Fatigue Data Analysis Using Continuous Wavelet Transform and Discrete Wavelet Transform

Abstract:

The wavelet transform is well known for its ability in vibration analysis in fault detection. This paper presents the ability of wavelet transform in fatigue data analysis starts from high amplitude events detection and it is then followed by fatigue data extraction based on wavelet coefficients. Since the wavelet transform has two main categories, i.e. the continuous wavelet transforms (CWT) and the discrete wavelet transform (DWT), the comparison study were carried out in order to investigate performance of both wavelet for fatigue data analysis. CWT represents by the Morlet wavelet while DWT with the form of the 4th Order Daubechies wavelet (Db4) was also used for the analysis. An analysis begins with coefficients plot using the time-scale representation that associated to energy coefficients plot for the input value in fatigue data extraction. Ten extraction levels were used and all levels gave the damage difference, (%∆D) less than 10% with respect to original signal. From the study, both wavelet transforms gave almost similar ability in editing fatigue data but the Morlet wavelet provided faster analysis time compared to the Db4 wavelet. In comparison to have the value of different at 5%, the Morlet wavelet achieved at L= 5 while the Db4 wavelet at L=7. Even though it gave slower analysis time, both wavelets can be used in fatigue data editing but at different time consuming.

• Periodical: Key Engineering Materials (Volumes 462-463)
• Pages: 461-466
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.462-463.461
• Online since: January 2011
• Authors: Shahrum Abdullah, S.N. Sahadan, Mohd Zaki Nuawi, Zulkifli Mohd Nopiah
• Keywords: Continuous Wavelet, Daubechies, Discrete Wavelet, Fatigue Data Extraction, Morlet

References

[1] C-S. Oh : International Journal of Fatigue Vol. 23 (2001) p.241.

[2] S. Abdullah, J. C. Choi, J. A. Giacomin and J. R. Yates: International Journal of Fatigue Vol. 28 (2006) p.675.

[3] T.E. Putra, S. Abdullah, M.Z. Nuawi and Z.M. Nopiah: Journal of Applied Sciences Vol. 10 (2010) p.277.

[4] J. -D. Wu and C. -H. Liu. Expert Systems with Applications (2007).

[5] S. Abbasion, A. Rafsanjani A. Farshidianfar, and N. Irani : Mechanical Systems and Signal Processing Vol 21 (2007) p.2933.

DOI: 10.1016/j.ymssp.2007.02.003

[6] H. Chen, M.J. Zuo , X. Wang and M.R. Hoseini: Measurement 43 (2010) p.570.

[7] N. Saravanan, V.N.S. Kumar Siddabattuni, K.I. Ramachandran: Expert Systems with Applications 35 (2008) p.1351.

[8] W. J Staszewski: Journal of Sound and Vibration Vol. 211 (1998) p.735.

[9] P. S Addison: The Illustrated Wavelet Transform Handbook (Institute of Physics Publishing, (2002).

[10] N. E. Dowling: Journal of Materials, JMLSA, Vol. 7 (1972) p.71.

[11] K. N. Smith, P. Watson, T.H. Topper: Journal of Materials Vol. 5 (1970) p.767.

[12] M. Khalil and T. H Topper : International Journal of Fatigue Vol. 25 (2003) p.149.

[13] Wavelet Toolbox. Matlab 2006a.

[14] Information on http: /www. wavelet. org/tutorial/wbasic. htm.