Improved Multiobjective Optimization for Interior Permanent Magnet Synchronous Motors

A multiobjective optimization model for IPMSM is innovatively proposed. To achieve simultaneous improvements in multiphysics performance, magneto-thermal coupling characteristics are considered in this model. First, in the sample data collection stage, segmented orthogonal experiments and correlation analysis are introduced to improve optimization efficiency. After that, a neural network algorithm optimized is selected to ensure the model prediction accuracy. Further, the optimal design scheme is obtained by an improved non-dominated sorting GA. Finally, a comprehensive comparative analysis including electromagnetic characteristics and temperature rise performance is implemented through finite element method (FEM) and computational fluid dynamics (CFD). The validity of proposed optimization model is excellently verified.