A Fault Diagnosis Method for Rolling Bearings Based on Composite Multiscale Permutation Entropy

Abstract

The invention discloses a rolling bearing fault diagnosis method based on composite multi-scale permutation entropy, which belongs to the technical field of fault diagnosis. The steps of the present invention are: measuring the vibration signal of the faulty object; extracting the composite multi-scale permutation entropy from the vibration signal; reducing the dimension of the composite multi-scale permutation entropy value by using the Laplacian score; reducing the score after dimension reduction The first several composite multi-scale permutation entropies are divided into multiple training samples and multiple test samples as fault feature vectors; multiple training samples are input into a multi-fault classifier based on support vector machine for learning to classify test samples ; Identify the working state and fault type of the fault object according to the classification result. The fault diagnosis method proposed by the invention has high innovation in the feature extraction process and high recognition degree in the fault pattern recognition process.

Description

technical field [0001] The present invention relates to the technical field of rolling bearing fault diagnosis, in particular to a feature selection based on composite multi-scale permutation entropy (Composite multi-scale permutation entropy, CMPE), Laplacianscore (Laplacianscore, LS) and support vector machine (Support vector) machine, SVM) rolling bearing fault diagnosis method. Background technique [0002] Due to the complexity of the mechanical system, friction, vibration and load will inevitably occur during the operation of the equipment, and the vibration signal of the system often exhibits nonlinear behavior. Therefore, the nonlinear analysis method has more unique advantages than the linear analysis method in extracting fault features, and can extract fault feature information hidden in vibration signals that cannot be extracted by other methods. In recent years, many nonlinear analysis methods, such as fractal, approximate entropy, sample entropy and permutation...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the problem that the permutation entropy value in the multi-scale permutation entropy has a large deviation with the increase of the scale factor, and at the same time to improve the efficiency of fault diagnosis and reduce the influence of human experience factors on the diagnosis result, the present invention provides a method based on composite multi-scale permutation entropy. Rolling bearing fault diagnosis method based on scale permutation entropy, Laplacian score and support vector machine; the present invention can better extract the nonlinear features of vibration signals in the feature extraction process, and also has higher accuracy in the pattern recognition process Fault identification

Images