核主成分分析和粒子群优化支持向量机在电力机车笼型异步牵引电机故障诊断中的应用研究

提出了一种采用核主成分分析和粒子群优化支持向量机的电力机车笼型异步牵引电机故障诊断方法。先利用核主成分分析对故障数据进行特征提取,以获得的故障特征子集作为支持向量机故障分类器的训练样本,然后设计和构建了支持向量机多故障诊断系统。其中,支持向量机的参数通过粒子群优化算法进行了优化,最后实现对笼型异步牵引电机的故障诊断。实验结果分析表明,该方法能够有效地应用于电力机车笼型异步牵引电机的故障诊断。

Application Research on KPCA and PSO-SVM in Fault Diagnosis of Cage Asynchronous Traction Motor on Electric Locomotive

This paper proposes a fault diagnosis method of cage asynchronous .Traction motor on electric locomotive using kernel principal component analysis（KPCA）, particle swarm optimization（PSO） and support vector machine（SVM） . The kernel principal com- ponent analysis is first employed to extract main feature from fault data in order to ob- tain the fault feature subset which is used as training sample of SVM fault classifier, and then designing and building the mult-fault diagnosis system of SVM, the SVM parameter of which is optimized by PSO algorithm. The experimental result shows that the method can effectively be used to fault diagnosis of cage asynchronous traction motor on electric locomotive.