Structural Optimization of the Permanent Magnet Drive Based on Artificial Neural Network and Particle Swarm Optimization

The objective of this paper is to optimize the structural parameters of the Permanent Magnet Drive by minimizing the manufacturing costs while maintaining the output performance. The electromagnetic analysis of the Permanent Magnet Drive is carried out by the Finite Element Method. In order to debasing the cost of calculation and time, the rapid calculation model which can map the relationship between the structural parameters and output torque is established by the Artificial Neural Network. The sample data for training are obtained by combining the Finite Element Method with the Orthogonal Test Design. The purpose of optimization is gained by using the Particle Swarm Optimization, and then the output performance of the new designs is calculated and compared with that of the orthogonal optimization. The results show that the new design obtained from the optimal method proposed in this paper has a reduction of approximately 20% in the magnet material but in the mean time with no loss of the output torque.