Loading [MathJax]/extensions/TeX/mhchem.js
Detection of gear defects by mixed Discrete Wavelet Transform and FastEMD | IEEE Conference Publication | IEEE Xplore

Detection of gear defects by mixed Discrete Wavelet Transform and FastEMD


Abstract:

Detection of default gear in an early stage is still a major problem, especially when gears are subjected to non-stationary phenomena. The purpose of this paper is to pre...Show More

Abstract:

Detection of default gear in an early stage is still a major problem, especially when gears are subjected to non-stationary phenomena. The purpose of this paper is to presents the importance of empirical mode decomposition (EMD) and the wavelet transform in the improvement of the sensitivity of scalar indicators kurtosis in the context of defect detection by vibration analysis of gear transmissions. If the gear is damaged, the signal is disturbed by the appearance of a crack in the gear tooth. This is due to the presence of periodic pulses. Nonetheless, the existence of background noise induced by random excitation may affect the values of these temporal indicators. However de noising these signals through wavelet conversion improves the sensitivity of these indicators and increases the reliability of the survey. The proposed approach consists in applying a multi-resolution wavelet analysis with variable decomposition levels related to the severity of gear faults then fast empirical mode decomposition is used to early detect faults. The technique developed is tested on vibration signals recorded from the Technical Center for Mechanical Industries (CETIM) test bench for twelve days. The results obtained showed us the triggering of a defect appearing on the gear tooth from the 5th day and the evolution of the defect on the 8th day. These results are in accordance with the expert report carried out on this gear system.
Date of Conference: 17-20 May 2022
Date Added to IEEE Xplore: 30 June 2022
ISBN Information:

ISSN Information:

Conference Location: Istanbul, Turkey

References

References is not available for this document.