Abstract
This paper presents different graphical representations of wavelet transform (WT), their interpretations, advantages and disadvantages and limitations in vibration signal analysis. Empirical expressions between wavelet scale and frequency for Daubechies continuous wavelet transform (CWT) and dyadic discrete wavelet transform (DWT) are also presented. The obtained mathematical expressions are evaluated by applying them to the analysis of vibration signal of an unbalanced rotor. The results show that dyadic discrete wavelet transform (DWT) cannot accurately analyze signals whose frequency contents are not multiples of 2; the empirical equation between wavelet scale and frequency is valid for DWT only when the sampling frequency fs = 2.875 fmax, fmax is the maximum frequency in the signal, which is different from Nyquist frequency. CWT does not have these limitations but the accuracy of the estimated frequency from CWT scalogram is limited by the reading of the scale, a , from CWT scalogram.
| Original language | English |
|---|---|
| Pages (from-to) | 183-191 |
| Number of pages | 9 |
| Journal | American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP |
| Volume | 426 |
| State | Published - 2000 |
ASJC Scopus subject areas
- Mechanical Engineering