小波包和SVM在轴承故障识别中的应用

For the sake of solving the problem of bearing working state on identification , a method of bearing intelligent fault identification is proposed by means of the wavelet package-support vector machine. According to the method , the energy of frequency bands after wavelet packet decomposition of the vibration signals in different working states is taken as the eigenvectors and also as training samples of SVM multi-fault classifier. The result indicated that the method fused benefits of wavelet packet and support vector machine ,so can conduct identification of equipment fault effectively and achieve the purpose of precise fault diagnosis for mechanical system.