Application of the horizontal slice of cyclic bispectrum in rolling element bearings diagnosis

• Published in: Mechanical systems and signal processing Vol. 26; no. JAN; pp. 229 - 243
• Main Authors: Zhou, Y., Chen, J., Dong, G.M., Xiao, W.B., Wang, Z.Y.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 2012; Elsevier
• Subjects: Applied sciences; Bearings; Bearings, bushings, rolling bearings; Cyclic bispectrum; Drives; Exact sciences and technology; Fault diagnosis; Faults; Fundamental areas of phenomenology (including applications); Horizontal; Industrial metrology. Testing; Mechanical engineering. Machine design; Mechanical systems; Physics; Rolling element bearing; Solid mechanics; Spectrum analysis; Statistical methods; Structural and continuum mechanics; Three dimensional; Vibration, mechanical wave, dynamic stability (aeroelasticity, vibration control...); Fault diagnosis; Rolling element bearing; Cyclic bispectrum; Statistical analysis; Vibration; Probabilistic approach; Redundancy; Amplitude spectrum; Periodicity; Cyclic method; Frequency domain method; Non stationary process; Background noise; Experimental study; Cyclostationary; Spectral analysis; Rolling bearing; Statistical method; Signal envelope; Order parameter; Non linear effect; Fault diagnostic
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2011.07.006

The vibration signals of rolling element bearings are random cyclostationary when they have faults. However, because the background noise is very heavy when the early fault occurs, it is difficult to disclose the latent periodic components successfully even using the second order cyclostationary analysis. To overcome this difficulty, the cyclic bispectrum (CBS), an alternative approach based on bispectrum and cyclostationarity analysis, is discussed in this paper. Furthermore, the slice spectrum analysis of the CBS is proposed. The CBS is a third-order cyclic statistical parameter, in the frequency domain. The CBS gives full play to the advantage, which is provided from the higher order cyclic statistical methods. It can restrain noise and provide more information than classical methods such as amplitude spectrum analysis and envelope analysis when the fault is at an early stage. However, the CBS is four-dimensional. So, the Slices Spectrum Analysis of the CBS is introduced to fault diagnosis. Firstly, the slice of the CBS, which is 3D structure, is sliced along the cyclic frequency axis. The CBS corresponding to one cyclic frequency is called the once slice of the CBS. However the information in the once slice is redundant and indirect for fault diagnosis. So the twice slice of the CBS, which is sliced along one frequency axis is studied. By analysis, the horizontal slice of the once slice of the CBS (HSCBS), which is at a special given cyclic frequency is proposed to resolve the contradiction. Simulation and experiment of the rolling element bearing fault diagnosis are performed, and the results indicate the feasibility and validity of the HSCBS analysis in rolling element bearing early fault diagnosis. ► Twice slice of cyclic bispectrum (CBS) presents a method for displaying the CBS. ► Twice slice of CBS has less calculation than the CBS. ► Twice slice of CBS inherits the advantage of the CBS, and is immune to noise. ► Twice slice of CBS is more direct and clearly to show the features for diagnosis. ► Method of Slice Spectrum provides an attempt to research higher-order spectrum.