Wayside acoustic fault diagnosis of railway wheel-bearing paved with Doppler Effect reduction and EEMD-based diagnosis information enhancement

Wayside acoustic monitoring technique is promising for health monitoring of wheel-bearing for railway vehicles. However, due to the high relative moving speed between the railway vehicle and the wayside mounted microphones, the recorded signal is embedded with Doppler Effect. What's more, the background noise is relatively heavy which makes it difficult to extract the diagnosis relevant information. To solve these problems, this paper introduced a railway wheel-bearing wayside acoustic fault diagnosis scheme based on Doppler Effect reduction and Ensemble Empirical Mode Decomposition (EEMD). Firstly, an improved Doppler Effect reduction method is introduced incorporating with the kinematic parameters estimation and signal re-sampling method. Secondly, the EEMD is employed to extract the diagnosis relevant Intrinsic Mode Function (IMF). Finally, the envelope spectrum analysis is employed to identify the local fault. The effectiveness of the proposed method is verified by experimental cases analysis.