Abstract
The acquired bearing fault signal usually reveals nonlinear and non-stationary nature. Moreover, in the actual environment, some other interference components and strong background noise are unavoidable, which lead to the fault feature signal being weak. Considering the above issues, an effective bearing fault diagnosis technique via local robust principal component analysis (LRPCA) and multi-scale permutation entropy (MSPE) was introduced in this paper. Robust principal component analysis (RPCA) has proven to be a powerful de-noising method, which can extract a low-dimensional submanifold structure representing signal feature from the signal trajectory matrix. However, RPCA can only handle single-component signal. Therefore, in order to suppress background noise, an improved RPCA method named LRPCA is proposed to decompose the signal into several single-components. Since MSPE can efficiently evaluate the dynamic complexity and randomness of the signals under different scales, the fault-related single-components can be identified according the MPSE characteristic of the signals. Thereafter, these identified components are combined into a one-dimensional signal to represent the fault feature component for further diagnosis. The numerical simulation experimentation and the analysis of bearing outer race fault data both verified the effectiveness of the proposed technique.
Highlights
The bearing as an essential element has been widely used in rotating machinery [1,2]
The vibration signals of the bearings are often used for fault diagnosis for their containing abundant equipment operation information [6]
The bearing’s weak fault feature exhibits the nature of nonlinear and non-stationary, which is hard to be extracted under the situation of existing strong background noise and interference components
Summary
The bearing as an essential element has been widely used in rotating machinery [1,2]. In order to realize accurate diagnosis of bearing faults, suppressing the background noise and extracting weak fault features from multi-component signals are becoming an urgent work to be solved. For the early weak faults of the bearings, the singular values representing different feature components are almost impossible to be identified [25]. As a widely used dimension reduction method, the RPCA can extract this submanifold structure through a rank function constraint based on low-rank matrix approximation (LRMA) and simultaneously suppress background noise through a l0 -norm regularization strategy [28]. An effective bearing fault diagnosis technique via local robust principal component analysis (LRPCA) and MSPE is introduced. The identified low-rank matrices were transformed into a one-dimensional signal to represent the global approximation of the fault feature component for further diagnosis via weighted.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
