Mao-fa GONG(公茂法), Qing-xue LIU(刘庆雪), Ming LIU(刘明), Xiao- li ZHANG(张晓丽)
College of Information and Electrical Engineering, Shandong Unive rsity of Science and Technology, Qingdao 266510, China
Abstract-This paper expounded in detail the principle of ener gy spectrum analysis based on discrete wavelet transformation and multi-resolut ion analysis. In the aspect of feature extraction method study, with investigati ng the feature of impact factor in vibration signals and considering the non-pl acidity and non-linear of vibration diagnosis signals, the authors import wavel et analysis and fractal theory as the tools of faulty signal feature description . Experimental results proved the validity of this method. To some extent, this method provides a good approach of resolving the wholesome problem of fault feat ure symptom description.
Key words-discrete wavelet transform(DWT); multi-resol ution analysis; fault diagnosis; rotating machinery; feature extraction
Manuscript Number: 1674-8042(2010)04-0312-03
dio: 10.3969/j.issn.1674-8042.2010.04.02
References
[1]He-ji Yu, Qing-da Han, Shen Li, etl al, 2001. Equipment Fault Diag nosis Engineering. Metallurgy Industry Press, Beijing, p.3-10.
[2]Menderes Kalkat, 2005. Design of artificial neural networks for roto r dynamics analysis of rotating machine systems. Mechatronics, 15: 573-588.
[3]V. Kreinovich, O. Sirisangtaksin, 1994. Wavelet Neural Networks are Asymptotically Optimal Approximators for Functions of One Variable. Florida: Pro c of IEEE ICNN, p. 299-304.
[4]T.Poggio, F.Girosi, 1990. Networks for Approximation and Leaning. Proc. IEEE, p.1481-1497.
[full text view]
