Chemical fiber winding machine bearing fault feature extraction method combining GEP and mutual information

Abstract

The invention provides a chemical fiber winding machine bearing fault feature extraction method combining GEP and mutual information. Feature extraction for rolling bearings is programmed using improved gene expressions with evolutionary search performance. In order to improve the availability rate of gene expression programming fitting features and the combination degree of bearing features in an industrial scene, a feature selection method based on mutual information is provided, correlation between current gene expression programming features and fault tag sequences and redundancy between the current gene expression programming features and current candidate gene expression programming feature sets are analyzed respectively, and the fault tag sequences and the current candidate gene expression programming feature sets are analyzed respectively. And on this basis, the complementarity between the current candidate subset and the time-frequency domain discrete feature set is analyzed. Compared with the original feature pair classification precision, the method has the advantage that dominant features can be better reserved in the gene expression programming training process. The method can also be applied to similar industrial scenes with changeable working conditions, fault feature extraction is helped, and efficient fault diagnosis is achieved.

Description

technical field [0001] The invention relates to a bearing fault feature extraction method of chemical fiber winding machine combining gene expression programming (GEP) and mutual information, which belongs to the field of fault diagnosis and machine learning. Background technology [0002] The normal operation of chemical fiber equipment plays a vital role in the quality of chemical fiber products, the safety of chemical fiber production and the normal use of chemical fiber equipment. Chemical fiber production includes melting, extrusion, wire spraying, bunching, winding, texturing and other steps. Chemical fiber winding is a very important and complex process in the process of chemical fiber production. The equipment used for chemical fiber winding - chemical fiber winding machine has the characteristics of complex structure, many moving parts and continuous operation. Once failure occurs, it will directly affect the forming quality of chemical fiber thread. Therefore, the fault...

AI Technical Summary

Problems solved by technology

However, GEP continues to use common selection methods such as roulette and tournaments in the selection of offspring, which cannot be fully combined with fault information and cannot be directly used to extract the fault characteristics of winding chuck bearings.

Images