Condition Feature Extraction of Machine Tools Based on Wavelet Packet Energy Spectrum Analysis and Bispectrum Analysis of Current Signal

Abstract:

Based on the study of the characteristics of load current signal, this article develops a method to extract features that can be use to distinguish the different working status of machine tools in real-time manner. The features are extracted from wavelet packet energy spectrum and bispectrum of the load current signal, and thus can take advantages of both wavelet packet transforms and bispectrum in signal analysis. Experimental results show that, compared with the features extracted from wavelet packet energy spectrum or bispectrum alone, the features extracted by applying the proposed method can provide better performance in term of identifying the machine working status.

• Periodical: Applied Mechanics and Materials (Volumes 101-102)
• Pages: 847-850
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.101-102.847
• Online since: September 2011
• Authors: Teng Fei Fang, Guo Fu Li
• Keywords: Bispectrum, Energy Spectrum, Feature Extraction, Wavelet Packet Transform

References

[1] G.F. Li, T.F. Fang and L. Luo: Applied Mechanics and Materials, Vol. 37-38 (2010), pp.1540-1543.

[2] A. Verl, U. Heisel and M. Walther: CIRP Annals - Manufacturing Technology, Vol. 58 (2009 ), pp.375-378.

[3] D. Kim and D. Jeon: Precision Engineering, Vol. 35 (2011), pp.143-152.

[4] D.G. Ece and M. Basaran: Expert Systems with Applications, Vol. 38 (2011), pp.8079-8086.

[5] Y.H. Feng and F.S. Schlindwein: Mechanical Systems and Signal Processing, Vol. 23 (2009), pp.712-723.

[6] J. Treetrong, J.K. Sinha and F.S. Gu: ISA Transactions, Vol. 48 (2009), pp.37-382.

[7] F. Gu, Y. Shao and N. Hu: Mechanical Systems and Signal Processing, Vol. 25 (2011), pp.360-372.