the CPTFM, the time-consuming 3-D finite-element method (FEM) should be used to accurately analyze the performance of the machine. In order to reduce the calculation time for the optimization of the complex CPTFM, the Taguchi method is first applied to evaluate the importance of all the design parameters in the consideration of the predefined five objectives. Subsequently, a mathematical model is developed to replace the 3-D FEM, so that the objectives depending on the selected four critical parameters can be analytically calculated. With its help, the four parameters are further optimized with the multi-objective particle swarm optimization method and nondominated sorting genetic algorithm 2. Finally, the optimal design of the CPTFM is selected from all the searched Pareto solutions with the help of fuzzy logic. It can be concluded that the developed method is effective and efficient, which can make an appropriate compromise between the calculation time and the improvement of the machine performance.