电机故障诊断支持向量机

基于数据的机器学习是现代智能技术中的重要方面。统计学习理论(Statistical learnmgtheory SLT)是研究小样本情况下机器学习规律的新理论。支持向量机(Support vector machine SVM)是在这一理论体系基础上发展起来的一种通用学习方法。SLT和SVM正成为继神经网络研究之后新的研究热点。通过对鼠笼式异步电动机转子断条故障进行实验模拟，对实验获取的采样电流信号经FFT分析，构造以低频到高频的频谱特性为分量的学习样本向量，通过支持向量机SVM对故障电流样本的训练，使SVM具有分类功能。最后，采用SVM对电动机各种转子断条故障进行诊断分类，取得较满意的结果，说明支持向量机SVM是进行故障诊断的一种新方法。

Support Vector Machine for Fault Diagnosis of the Rotor Broken Bars of IM

The data-based machine learning is an important aspect of modern intelligent technology, while statistical learning theory(SLT) is a new tool that studies the machine learning methods in the case of a small number of samples. As a common learning method, support vector machine(SVM) is derived from SLT. Some analogical experiments of the rotor broken bar faults of induction motors were done. The signals of the sample currents are analyzed with FFT, and the spectral characteristics from low frequency to high frequency are constructed and used as learning sample vectors for SVM. Accordingly a SVM is trained with learning sample vectors, so that each kind of the rotor broken bars faults of induction motors had classified for SVM. Finally the retest proves that SVM really has preferable ability of classification, and the results suggested that SVM could yet be regarded as a new method in the fault diagnosis.