Study on Mechanical Fault Diagnosis Based on IMF Complexity Feature and Support Vector Machine

Abstract:

According to the non-stationarity characteristics of the vibration signals from reciprocating machinery,a fault diagnosis method based on empirical mode decomposition,Lempel-Ziv complexity and support vector machine(SVM) is proposed.Firstly,the vibration signals were decomposed into a finite number of intrinsic mode functions(IMF), then choosed some IMF components with the criteria of mutual correlation coefficient between IMF components and denoised signal.Thirdly the complexity feature of each IMF component was calculated as faulty eigenvector and served as input of SVM classifier so that the faults of machine are classified.Practical experimental data is used to verify this method,and the diagnosis results and comparative tests fully validate its effectiveness and generalization abilities.

• Periodical: Applied Mechanics and Materials (Volumes 246-247)
• Pages: 37-42
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.246-247.37
• Online since: December 2012
• Authors: Wei Dong Liu, Hu Sheng Wu
• Keywords: Empirical Mode Decomposition (EMD), Fault Diagnosis, Lempel-Ziv Complexity, Support Vector Machine (SVM)

References

[1] Jinming Lu, Chuntao Wang, Haigang Zhou. Misfire fault diagnosis of diesel engiie based on EMD. Journal of Jiangsu University of Science and Technology, vol. 23, no. 1, pp.234-238, Jun, (2009).

[2] Norden E. Huang, etc. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proc.R. Soc. Lond. A, 1998, pp.903-995.

[3] J. Jeong, S J. Kim, Non-linear dynamical analysis of the EEG in Alzheimers disease with optimal embedding dimension. Electroenceph Clin Neurophysiol, 1998, pp.220-228.

DOI: 10.1016/s0013-4694(97)00079-5

[4] Anishchenko V. Stochastic R: noise-enhanced order. Physics Uspekhi, vol. 42, no. 1, pp.7-36, (1999).

[5] Sun Bin, Zhou Yunlong, Zhong Jinshan. Identification method of gas- liquid two-phase flow regime based on complexity feature. Chemical Engineering (China), vol. 36, no. 4, pp.27-30, Apr, (2008).

[6] Lempel A, Ziv J. On the complexity of finite sequenced. IEEE Transactions on Information Theory, vol. 22, no. 1, pp.75-81, Jan, (1976).

DOI: 10.1109/tit.1976.1055501

[7] D. Abasolo, Analysis of EEG background activity in Alzheimer's disease patients with Lempel-Ziv complexity and central tendency measure, Medical Engineering&Phsics vol. 28, 2006, pp.315-322.

DOI: 10.1016/j.medengphy.2005.07.004

[8] X.S. Zhang R.J. Roy, and E.W. Jensen, EEG complexity as a measure of depth of anesthesia for patients, IEEE Trans. Biomed. Eng, vol. 48, no. 12, pp.1424-1433, Dec. (2001).

DOI: 10.1109/10.966601

[9] R. Nagarajan, Quantifying physiological data with Lempel-Ziv complexity-certain issues, IEEE Trans. Biomed. Eng, vol. 49, no. 11, pp.1371-1373, Nov. (2002).

DOI: 10.1109/tbme.2002.804582

[10] Vapnik V N. The Nature of Statistical Learning Theory. New York: Spring Verag, (1995).

[11] Kressel U. Pairwise classification and support vector machines. In Scholkopf B. et al(Eds), Advances in kernel Methods Support vector learning. Cambridge, MA, MIT Press, 1999, pp.255-268.

DOI: 10.7551/mitpress/1130.003.0020

[12] Platt J, Cristianini N, Shawe-Taylor J. Large margin DAG's for multiclassification. In: Advances in Neural Information Processing Systems Cambridge. MA, MIT Press, Vol. 1, No. 12, pp.547-553, Jan, (2000).

[13] Xuedong Zhu, Ping Hu. The design of Multi-fault classification based on Huffman binary tree. Journal of Beijing Union University(Natural Sciences), Vol. 23, No. 2, pp.26-29, Jun, (2009).

[14] Xu Renlin, An Wei. Wavelet Denoise Application in the Signal Hilbert Transform Based on EMD. Noise and Vibration Control, vol. 74, no. 4, pp.74-77, Jun, (2008).