基于小波包和支持向量机的故障诊断方法研究

基于支持向量机（SVM,support vector machine）对小样本决策具有较好的学习推广性,本文提出一种基于小波包和支持向量机的故障诊断方法,通过小波包分解系数求取频带能量,并根据各个频带的能量的变化提取故障特征,应用LSSVM（least squares support vector machines）进行故障分类。实验结果表明,支持向量机分类器优于传统的BP神经网络和RBF神经网络分类器,识别率较高,具有更强的泛化推广能力。

Fault diagnosis method based on wavelet packet and support vector machine

For support vector machine (SVM) exhibits good generalization characteristics when fault samples are few,the paper present a fault diagnosis method based on wavelet packet and SVM.This method strikes the energy of frequency band through the decomposition coefficient of wavelet packet and gains the fault characteristics from various changes in the energy of each frequency band,and classify fault with least squares support vector machines(LSSVM).The experimental results show that the capability of SVM is more...