A Weak Fault Diagnosis Method Based on Sparsity Overlapping Group Lasso for Rolling Bearing

The rolling bearing is a vital part of modern equipment. It has great significance to effectively extract fault periodic pulses from bearing signals. Due to the noise mixed in bearing signal, the extraction of weak fault feature is facing difficult. For this, a new method for diagnosis of rolling bearing is developed in this paper, based on weighted sparsity norm and overlapping group sparse/shrinkage (OGS), which called sparsity overlapping group lasso (SOGL).The signal obtained from working rolling bearing is prone to be interfered with noise and occur coupling in time domain, so the algorithm processes signal in the Fourier domain, and combines with iterative shrinkage threshold (ISTA) and majorization–minimization (MM) to solve the model proposed in this study. Because signal mixed with noise will reduce the processing accuracy, based on the variational mode decomposition (VMD), this paper derives a simple and fixed center frequency filtering algorithm to reduce the useless components in the signal and improve the effect of SOGL. The SOGL is applied to simulated signals and measured signals. From the experimental results, the weak fault periodic pulse obtained by SOGL is obviously prominent, and this method has better performance in feature extraction compareng with the method of fast Fourier transform and OGS (FFT-OGS).

[1]  M. Yuan,et al.  Model selection and estimation in regression with grouped variables , 2006 .

[2]  K. Gröchenig.  Describing functions: Atomic decompositions versus frames , 1991 .

[3]  Robert B. Randall,et al.  Rolling element bearing diagnostics—A tutorial , 2011 .

[4]  Enrico Zio,et al.  Artificial intelligence for fault diagnosis of rotating machinery: A review , 2018, Mechanical Systems and Signal Processing.

[5]  Ivan W. Selesnick,et al.  Translation-invariant shrinkage/thresholding of group sparse signals , 2013, Signal Process..

[6]  Stéphane Mallat,et al.  Matching pursuits with time-frequency dictionaries , 1993, IEEE Trans. Signal Process..

[7]  Fulei Chu,et al.  Recent advances in time–frequency analysis methods for machinery fault diagnosis: A review with application examples , 2013 .

[8]  Zhibin Zhao,et al.  Enhanced Sparse Period-Group Lasso for Bearing Fault Diagnosis , 2019, IEEE Transactions on Industrial Electronics.

[9]  Ivan W. Selesnick,et al.  Translation-invariant shrinkage/thresholding of group sparse signals , 2014, Signal Process..

[10]  Michael Elad,et al.  On the Role of Sparse and Redundant Representations in Image Processing , 2010, Proceedings of the IEEE.

[11]  Joel A. Tropp,et al.  Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit , 2007, IEEE Transactions on Information Theory.

[12]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[13]  Y. C. Pati,et al.  Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[14]  Yaguo Lei,et al.  A review on empirical mode decomposition in fault diagnosis of rotating machinery , 2013 .