Rail crack detection based on the adaptive noise cancellation method of EMD at high speed

In order to improve the accuracy of rail crack detection, especially for the high-speed condition, this paper proposes a rail crack detection method based on the adaptive noise cancellation method of empirical mode decomposition (EMD). The acoustic emission (AE) signals of simulated crack signals are obtained by the rail-wheel test rig, and these signals are decomposed into intrinsic mode functions (IMFs) by EMD. The cross-correlation analysis is used to reserve the IMFs which contains useful information. Then, the denoising signals are reconstructed based on the reserved IMFs, and wavelet transform is employed to detect the crack signals from the denoising signals. Signals at different speeds are investigated by the proposed method and the interference of noise signals is suppressed effectively. The results illustrate that the proposed method can detect the rail crack signals effectively at high speed with noise.