Laser & Optoelectronics Progress, Volume. 59, Issue 21, 2107001(2022)
Feature Extraction of Bearing Faults Under Nonlinear Equalization of Variance Based on Wavelet Packet Decomposition
When the bearing has various faults, the vibration signal's variance distribution in various frequency bands is not balanced. In order to extract the features of various fault signals efficiently and for the frequency components generated from the bearing fault signal's wavelet packet decomposition, this study proposed a variance equalization method of nonlinear equalization. The higher discrimination degree of fault characteristics can In order to extract the features of various fault signals efficiently and for the frequency components generated from the bearing fault signal's wavelet packet decomposition, this study proposes a variance equalization method of nonlinear equalization. The higher discrimination degree of fault characteristics can be achieved. Based on the data from Case Western Reserve University's bearing data center's collected bearing vibration in the experiment, the variance parameters extracted from a normal, inner ring fault, outer ring fault, and rolling element fault bearing signal under four speeds are investigated using this approach. The results reveal that the variance parameters of various fault signals after equalization have better discrimination. It can efficiently differentiate the types of bearing faults.
Get Citation
Copy Citation Text
Zhiqing Zheng, Haiyan Quan, Junbing Qian. Feature Extraction of Bearing Faults Under Nonlinear Equalization of Variance Based on Wavelet Packet Decomposition[J]. Laser & Optoelectronics Progress, 2022, 59(21): 2107001
Category: Fourier Optics and Signal Processing
Received: Aug. 30, 2021
Accepted: Oct. 25, 2021
Published Online: Oct. 24, 2022
The Author Email: Qian Junbing (1226160701@qq.com)