Bearing fault diagnosis based on Lie group classifier

This paper briefly describes the framework of Lie group classifier, then Lie group classifier is introduced to detect fault of bearings, aiming at the characteristics of bearing fault vibration signals. Firstly, training feature set and test feature set are constructed from fault vibration signal. The two sets consist of mean value, energy, root-mean-square value, peak value, crest factor, kurtosis, shape factor, clearance factor. Secondly, training feature set is applied to Lie group classifier to compute classifier parameters. Thirdly, bearing fault is diagnosed by Lie group classifier based on test feature set. The results show that this method can detect fault with high accuracy rate and it presents a new method for bearing fault diagnosis.