A Fault Diagnosis Method of Axle Box Bearing Based on Three-Direction Vibration Acceleration Fusion and Adaptive Bandpass Filtering

Axle box bearing is a key component of the bogie transmission system and its health state directly affects the safety of train operation. It is necessary to establish a more scientific and efficient fault diagnosis method for axle box bearing, so as to effectively extract bearing fault characteristic information under strong interference noise. In this paper, the vibration acceleration signal is taken as target signal, and a bearing fault characteristic frequency extraction method based on data-level fusion and adaptive band-pass filtering denoising signal demodulation spectrum is proposed. The principal components analysis method is used to realize the fusion of the three-direction vibration acceleration, and reduce the amount of irrelevant noise. The improved kurtosis index is used to select the filter signals of different filters, and the optimal filter signal is selected. Finally, the bearing fault simulation signal and experimental signal are used to verify the effectiveness of the proposed method. The analysis results show that the proposed method can extract the fault characteristic information from the axle box bearing vibration acceleration signal under strong interference noise.