Realizing Method of MATLAB Software for Fault Diagnosis of Freight Bearings

Abstract:

In order to realize the online fault diagnosis of freight rolling bearing without disassembling, a simulation test platform was established in the laboratory and acoustic emission (AE) sensor of AE-98/R15 was used to acquire AE signals. According to the signal characteristics, MATLAB software was used to analyze features of signals with wavelet packet and to recognize bearing state with probabilistic neural network. These methods have a very good effect for fault diagnosis in the laboratory. The innovation of this paper is that the above methods are effectively used for fault diagnosis and provide a feasible scheme for field application.

• Periodical: Applied Mechanics and Materials (Volume 273)
• Pages: 245-249
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.273.245
• Online since: January 2013
• Authors: Chun Chao Chen
• Keywords: Fault Diagnosis, MATLAB, Neural Network (NN), Wavelet Packet Analysis

References

[1] Zhao Yuanxi, Xu Yonggang: Fault Pattern Recognition Technology of Acoustic Emission of Rolling Bearing based on Harmonic Wavelet and BP Neural Network . Journal of vibration and shock. Vols. 29(2010) , pp.162-165.

[2] Liao Chuanjun, Li Xuejun: applications of reassigned wavelet scalogram in feature extraction of acoustic emission signal. Journal of mechanical engineering. Vols. 45(2009), pp.273-279.

DOI: 10.3901/jme.2009.02.273

[3] Zhang Wenbin, Zhou Xiaojun: Machinery Fault Signal Extraction Based on Harmonic Wavelet Packet Method of Rotating. Journal of vibration and shock, Vols. 28(2009) , pp.87-89.

[4] Tang Wei, Wang Haili, Zhuang Zijie: State Recognition of Tool Wear Based on Wavelet Multi resolution Analysis and RBF Neural Network Tool technology. Vols. 43(2009), pp.15-19.

[5] Flying science and technology product research and Development Center. Neural network theory and MATLAB7 realizition. Beijing: Publishing House of electronics industry(2005).