Railway rolling bearing fault diagnosis based on multi-scale intrinsic mode function permutation entropy and extreme learning machine classifier

The application of the multi-scale intrinsic mode function permutation entropy and extreme learning machine classifiers in railway rolling bearing fault diagnosis is here proposed in this article. The original signal is first denoised using wavelet de-noising as a pre-filter, which improves the subsequent decomposition into a number of intrinsic mode functions using ensemble empirical mode decompose. Second, the multi-scale intrinsic mode function permutation entropy is extracted as feature parameters. Finally, the extracted features are entered into extreme learning machine for an automated fault diagnosis procedure. Case studies have been carried out to evaluate the validity of the approach. The results demonstrate its effectiveness for diagnosis of faults in railway rolling bearings.