Application of Fault Diagnosis Method Based on cICA to Gearbox

Abstract:

Constrained independent component analysis (cICA) is a new theory and new method derived from the independent component analysis (ICA).It can extract the desired independent components (ICs) from the data based on some prior information, thus overcoming the uncertainty of the traditional ICA. Early gearbox fault signals is often very weak ,characterized by non-Gaussian,low signal-to-noise ratio (SNR), which make the existing diagnosis methods in the diagnosis of early application restricted. In this paper,cICA algorithm is applied to gear fault diagnosis. Through the case studies verify the feasibility of this method to extract the desired independent components (ICs), indicating the applicability and effectiveness of the method.

• Periodical: Applied Mechanics and Materials (Volume 664)
• Pages: 148-152
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.664.148
• Online since: October 2014
• Authors: Shuang Xi Jing*, Song Tao Guo, Jun Fa Leng, Xing Yu Zhao
• Keywords: Constrained Independent Component Analysis, Fault Diagnosis, Gear, Independent Component Analysis, Reference Signal

References

[1] Zhong Binglin, Huang Ren：Introduction to machine fault diagnosis(Mechanical Industry Publications, Beijing 2002).

[2] Aapo Hyvarinen, Juha Karhunen, Erkki Oja: Zhou Zongtan , Dong Guohua, et, al Translated: Independent component analysis( Electronic Industry Publications, Beijing 2007).

[3] Q. He, R, Du, F, Kong: Phase space feature based on independent component analysis for machine health diagnosis. Journal of Vibration and Acoustic, vol. 133(2011), 011106.

[4] X. Liu, et al: Blind source separation of internal combustion engine piston slap from other measured vibration signals. Mechanical Systems and Signal Processing , no. 19(2005), pp.1196-1208.

DOI: 10.1016/j.ymssp.2005.08.004

[5] R. Peled,S. Braun, et. al: A blind deconvolution separation of multiple sources with application to bearing diagnostics. Mechanical Systems and Signal Processing , no. 19(2005), pp.1181-1195.

DOI: 10.1016/j.ymssp.2005.08.019

[6] Wei Lu, Jagath C. Rajapakse : ICA with Reference. Neurocomputing , vol. 69(2006), pp.2244-2257.

DOI: 10.1016/j.neucom.2005.06.021

[7] W. Lu, J.C.R.: Approach and applications of constrained ICA. IEEE TRANSACTIONS ON NEURAL NETWORKS , vol. 16(2005), pp.203-212.

DOI: 10.1109/tnn.2004.836795

[8] Changli Lia, Guisheng Liaoa, Yuli Shenc: An improved method for independent component analysis with reference. Digital Signal Processing , vol. 20(2010), pp.575-580.

DOI: 10.1016/j.dsp.2009.08.009

[9] Akhtar, M. T., et al: Employing spatially constrained ICA and wavelet denoising, for automatic removal of artifacts from multichannel EEG data. Signal Processing, vol. 92 (2012), pp.401-416.

DOI: 10.1016/j.sigpro.2011.08.005

[10] Zhiyang Wang, Jin Chen, Guangming Dong, Yu Zhou: Constrained independent component analysis and its application to machine fault diagnosis. Mechanical Systems and Signal Processing, vol. 25 (2011), pp.2501-2512.

DOI: 10.1016/j.ymssp.2011.03.006

[11] Wu Chuanhui , Liang Yu, Guo Yu：Extracting fault features of rolling bearing during speed variation based on Cica. Mechanical Science and Technology, vol. 8(2013).

[12] Liang Yu, Guo Yu, Wu Chuanhui：Rolling element bearing fault feature extraction based on spectral kurtosis and cICA. Journal of Mechanical Strength , vol. 5(2013).

[13] Liu Ju, Sun Jiande, Xu Hongji: Blind signal processing theory and application (Science Publications, Beijing 2012).

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

DOI: 10.1109/72.761722

[15] Nguyen Duc Thanga, Tahir Rasheeda, Young-Koo Leea, Sungyoung Leea, Tae-Seong Kimb: Content-based facial image retrieval using constrained independent component analysis. Information Sciences, vol. 181 (2011), pp.3162-3174.

DOI: 10.1016/j.ins.2011.03.021