基于正交设计和支持向量机的横磁通永磁电机参数优化  被引量:4

Transverse Flux Permanent Magnet Machine Optimization Based on Orthogonal Design and Support Vector Machines

在线阅读下载全文

作  者:包广清[1] 郑文鹏[2] 

机构地区:[1]兰州理工大学电气工程与信息工程学院,甘肃兰州730050 [2]中国电子科技集团公司第21研究所,上海200233

出  处:《应用基础与工程科学学报》2012年第5期912-921,共10页Journal of Basic Science and Engineering

基  金:国家自然科学基金项目(50877034);国家人力资源和社会保障部"留学人员科技活动优秀项目"(2009)

摘  要:漏磁系数是永磁电机的一个重要参数,它标志着磁路设计的合理性和永磁体的利用率.本文将正交设计、支持向量机以及粒子群智能优化算法相结合,提出一种横磁通永磁电机漏磁优化设计方法.首先,在基于标量磁位求解的三维有限元模型基础上,确定对空载漏磁系数敏感度较高的设计参数;其次,通过正交试验建立参数样本空间,并运用固定尺度最小二乘支持向量机实现电机漏磁系数的非线性回归建模,为电机优化所需的大规模迭代运算提供高效的计算模型;最后利用粒子群算法进行以空载漏磁系数最小为目标的寻优操作,并得到最优方案.对一台30kW聚磁式横磁通永磁同步电机进行参数优化,结果表明在满足输出转矩要求的约束条件下降低了漏磁,充分证明了建模和优化设计的正确性和工程实用价值.The flux leakage is a key factor for permanent magnet machine design, which closely related to the magnetic circuit efficiency and permanent-magnet material utilization. This paper presented an optimum design of a transverse flux permanent magnet machine (TFPMM) to reduce the flux leakage of machine with constraints of thrust torque. An orthogonal design method (ODM) combined with fixed size least squares support vector machines (FS-LSSVM) and particle swarm optimization (PSO) algorithm was proposed to reduce the leakage coefficient and improve the motor performance. The sensitive design parameters to flux leakage were investigated by the three-dimensional (3D) finite element analysis with magnetic scalar potential. The ODM yielded the most motor geometric information of trial samples from the fewest runs. The FS-LSSVM regression analysis addresses the difficulty of TFPMM modeling which enables objective function to be easily created and a great computational time to be saved. With the global search ability as well as the independence on original solution characteristics, PSO algorithm was employed as a searching tool for optimal solution. The numerical results obtained from a 30 kW prototype evaluated the effectiveness of the nonparametric modeling and the optimization method.

关 键 词:横磁通永磁电机 优化设计 正交设计 支持向量机 粒子群算法 

分 类 号:TM301[电气工程—电机]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象