
An impact feature extracting method based on S transform spectrum threshold denoising
GUO Yuan-jing WEI Yan-ding ZHOU Xiao-jun TANG Fang
Journal of Vibration and Shock ›› 2014, Vol. 33 ›› Issue (21) : 44-50.
An impact feature extracting method based on S transform spectrum threshold denoising
Extracting the impact feature from rotating machinery fault vibration signal, is significant to some faults dignosis. Based on the fact that S tansform is suitable for processing the signal with impact feature, an effective impact feature extracting method with the use of threshold denoising for S transform spectrum is proposed. Firstly, the time-frequency spectrum of a time-domain signal is obtained by S transform. As the spectrum coefficients are complex numbers, the spectrum threshold denoising method is carried out according to the modulus magnitude of coefficients. During denoising both the hard and the soft threshold functions are used, respectively. Most importantly, the optimal threshold estimation is available between zero to the maximum modulus of the coefficients by step iterative algorithm, evaluated with the introduced modified risk function. Finally, inverse S transform is used on the denoised time-frequency spectrum to reconstruct the time-domain impact feature. The processing results of simulated signal and rolling bearing fault vibration signal have shown that using the acquired optimal threshold, S transform spectrum threshold denoising method can successfully extract the impact feature from noise mixed signal, so as to realize related fault dignosis.
fault dignosis / impact feature / S transform / step iterative algorithm / modified risk function / optimal threshold estimation / threshold denoising {{custom_keyword}} /
/
〈 |
|
〉 |