基于最小二乘支持向量机的五相容错永磁电机匝间短路故障诊断

Short circuit fault diagnosis of five-phase fault-tolerant permanent-magnet motor based on least squares support vector machine

随着电机驱动系统的应用领域越来越广泛,对电机的安全性能提出越来越高的要求,多相电机比三相电机具有更好的可靠性和容错性,其故障诊断技术成为研究的热点.针对五相容错永磁电机的匝间短路故障进行研究,分析支持向量机(SVM)与最小二乘支持向量机(LS-SVM)在该电机匝间短路故障诊断中的应用机理,构造多故障诊断分类器,建立基于LS-SVM的多故障诊断模型,采用粒子群优化算法进行参数寻优,测试结果表明:LS-SVM方法在快速性和精度上均有所提高,证明此方法用于五相容错永磁电机匝间短路故障诊断的可行性.

Since motor drive has wide applications in many fields, high reliability of motor drive is required in industrial applications. The Multi-phase motor has better reliability and fault tolerance performance than the three-phase motor, and its fault diagnosis technology has become a hot research. In this paper, five-phase fault- tolerant permanent-magnet motor of inter-turn short circuit fault is investigated. Support vector machine （SVM） and least squares support vector machines （LS-SVM） are analyzed and discussed. The multi-fault diagnosis clas- sifier is constructed and the multi-fault diagnosis model based on LS-SVM is established. The PSO algorithm is applied for parameter optimization. Test results show that LS-SVM method has improved the speed and precision. The feasibility of the short-circuit fault diagnosis method for five-phase fault-tolerant permanent-magnet motor is proved.