A Fault Diagnosis Method for Bearing Equipment

Abstract

The invention discloses a fault diagnosis method for bearing equipment, and relates to the technical field of bearing equipment fault diagnosis. The present invention first uses the resampling method to process the unbalanced data samples, then uses the deep learning algorithm to extract the feature patterns of the newly added data, then inputs the newly added patterns into the original integrated model for classification, and further proposes a dynamic forgetting weight algorithm , an effective incremental screening algorithm, an incremental information fusion algorithm, and dynamically adjusts the weight of the incremental merge mode according to the calculated value. Finally, the weighted patterns are classified using incremental ensemble SVM in a supervised manner, which enables real-time extraction of bearing equipment state features and reliable classification of failure modes.

Description

technical field [0001] The invention relates to the technical field of bearing equipment fault diagnosis, in particular to a bearing equipment fault diagnosis method based on incremental fusion and dynamic weight integrated learning. Background technique [0002] Bearing equipment is mostly used in important fields such as industry, aviation, and national defense, and its failure consequences are relatively serious. Therefore, how to extract fault feature information from the operating state data of bearing equipment and conduct effective analysis, so as to complete fault diagnosis and prediction is becoming increasingly important, and has become an intelligent manufacturing industry. research hotspots in the field. Especially in the context of Industry 4.0, with the development of industrial Internet of Things and information technology, a large amount of operating status data emerges in the production process, making it possible to use big data analysis methods for fault d...

AI Technical Summary

Problems solved by technology

For example, the influence of noise data is not considered in the process of resampling, resulting in a serious decrease in classification accuracy

Although some methods take into account the influence of sample distribution information on the data processing process, they ignore the changes in the sample distribution state with the generation of massive new data.

These problems hinder the further application of the above-mentioned unbalanced processing method in the field of equipment fault diagnosis

Images