Doppler Effect removal based on instantaneous frequency estimation and time domain re-sampling for wayside acoustic defective bearing detector system

Abstract

The phenomenon of Doppler Effect in the acoustic signal recorded by the wayside acoustic defective bearing detector (ADBD) leads to the difficulty for fault diagnosis of train bearings with a high moving speed, which is a barrier that would badly reduce the effectiveness of online defect detection. In order to improve the performance of condition monitoring of the bearings on a passing train with microphones amounted besides the railway, the elimination of the Doppler Effect should be solved firstly. An effective method for removing the Doppler Effect embedded in the source signal is presented in this paper. The Short Time Fourier Transform-Viterbi Algorithm (STFT-VA) is applied to obtain instantaneous frequency estimation of the distorted signal. According to the acoustic theory of Morse, the non-linear data fitting is then carried out to get the fitting instantaneous frequencies. The necessary parameters for time domain interpolation re-sampling, which is totally based on the kinematic analysis, are acquired from the fitting curve and the re-sampling sequence could be established in the time domain. As a result of the preceding steps, the fault diagnosis for the train bearings could be implemented with the restored signal. The effectiveness of this proposed method is verified by means of a simulation with three adjacent frequencies and an experiment with practical acoustic signals of train bearings with a crack on the outer race and the inner race. The results of the simulation and the experiment indicate that the proposed method has an excellent performance in removing Doppler Effect, and could be well employed to the condition monitoring and fault diagnosis of train bearings with a high moving speed.