Minority-class identification-oriented multi-strategy joint fault diagnosis method

Abstract

The invention discloses a minority-class identification-oriented multi-strategy joint fault diagnosis method, which comprises the following steps of: performing equalization processing on sample data,performing step-by-step training on a constructed multi-strategy joint fault diagnosis model by adopting the equalized sample data, and constructing a DBN-based feature extractor, so that the deep features of most types of samples can be extracted, the shallow and deep features of minority types of samples are fused, and the minority type fault recognition rate is improved. Starting from multiplelevels of data, features and classifiers, the powerful data representation and feature extraction capacity of deep learning is fully utilized, the problem that minority class faults are difficult torecognize due to data class imbalance is solved, and the recognition effect of the minority class faults is comprehensively improved.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of industrial equipment, and in particular relates to a multi-strategy combined fault diagnosis method for minority class recognition. Background technique [0002] With the development of production and technological progress, the large-scale and complex modern industrial equipment has greatly increased the difficulty and cost of equipment maintenance. Problems such as equipment wear and tear, complex operating conditions, and dynamic and changeable production environments make failures inevitable. Once a failure occurs or is not repaired in time, the "light one" will affect the equipment's operating performance, production progress and even product quality. The "insider" may lead to catastrophic accidents such as production system paralysis, casualties, etc., causing huge losses to enterprises and society. Therefore, using deep learning to evaluate and predict the operating status of in...

AI Technical Summary

Problems solved by technology

Although these methods have achieved certain results, they still have the following deficiencies: First, when the existing sampling techniques balance the data set, it is easy to introduce too much noise, or it is difficult to increase the diversity of minority class samples, resulting in the performance of minority class recognition. The improvement is very limited; secondly, the existing classifier improvement method is difficult to take into account the diagnostic effect on other majority classes while improving the attention to the minority class, thus affecting the overall accuracy

Images