Application of the Blind Source Separation Based on Time-Frequency Analysis in Mechanical Fault Diagnosis

Abstract:

Blind source separation provides a new method for the separation of mechanical sources under high level background noise, as well as the diagnosis of the compound fault. At present, the blind source separation has been successfully applied to the mecanical fault diagnosis. But the traditional mechanical source separation methods are restricted to non-gauss, stationary and mutually independent source signals. However, the mechanical fault signals do not suffice to these conditions, and generally exhibit non-stationarity and non-independence. For the non-stationary signal, its spectral feature is time-varying. Thus only the time-domain or frequency-domain analysis is not sufficient to describe the characteristics of non-stationary signal. The time-frequency analysis, which can provide the information about that the spectrum of the signal varies with the time, is a useful tool for non-stationary signal analysis. In this paper, combined time-frequency analysis with blind source separation, a blind source separation method for the non-stationary signal of the mechanical equipment based on time-frequency analysis is proposed and studied. The simulation and experimental results show that the proposed approach is feasible and effective.

• Periodical: Advanced Materials Research (Volumes 945-949)
• Pages: 1054-1062
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.945-949.1054
• Online since: June 2014
• Authors: Zhi Nong Li*, Fen Zhang, Xu Ping He, Yao Xian Xiao
• Keywords: Blind Source Separation (BSS), Fault Diagnosis, Non-Stationary Signal, Smoothed Pseudo Wigner-Ville Distribution (SPWVD), Time-Frequency Analysis

References

[1] Li Z.N., He Y.Y., He Y. Y, Chu F.L., Han J., Hao W. Fault Recognition Method for Speed-up and Speed-down Process of Rotating Machinery Based on Independent Component Analysis and Factorial Hidden Markov Model, Journal of Sound and Vibration, 291(1-2), PP. 60-71, (2006).

DOI: 10.1016/j.jsv.2005.05.020

[2] Li Z.N., Han J., Sun J.J., He Y. Y, Chu F.L. Fault Recognition Method Based on Independent Component Analysis and Hidden Markov Model. Journal of Vibration and Control, 13(2), pp.125-137, (2007).

DOI: 10.1177/1077546307066964

[3] Shen Y.J., Yang S.P. A New Blind Source Separation Method and its Application to Fault Diagnosis of Rolling Bearing. International Journal of Nonlinear Sciences and Numerical Simulation, 7(3), pp.245-250, (2011).

DOI: 10.1515/ijnsns.2006.7.3.245

[4] Yang X.Y. ，Zhou X. A New Method for Adaptive Blind Source Separation Based on the Estimated Number of Dynamic Fault Sources. Mechanical and Electrical Control Engineering and Automation, Vol. 233, pp.211-217, (2012).

DOI: 10.4028/www.scientific.net/amm.233.211

[5] Nguyen V. H., Rutten C., Fault Diagnosis in Industrial Systems Based on Blind Source Separation Techniques Using One Single Vibration Sensor. Shock and Vibration, 19(5), pp.795-801, (2010).

DOI: 10.1155/2012/183541

[6] He Z.J., Zi Y.Y. Meng Q.F., Zhao J.Y., et al. Fault Diagnosis Principles of Non-stationary Signal and Applications to Mechanical Equipment. Beijing: Higher Education Press, 2001(In Chinese).

[7] Li Z. N, Several key problems in the machine fault diagnosis based on blind source separation. Postdoctoral Research Report of Tsinghua University, 2005(In Chinese).

[8] Chu F. L, Peng Z. K, Feng Z. P, Li Z. N, Morden Signal Processing in Mechanical Fault Diagnosis. Beijing: Science Press, (2009).

[9] Cardoso,J. -F. and Souloumiac, A., Blind beamforming for non-Gaussian signals, Radar and Signal Processing, IEE Proceedings F 140(6), pp.362-370, (1993).

DOI: 10.1049/ip-f-2.1993.0054

[10] Ypma A., Learning methods for machine vibration analysis and health monitoring, PhD-dissertation, ISBN 90-9015310-1, Pattern Recognition Group, Dept. of Applied Physics, Delft University of Technology, November, (2001).

[11] Cardoso, J. -F., Laheld, B., Equivariant adaptive source separation, IEEE Trans. on Signal Processing 44(12), pp.3017-3030, (1996).

DOI: 10.1109/78.553476

[12] Bell, A. J. and Sejnowski,T., J., An information maximization approach to blind separation and blind deconvolution, Neural Computation 7, p.1129–1159, (1995).

DOI: 10.1162/neco.1995.7.6.1129

[13] Hyvärinen A., Fast and robust fixed-Point algorithms for independent component analysis, IEEE Transactions on Neural Networks 10(3), pp.626-634, (1999).

DOI: 10.1109/72.761722

[14] Jiao Weidong, Research on methods for fault diagnosis of rotating machinery based on independent component analysis, PhD thesis, Zhejiang University, (2003).