Application of the refined multiscale permutation entropy method to fault detection of rolling bearing

Abstract

To detect the fault status and identify the fault category of rolling bearings in a timely manner, a method based on refined composite multiscale permutation entropy (RCMPE) and a support vector machine is proposed. Multiscale permutation entropy (MPE) is an improved method based on permutation entropy. Although this method can effectively perform dynamic analysis of nonlinear signals and fault monitoring of rolling bearings, the method is likely to lose some key information when shortening the time series during the identification of coarse-grained fault signals, which reduces the reliability of the results. To overcome this obstacle, the RCMPE algorithm is utilized to extract bearing feature information, and it is compared and analyzed with MPE and the multiscale entropy (MSE). Through simulation and experimental verification of the signal, it is found that as the scale factor increases, RCMPE can retain more useful information. The calculation results of fault identification are more accurate and have better robustness and stability. This method has less dependence on data length and is more stable in entropy estimation. This new diagnostic method is verified by actual bearing test data. The results show that the method can successfully distinguish the fault type and damage degree of rolling bearings and verify the effectiveness and superiority of the proposed method.