基于小波包分解和支持向量机的机械故障诊断方法

提出应用小波包分解和支持向量机进行机械故障诊断的方法。该方法将振动信号小波包分解后的频带能量作为特征向量，输入到由多个支持向量机构成的多故障分类器中进行故障识别和分类。试验结果表明，与神经网络相比，采用支持向量机进行故障诊断可以获得更高的诊断精度，表明该方法是有效的、可行的。

METHOD OF MACHINERY FAULT DIAGNOSIS BASED ON WAVELET PACKET DECOMPOSITION AND SUPPORT VECTOR MACHINE

Since there are some problems in machinery fault diagnosis, such as difficulty in getting adequate fault data samples, extracting feature vectors and acquiring fault diagnosis knowledge, a novel method of machinery fault diagnosis based on wavelet packet decomposition and support vector machine is proposed. According to the method, the energy of different frequency bands after wavelet packet decomposition constitutes the input vectors of support vector machine as feature vectors. And these feature vectors are inputted into multiple fault classifiers to identify faults. The new method, by which multiple faults can be diagnosed, only requires a small quantity of fault data samples and doesn't know the empirical knowledge of fault diagnosis. Compared with artificial neural networks(ANN),the method can achieve excellent diagnosis accuracy, and the results prove the method is efficient and feasible.