基于支持向量机的旋转机械故障诊断

为了解决旋转机械故障的在线诊断识别问题，用小波包从旋转机械的震动信号中提取特征向量，给出了一种基于支持向量机的故障诊断分类方法。该方法通过有限的学习样本，建立旋转机械故障特征与其运行状态之间的关系。利用获得的矿井提升机减速箱齿轮数据建立了多级故障分类器，通过对样本的分类输出检验，验证了该故障诊断方法的可行性。

Rotating machinery fault diagnosis based on support vector machine

To identify the rotating machinery fault,the method based on support vector machine(SVM),by extracting feature vector from it with wavelet packet,was proposed.The principle and algorithm of this method were introduced.Through a finite learning sample the relation between the rotating machinery fault signature and its' running state was established.A multi-faults classifier was constructed by using the gearbox data of the mine hoist.The testing results show that this method can successfully be applied to the di-agnosis of gear faults.