Wayside Acoustic Diagnosis of Axle Box Bearing Based on Fault Feature Extraction Algorithm

Abstract

As one of the key components of railway vehicles, the operation condition of the axle box bearing has a significant effect on traffic safety. The wayside monitoring sound of train axle box bearing is an amplitude modulation and frequency modulation signal with complex train running noise. Although empirical mode decomposition (EMD) and some improved time-frequency algorithms have been proved to be useful in bearing vibration signal processing, it is hard to extract the bearing fault signal from serious trackside acoustic background noises by using those algorithms. Therefore, a kurtosis-optimization-based wavelet packet (KWP) feature extraction algorithm is proposed, as the kurtosis is the key indicator of bearing fault signal in time domain. After beamforming of microphone array, the assessment of KWP is conducted by comparing with exiting algorithms. The test results of 50 fault bearing data indicate that the KWP is more efficient than high frequency resonance technique (HFR) and EMD in an environment where authentic railway noise were present.