Vibration Monitoring Using Wavelets Transform Feature Extraction Algorithm and Technique

Abstract:

Vibration is a mechanical phenomenon whereby oscillations occur about an equilibrium point. The oscillation may be periods such as the motion of a pendulum or random such as the movement of tire on a gravel road. Vibration causes waste of energy and creates unwanted sound-noise. Monitoring such process generally possess a big problem especially for a system. The present traditional single resolution techniques could not solve this problem, coupled with the Fourier transform which seems to be one of the best method in analyzing and monitoring vibration in machineries or machinery components.This paper present a new algorithm using wavelet- packet based feature in classification of vibration signals. This study explores the feasibility of the wavelet packet transform as a tool in search for features that may be used in the detection and classification of machinery vibration signals. By formulating a systematic method of determining wavelet packet based features that exploit class specific differences among interested signals, which avoid human interaction. This new algorithm provide more effective method to achieve robust classification than traditional single resolution techniques. The new algorithm in wavelet transform techniques proved to be more efficient, better analysis, and provides better results with minimum error than any existing method.

• Periodical: Applied Mechanics and Materials (Volume 666)
• Pages: 256-266
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.666.256
• Online since: October 2014
• Authors: Okuwobi Idowu Paul, Yong Hua Lu
• Keywords: Extraction, Feature, Transform, Vibration Monitoring, Wavelets

References

[1] G.G. Yen, "Health Monitoring of Vibration Signatures in Rotorcraft Wings, /I Neural Processing Letters, vol. 4, pp.127-137, (1996).

DOI: 10.1007/bf00426021

[2] RJ. Ferlez and D,C. Lang, Gear-Tooth Fault Detection and Tracking Using the Wavelet Transform, Prognosis of Residual Life Machinery and Structures, Proceedings of the 52nd Meeting of the Society for Machinery Failure Prevention Technology, pp.451-460, March, (1998).

[3] Juei-Cheng La, et aI, Fault Prediction in Transmissions Using Wavelet Analysis, Prognosis of Residual L~leMachinery and Structures, Proceedings of the 52nd Meeting ofthe Society for Machinery Failure Prevention Technology, pp.441-450, March, (1998).

[4] P.D. Samuel, et. ai., Fault Detection in the OH-58A Main Transmission Using the Wavelet Transform, Prognosis of Residual Life Machinery and Structures. Proceedings of the 52nd Meeting of the Society for Machinery Failure Prevention Technology, pp.323-335, March. (1998).

[5] B. Liu, etaI, Machinery Diagnosis Based on Wavelet Packets., Journal of Vibration and ControL vol. 3, pp.5-17, (1997).

[6] J.E. Lopez and K. Oliver, Overview of Wavelet Neural Network Fault Diagnostic Methods Applied to Rotating Machinery, Technology Showcase Integrated Monitoring, Diagnostics and Failure Prevention. Proceeding ofa Joint Conference, pp.405-417, April (1996).

[7] B.E. Parker, etaI, Helicopter Transmission Diagnostics using Vibration Signature Analysis, Technology Showcase Integrated Monitoring, Diagnostics and Failure Prevention, Proceeding of a Joint Conference, pp.419-430, April (1996).

[8] M.A. Essawy, S. Diwakar and S. Zein-Sabatto, Fault Diagnosis of Helicopter Gearboxes Using Neuro-Fuzzy Techniques, Prognosis ofResidual Life Machinery and Structures, Proceedings of the 52nd Meeting of the Society for Machinery Failure Prevention Technology, pp.293-303, March, (1998).

[9] D. Paul, Condition monitoring and defect diagnosis in manufacturing process using DDS and wavelets, Ph. d dissertation, Michigan Technological University, (1995).

[10] O. Rioul and M. Vetterli, Wavelets and Signal Processing, IEEE Signal Processing Magnine, pp.14-38, Oct. (1991).

DOI: 10.1109/79.91217

[11] F. Hlawatsch and G.F. Boudreaux-Bartels, Linear and Quadratic Time-Frequency Signal Representations, IEEE Signal Processing Magnine, pp.21-67. Apr. (1992).

DOI: 10.1109/79.127284

[12] R.R. Coifman and M.V. Wickerhauser. Entropy-Based Algorithms for Best Basis Selection, IEEE Transactions on Information Theory, vol. 38, no. 2, pp.713-718. March (1992).

DOI: 10.1109/18.119732