Fault Diagnosis of Gearbox Based on ICA and Envelope Analysis

Abstract:

The traditional envelope analysis methods are usually ineffective in the fault diagnosis of Gearbox on the condition that there are multi-fault sources in mechanic systems. In the paper, an independent component analysis (ICA) approach is proposed incorporating with the envelope analysis to handle the difficulty. The gearbox envelope signal can be separated into different components according to different vibration sources, and then be further studied by the envelope spectrum analysis in terms of the vibrations in the gearbox. Extensive practical tests on the gearbox including two types of faults are performed to verify the validity of the proposed method. The test results show that the fault frequency characteristics can be separated successfully and the faults are identified precisely.

• Periodical: Advanced Materials Research (Volumes 430-432)
• Pages: 2054-2057
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.430-432.2054
• Online since: January 2012
• Authors: Bin Li, Yu Guo, Ting Wei Liu, Yan Gao
• Keywords: Envelope Analysis, Fault Diagnosis, Gearbox, Independent Component Analysis (ICA)

References

[1] Chen Jin. Machinery vibration monitoring and fault diagnosis [M]. Shanghai: Shanghai Jiao tong University Press 1997: 98-102(In Chinese).

[2] A. Hyvrainen, J. Karhunen and E. Oja: Independent Component Analysis [J], John WILEY & Sons, Inc, Finland (2001).

[3] A. Hyvrainen: Fast and robust fixed-point algorithms for independent component analysis, IEEETrans. Neural Netw., Vol. 10(1999), p.626.

DOI: 10.1109/72.761722

[4] Li Qiang, Wang TaiYong, Huang Yi, Wang ZhengYing. Fault signal' blind source separation in frequency donmain and its application [J]. Journal of Mechanical Strength, 2009, 31(1): 014-018 (In Chinese).

[5] McFadden P D, Smith J D. Vibration monitoring of rolling element bearings by the high-frequency resonance technique—a review [J]. Tribology International, 1984, 1(17): 3-10.

DOI: 10.1016/0301-679x(84)90076-8

[6] Fan Yongsheng, Mechanical equipment diagnosis method of modern signal processing [M]. National Defense Industry Press, 2009, 3(3): 41~44(In Chinese).