The aim of this study is to assess the effectiveness of the Wavelet Transform (WT) for machine condition monitoring purposes. In this chapter the WT is set up specifically for vibration signals captured from real life complex case studies having a poor extent in literature: marine couplings and i.c. engines tested in cold conditions. Both Continuous (CWT) and Discrete Wavelet Transform (DWT) are applied. The former has been used for faulty event identification and impulse event characterization through the analysis of the three-dimensional representation of the CWT coefficients. The latter has been applied for filtering and feature extraction purposes and for detecting impulsive events strongly masked by noise. Comparing the results from both the CWT and DWT analyses it has been clearly demonstrated the ability of the WT in satisfying both the condition monitoring and fault detection requirements for all tested cases. This means that the application of the WT not only permit to recognize the change of the state of the tested machine but it is also able to localise the source of the alteration.
On the Use of Wavelet Transform for Practical Condition Monitoring Issues
DELVECCHIO, Simone
2012
Abstract
The aim of this study is to assess the effectiveness of the Wavelet Transform (WT) for machine condition monitoring purposes. In this chapter the WT is set up specifically for vibration signals captured from real life complex case studies having a poor extent in literature: marine couplings and i.c. engines tested in cold conditions. Both Continuous (CWT) and Discrete Wavelet Transform (DWT) are applied. The former has been used for faulty event identification and impulse event characterization through the analysis of the three-dimensional representation of the CWT coefficients. The latter has been applied for filtering and feature extraction purposes and for detecting impulsive events strongly masked by noise. Comparing the results from both the CWT and DWT analyses it has been clearly demonstrated the ability of the WT in satisfying both the condition monitoring and fault detection requirements for all tested cases. This means that the application of the WT not only permit to recognize the change of the state of the tested machine but it is also able to localise the source of the alteration.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
