基于Kriging模型的永磁电机优化设计  

Optimization of Permanent Magnet Motor Based on Kriging Model

作　　者：应志平 卢琴芬 包广清[3] 任彦浩 YING Zhiping;LU Qinfen;BAO Guangqing;REN Yanhao(College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China;College of Electrical Engineering,Zhejiang University,Hangzhou 310027,China;School of Electronics and Information Engineering,Southwest Petroleum University,Chengdu 610500,China)

机构地区：[1]兰州理工大学电气工程与信息工程学院,兰州730050 [2]浙江大学电气工程学院,杭州310027 [3]西南石油大学电气信息学院,成都610500

出　　处：《微特电机》2023年第5期7-12,共6页Small & Special Electrical Machines

基　　金：国家自然科学基金项目(51967012)。

摘　　要：针对空压机用高效电机采用有限元法进行优化耗时长、效率低的问题,提出了一种基于Kriging模型的优化方法,以减少有限元调用次数和提高优化效率。以功率因数和效率为优化目标,同时考虑永磁体用量,对电机的结构参数进行敏感性分析,选择永磁体磁化方向长度、永磁体宽度和轴向铁心长度作为优化变量;基于改进的拉丁超立方抽样和有限元获得样本集,建立Kriging模型代替有限元进行迭代优化,优化过程中有限元调用次数减少了83%,节省了计算成本;通过有限元进行优化前后的对比分析,验证了该优化结果的可行性和正确性。优化后电机的功率因数与效率得到改善,且永磁体用量减少了9.94%,降低了电机的材料成本。An optimization method based on Kriging model was proposed to solve the time-consuming problem of using finite element method to optimize the high-efficiency motor for air compressor by reducing the number of finite element calls.With power factor and efficiency as the optimization objectives considering the amount of permanent magnet,sensitivity analysis was carried out on the structural parameters of the motor,and the magnetization direction length of permanent magnet,permanent magnet width and axial core length were selected as the optimization variables.The Kriging model based on the improved Latin hypercube design and finite element method to obtain the sample set was established to replace the finite element method for iterative optimization,which reduced the number of finite element calls by 64%during the optimization process and saved the calculation cost.The feasibility and correctness of the optimization results were verified by comparison before and after optimization with the finite element.The power factor and efficiency of the optimized motor were improved,and the amount of permanent magnet was reduced by 9.94%,which can reduce the material cost.

关 键 词：内置式永磁同步电机 优化设计 Sobol模型 KRIGING模型 遗传算法 

分 类 号：TM351[电气工程—电机]