Feature extraction for rolling element bearing weak fault based on MOMEDA and ICEEMDAN

Yongxiang Zhang,Danchen Zhu,Lei Zhao

doi:10.21595/jve.2018.19309

Abstract

There are always the nonlinear and non-stationary characteristics and periodic pulse in vibration signals of rolling element bearings when there are partial faults in those bearings. Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) overcomes the presence of spurious modes and residual noise in Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), but it cannot clearly and accurately extract the weak fault feature of rolling element bearings under the strong background noise. Here, Multipoint Optimal Minimum Entropy Deconvolution Adjusted (MOMEDA) was proposed. A bearing simulator was used to collect vibration signals of bearing inner and outer race, which was enhanced by MOMEDA, decomposed into several Intrinsic Mode Functions(IMFs) by ICEEMDAN, and analyzed by the envelope demodulation, finally gaining the frequency of shaft speed, BPFI (ball pass frequency, inner race) and harmonics, sidebands spaced, BPFO (ball pass frequency, outer race) and harmonics. The results show that this method can be used to accurately extract different frequency components of bearing fault vibration signals and diagnose bearing different fault location.

Highlights

Rolling element bearings are one of the most common components that are widely used in rotating machinery, and their failure is one of the most frequent reasons for machine breakdown
In order to demonstrate the advantages of ICEEMDAN, The Empirical mode decomposition (EMD), ensemble EMD (EEMD), CEEMDAN, and ICEEMDAN are compared by simulation signal
In order to verify the validity of the method combining Multipoint Optimal Minimum Entropy Deconvolution Adjusted (MOMEDA) with ICEEMDAN, a simple fault model is used to simulate the periodic impact signal produced by the bearing fault, and the white noise is added to simulate the early fault signal of the bearing [12]: x(t) = x exp(−2πf ξt)sin2πf 1 − ξ t + n(t), (17)

Summary

Introduction

Rolling element bearings are one of the most common components that are widely used in rotating machinery, and their failure is one of the most frequent reasons for machine breakdown. There are always the nonlinear and non-stationary characteristics and periodic pulse in vibration signals of rolling element bearings when there are partial faults in those bearings. As an adaptive signal processing method, it remains the completeness of EMD, overcomes mode-mixing phenomenon of EMD and suppresses the reconstruction error [8]. It still has two drawbacks: the existence of spurious modes and the presence of residual noise in the modes. In order to enhance the periodic impact component in the vibration signal of bearing fault, and reduce the noise interference, MOMEDA was proposed [10]. As a non-iterative deconvolution approach, MOMEDA settles a deconvolution problem by an infinite impulse train as the goal and the optimal filter solution can be given directly

The Principle of the MOMEDA

The principle of the ICEEMDAN

Comparison analysis of simulation signals

Bearing fault simulation signals

Experimental verification

Conclusions