Multi-objective parameter optimization of permanent magnetic eddy-current couplings using multiple population genetic algorithm

作者：Wang, Dazhi; Shi, Tongyu^*; Wang, Xingyu; Wang, Xinghua

来源：International Journal of Applied Electromagnetics and Mechanics, 2017, 55(3): 359-372.

DOI：10.3233/JAE-160131

• Permanent magnet eddy-current couplings
• analytical method
• entropy-based weight
• multi-objective optimization
• multiple population genetic algorithm

The purpose of this paper is to obtain a 3D analytical model of permanent magnetic eddy-current couplings and optimize several parameters by using the multiple population genetic algorithm (MPGA). Firstly, on the basis of the analytical model, the formulas of key parameters are deduced. By using permanent magnet thickness, pole-arc coefficient and copper plate thickness as variables and taking output torque in steady state, rotational inertia and the volume of the couplings as optimization goals, this paper proposes a multi-objective optimization function with entropy coefficients and uses the MPGA to optimize structure parameters. The validity and feasibility of the proposed method is proved through 3D finite element analysis (3D-FEA) and experimental results. The results confirm that compared with another two optimization algorithms from the papers which had already published, optimization by the multiple population genetic algorithm based on the 3D analytical model has good effect on optimization of structural parameters.

• 出版日期2017
• 单位东北大学