Науковий вісник НГУ

Compound fault diagnosis of gearbox based on blind source separation and EEMD-SVD

• 
• 

User Rating:  / 0

PoorBest 

Details

Category: Geotechnical and mining mechanical engineering, machine building

Last Updated on 08 February 2016

Published on 08 February 2016

Hits: 4267

Authors:

Wang Lijun, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China

Ma Xiao, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China

Duan Zhichao, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China

Yao Xinying, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China

Wu Liping, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China

Abstract:

Purpose. Blind source separation (BSS) is a new theory and method of vibration signal analysis and processing. With the development of science, blind source separation is more and more applied in the field of mechanical equipment fault diagnosis. In the research, it was applied in gearbox fault diagnosis, which is advantageous to separate mixed signal and make fault diagnosis.

Methodology. Based on the gearbox experiment platform of laboratory, first, we used the Intrinsic Mode Function‒Singular Value Decomposition (IMF-SVD) source number estimation method to estimate the vibration source number, then we used the Ensemble Empirical Mode Decomposition-Singular Value Decomposition (EEMD-SVD) method to reduce noise and the Joint Approximate Diagonalization of Eigenmatrices (JADE) algorithm to separate the signals and extract the fault features. Lastly, we diagnosed the gearbox fault by further analysis.

Findings. Through many simulation experiments, we found that, in the case of high Signal to Noise Ratio (SNR)(0dB), the employment of the EEMD-SVD method followed by the JADE algorithm allows separating the vibration signal well and the separated signal only contains a small amount of noise. The gearbox fault diagnosis showed that the method has good results when diagnosing the mixed fault of the gearbox.

Originality. We made a study of combining the EEMD-SVD noise reduction method and the JADE algorithm to separate gearbox vibration signals. The further studies on this aspect may be helpful.

Practical value. The method has good accuracy and reliability when applied for the gearbox fault diagnosis. We will further study the application of blind source separation technology in gearbox fault diagnosis.

References:

1. Gelle, G., Colas, M. and Serviere, C. (2001), “Blind source separation: a tool for rotating machine monitoring by vibrations analysis”, Journal of Sound & Vibration, vol. 248, no.5, pp. 865−885.

2. Keziou, A., Fenniri, H., Ghazdali, A. (2014), “New blind source separation method of independent/dependent sources”, Signal Processing, vol. 104, no.6, pp. 319−324.

3. Gelle, G., Colas, M., Serviere, C. (2003), “Blind Source Separation: A New Pre-Processing Tool for Rotating Machines Monitoring?”, Instrumentation & Measurement IEEE Transactions, vol.52, no.3, pp. 790−795.

4. Z Li, X Yan, Z Tian (2013), “Blind vibration component separation and nonlinear feature extraction applied to the nonstationary vibration signals for the gearbox multi-fault diagnosis”, Measurement, vol.46, no.1, pp. 259−271.

5. Roan, M.J., Erling, J.G. and Sibul, L.H. (2002), “A new, non-linear, adaptive, blind source separation approach to gear tooth failure detection and analysis”, Mechanical Systems and Signal Processing, vol.16, no.5, pp.719−740.

6. Jiang, Y. and Zhi-Xiong, L.I. (2011), “Research on the Blind Source Separation of Gearbox vibration and Its Application in Gear Fault Diagnosis”, Journal of Hubei University of Technology, vol. 26, no.4, pp. 25−29.

2015_06_lijun
2016-02-08  2.16 MB  1026

• < Prev