论文信息 - Feature Extraction of Gearbox Compound Faults Based on Blind Source Separation

Feature Extraction of Gearbox Compound Faults Based on Blind Source Separation

Aimed at the diagnosis to compound faults of eccentricity and teeth broken in gear box, the paper presents a method of compound fault diagnosis based on the blind source separation to separate the coupling characteristics in compound faults. The method separates two collected signals including some compound faults into two independent source signals, and then extracts the signal characteristics, i.e. the fault characteristics of gear eccentricity and teeth broken, through resonance demodulation. The result of the experiment shows the validity and practicality of the method.

Wei Zhang | Xiao Wei Wang | Hui Li | Lin Suo Shi

[1] E. Oja,et al. Independent Component Analysis , 2001 .

[2] Dominique Martinez,et al. Nonlinear blind source separation using kernels , 2003, IEEE Trans. Neural Networks.

[3] S. Haykin. Unsupervised adaptive filtering, vol. 1: Blind source separation , 2000 .

[4] Mnaouar Chouchane,et al. Detection of rolling element bearing defects by adaptive filtering , 2005 .

[5] Mathias Johansson. Blind Source Separation , 2014, Encyclopedia of Social Network Analysis and Mining.

[6] A. Cichocki. Blind Signal Processing Methods for Analyzing Multichannel Brain Signals , 2004 .