Open AccessBearing fault diagnosis based on wavelet packet energy spectrum and SVM
Author(s) -
Cheng He,
Tao Wu,
Runwei Gu,
Huaying Qu
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1684/1/012135
Subject(s) - wavelet , wavelet packet decomposition , support vector machine , energy (signal processing) , pattern recognition (psychology) , fault (geology) , computer science , network packet , bearing (navigation) , vibration , frequency band , artificial intelligence , algorithm , speech recognition , wavelet transform , mathematics , telecommunications , acoustics , statistics , physics , bandwidth (computing) , computer network , seismology , geology
Aiming at the limitation of wavelet analysis in fault diagnosis, combining wavelet packet energy spectrum and support vector machine algorithm, a fault diagnosis method based on wavelet packet energy spectrum and support vector machine algorithm (SVM) is proposed. This method first performs wavelet packet transformation on the test data, and the vibration signal is decomposed into independent frequency bands. The signal energy changes in different frequency bands reflect the change of the operating state, and the wavelet packet energy spectrum of each frequency band is extracted as a feature vector. Finally, the SVM algorithm is used to test the bearing faults. The test results show that after processing and analysis of a large number of measured bearing data, the fault of the bearing can be diagnosed relatively accurately.