For the strong non-linear, non-Gauss and non-stationary vibration signal of rotating machinery, a time-frequency analysis method based on the wavelet transform technology and the traditional time-frequency analysis technology is proposed. In the method, the wavelet transform technology is used to achieve the signal decomposition of the variable scale time-frequency atom (wavelet functions), and FFT is used to complete time-frequency analysis. With the method, a simulation for the assumed non-stationary shock-vibration signal on bearing-rotor system is carried out. The simulated bearing vibration signal with the shock response component is analyzed, and the vibration characteristics in the time domain and frequency domain are obtained. The results show that the method can more quickly and accurately extract the non-stationary signal characteristics (frequency and magnitude), and its accuracy is higher than that of FFT and STFT.
[1]
Olivier Rioul,et al.
Time-scale energy distributions: a general class extending wavelet transforms
,
1992,
IEEE Trans. Signal Process..
[2]
William J. Williams,et al.
Improved time-frequency representation of multicomponent signals using exponential kernels
,
1989,
IEEE Trans. Acoust. Speech Signal Process..
[3]
Leon Cohen,et al.
Time Frequency Analysis: Theory and Applications
,
1994
.
[4]
Robert J. Marks,et al.
The use of cone-shaped kernels for generalized time-frequency representations of nonstationary signals
,
1990,
IEEE Trans. Acoust. Speech Signal Process..
[5]
Stéphane Mallat,et al.
Singularity detection and processing with wavelets
,
1992,
IEEE Trans. Inf. Theory.