A Feature Extraction Method of Gear Fault Based on the SVD EMD and Morphology

Abstract:

In order to extract the weak signal from strong background signal characteristics, a feature extraction method combined of the singular value decomposition (SVD), empirical mode decomposition (EMD) and mathematical morphology was proposed. The signal got through the singular value decomposition first. Next took the average value of the decomposed main components. And carried on the empirical mode decomposition and selected the main component to summate and refactor. Then morphological difference filter was used to extract the frequency characteristics of the fault signal. The results of numerical simulation test and gear fault simulation experiments show that the proposed method can clearly extract the frequency characteristics of weak signal from strong background signal and noise. Comparison has been done with the results of singular value decomposition (SVD) and morphological filtering method and empirical mode decomposition form of filtering method. It proves the effectiveness of the proposed method.

• Periodical: Applied Mechanics and Materials (Volumes 433-435)
• Pages: 477-482
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.433-435.477
• Online since: October 2013
• Authors: Gao Yan Hou*, Yong Lv, Hao Huang, Yi Zhu
• Keywords: Empirical Mode Decomposition (EMD), Feature Extraction, Mathematical Morphology, Singular Value Decomposition

References

[1] Kakarala R. Ogunbona P O. Signal Analysis Using a Multiresolution From of the Singular Value Decomposition[J]. IEEE Transactions on Image Processing, 2001, 10（5）: 724—735.

DOI: 10.1109/83.918566

[2] Maj J B, Royackers L, Moonen M, et a1. SVD-based optimal filtering for noise reduction in dual microphone hearing aids: a real time implementation and perceptual evaluation[J]. IEEE Transactions on Biomedical Engineering, 2005, 52(9): 1563—1573.

DOI: 10.1109/tbme.2005.851517

[3] Cempel C. Generalized singular value decomposition in multidimensional condition monitoring of machines—a proposal of comparative diagnostics[J]. Mechanical Systems and Signal Processing, 2009, 23(3): 701—711.

DOI: 10.1016/j.ymssp.2008.07.004

[4] Zhiming LV, Wujun ZHANG, Jinwu XU, et a1. The noise reduction method based on singular spectrum and its application in fault diagnosis technology[J]. Journal of mechanical engineering, 2000, 35(3): 95—99. (In Chinese).

[5] Xiangyang DUAN, Yongsheng WANG, Yongsheng SU, et a1. Signal feature extraction method based on singular value decomposition[J]. Journal of Vibration and Shock, 2009, 28(11): 30—33. (In Chinese).

[6] Yong LV, Yourong LI, Zhigang WANG, Ruisun ZHU, et a1. Based on singular value decomposition and envelope analysis of local gear fault feature extraction[J]. Machine Tool and Hydraulics, 2007, 35(5): 217—219. (In Chinese).

[7] Huang N E, Shen Z, Long S R, et a1. The Empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis[J]. Proc. R. Soc. Lond. A, 1998, 454: 903—994.

[8] Yanju ZHAO , Taiyong WANG , Chengzu REN, et a1. Under the background of strong noise empirical mode decomposition[J]. Journal of Vibration and Shock, 2009, 28(3): 149—152. (In Chinese).

[9] Naveed ur Rehman, Danilo P. Mandic, et a1. Empirical Mode Decomposition for Trivariate Signals[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 58, NO. 3, 1059–1068, MARCH. (2010).

DOI: 10.1109/tsp.2009.2033730

[10] J. Serra, Morphological filtering: an overview, Signal Processing , 38 (4) (1994) 3–11.

[11] N. G. Nikolaou, I. A. Antoniadis, Application of morphological operators as envelope extractors for impulsive-type periodic signals, Mechanical Systems and Signal Processing 17 (6) (2003) 1147–1162.

DOI: 10.1006/mssp.2002.1576

[12] A. Hu, G. Tang, L. An, De-noising technique for vibration signals of rotating machinery based on mathematical morphology filter, Chinese Journal of Mechanical Engineering 42 (4) (2006) 127–130.

DOI: 10.3901/jme.2006.04.127