Vibration analysis for slab track at different train speeds using Bayes wavelet denoising

With increasing train speed, it is necessary to study the relation between slab track vibration and travelling speed. In this paper, a wireless sensor network system is deployed on a typical ballastless slab track in China, which uses acceleration nodes to collect vibration information when the train is passing by at different constant speeds. Then, after comparing the denoising effects of a hard threshold method, a soft threshold method and a Bayes wavelet method, the Bayes wavelet denoising method is used to remove the noise, while the spatial variability characteristics of the signal are preserved. Finally, the wavelet energy spectrum is adopted to obtain the duration and energy of the non-stationary vibration data. The time–frequency function curve is obtained to further analyse the physical behaviour of the vehicle–track system. A hammering experiment is conducted to show the importance of the results. This work facilitates a better understanding of the track vibration characteristics for monitoring the status of the track.