关键词: 滚动轴承, 最小二乘支持向量机, 半监督学习, 故障诊断

Abstract:

To improve the rate of correct training of the fault identification sorter of rolling bearing and shorten the training time,LS_SVM is combined with semi-supervised learning according to the fact that the training sets have both labeled and unlabeled examples.Full use is made of the effective information in the training sets and a novel SLS_SVM based fault identification method for rolling bearing is proposed.A comparison of this method with the standard SVM and semi-supervised learning based on the SVM method shows that this method can not only improve the rate of correct training but also shorten the training time.The diagnostic tests show that it is an effective and efficient method.

Key words: rolling bearing;LS_[KG-2mm]SVM;semi-supervised learning;fault identification