单绕组磁悬浮开关磁阻电机结构优化设计

Optimization design of single winding bearingless switched reluctance motor

针对单绕组磁悬浮开关磁阻电机结构参数的优化问题,提出了基于相关向量机与混合粒子群优化的方法。通过有限元分析计算了电机悬浮力和转矩与结构参数的关系,构建基于相关向量机非参数模型。以电机平均悬浮力和平均转矩最大输出为优化目标,采用混合粒子群优化算法获取最优结构参数,通过对比仿真实验说明了该算法的准确性与优越性,电机性能得到明显提升。

For the parameters design of single winding bearingless switched reluctance motor（ SWBSRM）,a structure parameter optimal design method was proposed based on the relevance vector machine（ RVM） and the mixed particle swarm optimization（ MPSO）. It obtained sample space by finite element simulation and then built RVM nonparametric models of radial levitation force and torque. Based on these models,it selected average radial levitation force and torque as optimal objective. The optimal parameters of SWBSRM were obtained by MPSO searching. The simulation results prove that the algorithm is accuracy and superiority,and the performance of SWBSRM improved significantly.