Improved Multiphysics Multiobjective Optimization for U-Shaped IPMSM Equipped With Eccentric Rotor
Feng Liu,Xiuhe Wang,Hongye Wei
DOI: https://doi.org/10.1109/jestpe.2023.3298010
IF: 5.462
2023-01-01
IEEE Journal of Emerging and Selected Topics in Power Electronics
Abstract:Benefiting from high efficiency, high-cost performance, and high operational reliability, interior permanent magnet synchronous motors (IPMSM) have attracted much attention in many emerging technology fields such as electric vehicles. To further enhance the comprehensive properties and engineering practicability, a U-shaped IPMSM equipped with an eccentric rotor is introduced, and a novel multiphysics multiobjective optimization model is innovatively proposed, studied, and developed. In this model, the key characteristics of electromagnetic, temperature, and mechanical fields are all selected as optimization objectives, realizing multiphysics collaborative optimization. First, nine key parameters that determine topology are selected as design variables, and five multiphysics characteristics are selected as optimization objectives or constraints. After that, in the data collection stage, segmented orthogonal experiments and sensitivity analysis are employed to ensure optimization efficiency. Then, the machine learning algorithm based on support vector regression and improved non-dominated sorting genetic algorithm (NSGA-III) are perfectly combined, and an effective and reliable optimal scheme is obtained while fully guaranteeing prediction accuracy. Finally, a comprehensive comparative analysis is carried out through simulation analysis and prototype experiments. The validity and merits of the studied U-shaped IPMSM and optimization model are excellently verified.
engineering, electrical & electronic