Bearing composite fault diagnosis method based on multi-label field adaptive model

Abstract

The invention provides a method for diagnosing a bearing compound fault in an unknown target domain under a variable working condition. The method comprises the following steps: constructing a fault feature extractor consisting of a deep residual network based on a multi-layer domain adaptive method; inputting a preprocessed bearing vibration signal, and carrying out distribution difference matching on features, extracted through a plurality of residual blocks, of source domain data and target domain data to obtain migratable features; representing the composite fault as a combination of single faults through multi-label learning; and a binary association strategy is used to train corresponding binary classifiers for various single faults, and the features of the single faults are separated from the composite faults and are diagnosed respectively. According to the method, the problems that a traditional diagnosis scheme depends on expert knowledge and is difficult to effectively decouple and recognize the composite fault are solved, accurate diagnosis of the composite fault of the bearing under variable working conditions is achieved, meanwhile, dependence of an existing method on marked data is eliminated, and accurate diagnosis can be conducted on a related but invisible target domain.

Description

technical field [0001] The invention relates to the fields of mechanical fault diagnosis, artificial intelligence, signal processing and the like, in particular to a mechanical composite fault diagnosis method based on a self-adaptive model in a multi-label field. Background technique [0002] In the field of rolling bearing fault diagnosis, deep learning algorithms can intelligently mine the essential features related to bearing faults from vibration signals, thus achieving good diagnostic results. However, most of the current research is aimed at a single category of bearing failures. In practice, bearing failures often appear in the form of composite failures composed of multiple single failures. Compound failure usually occurs when multiple key parts of rotating machinery bearings are damaged at the same time, and it is also the main cause of catastrophic accidents in rotating machinery. Composite faults are more harmful to equipment and more difficult to diagnose than ...

AI Technical Summary

Problems solved by technology

[0003] At present, there are mainly the following problems that restrict the development of bearing composite fault diagnosis technology: 1) The bearing signal of composite fault usually contains multiple characteristic signals and strong aliasing noise, which is often nonlinear composite and interferes with each other, so it is difficult to accurately carry out theoretical analysis. decoupled and thus easily lead to false identifications during diagnostics

2) The composite fault signal has no obvious periodicity, and the amplitude changes are also large. The peak value at the fault characteristic frequency is not obvious, and the periodic component is almost completely covered under the influence of noise.

Traditional time domain frequency domain signal processing methods are difficult to extract the characteristics of complex faults, so it is difficult to be effective in complex fault diagnosis problems

3) Existing methods often regard composite faults as a brand-new failure mode, ignoring the potential relationship between them and single faults. Composite faults are actually a combination of single faults, and it is difficult for such methods to accurately classify them as multiple single failures

4) When facing actual engineering applications, because the distribution of the target data set to be diagnosed is different from that of the training data set, the effect of the model will often decline significantly

[0004] In order to solve the problem that complex fault features are difficult to extract and identify, the present invention introduces a multi-label learning (Multi-label Learning, ML) method to express and classify complex faults, which not only gets rid of the traditional signal processing methods that require professional technology and fault diagnosis The reliance on empirical knowledge also solves the problem that existing deep learning methods ignore the relationship between compound faults and single faults, and lack rigor and reliability.

Images