Rail Crack Detection Using Adaptive Threshold and Wiener Filter Method Based on Acoustic Emission Technology

As a new method to detect rail cracks, acoustic emission (AE) technology has the advantages of dynamic crack detection, unlimited geometric shape, high sensitivity and real-time online monitoring. However, it is easily contaminated by the noise signals which restrain the application of AE technology. In order to solve this problem, this paper adopts the adaptive threshold method and Wiener filter to study the denoising method. For adaptive threshold method, the optimal time window is obtained from the intersection of confidence intervals (ICI) method. Based on the prior information of the signal, the Wiener filter coefficients are calculated. The Wiener filter is used for signal preprocessing. And then the adaptive threshold denoising method is used to detect crack signal. The simulation crack signals are used to analyze the improved method. It is also verified by the AE signals acquired from the experiment in real operating environment. The results clearly illustrate that the proposed method is effective. It can provide a useful guidance for AE detection of rail cracks.