研究了一种结合独立成分分析和支持向量机的方法在交流电机故障诊断中的应用。首先通过检测各种电机振动和定子电流信号得到数据,利用独立成分分析对交流电机原始数据进行特征提取和压缩;主成分分析也同时应用于独立成分分析特征提取过程中,在完成故障识别时应用了支持向量机技术.采用的是连续最小优化算法和基于支持向量机分类的多类统计分类方法。同时分类过程选择了典型的核函数,以达到诊断电机故障的目的。试验分析的结果表明.该方法是一种简单而有效的方法。
It is studied the application of independent component analysis (ICA) and support vector machine (SVM) to fault diagnosis of induction motor. Data are collected from the detection of various motor vibration and stator current signals. The original data of the induction motor are extracted and compressed by way of ICA, and main component analysis. SVM is adopted for the fault identification, through continuous min. optimum algorithm and multiple statistics clarification based on classification of SVM. Typical kernel functions are selected during the clarification, for diagnosis of motor faults. The test and analysis results indicate that the method is simple and efficient.