Cogging Torque Reduction and Optimization in Surface-mounted Permanent Magnet Motor Using Magnet Segmentation Method

One of the main challenges in permanent magnet electrical machine design is cogging torque reduction. In this article, the magnet segmentation method is used for cogging torque reduction. For this end, each surface permanent magnet is divided into eight parts, and a symmetrical structure with equal angular widths and considering the angular gaps between them is used for minimizing a number of optimization parameters. In this article, three optimization algorithms—response surface methodology, genetic algorithm, and particle swarm optimization—are used to determine the optimal values of optimization parameters. Finally, the result is obtained that the optimum values of response surface methodology are more efficient than of those of the genetic algorithm and particle swarm optimization in cogging torque reduction, because the objective function of the response surface methodology is cogging torque that is calculated using the finite-element method, whereas the objective function in the genetic algorithm and particle swarm optimization is based on the analytical methods. However, the main objection of the magnet segmentation method is the simultaneous reduction of average torque with cogging torque.