Fault monitoring and diagnosis method based on multi-feature fusion and width learning

Abstract

The invention provides a fault monitoring and diagnosis method based on multi-feature fusion and width learning. The method comprises the following steps: acquiring a historical process signal in an actual industrial process; performing multi-feature extraction based on the historical process signal, and constructing a multi-feature vector; inputting the multi-feature vector into a width learning network for training to obtain a fault diagnosis model; applying the fault diagnosis model to the actual industrial process for real-time fault diagnosis; adopting an evaluation index F1-score to evaluate the diagnosis performance of the fault diagnosis model, and if Micro F1 is larger than a first set threshold value and Macro F1 is larger than a second set threshold value, continuing to use the fault diagnosis model; otherwise, performing training again. According to the invention, a multi-feature fusion method is adopted, more dynamic information is provided for fault diagnosis, the accuracy of fault diagnosis is improved, the problems that the calculation amount is increased and the fault diagnosis efficiency is reduced due to multi-feature fusion are solved, and the requirements of actual industrial real-time monitoring and diagnosis can be met.

Description

technical field [0001] The invention relates to the technical field of industrial process fault monitoring and diagnosis, in particular to a fault monitoring and diagnosis method based on multi-feature fusion and width learning. Background technique [0002] With the advancement of science and technology and the development of modern productivity, the degree of automation of production equipment has been greatly improved, with more complete functions and more complex structures. Due to many unavoidable unfavorable factors in the industrial process, the performance of the equipment is reduced or the original function is lost. If it is not discovered and intervened in time, it may even lead to catastrophic accidents. With the development of informatization and intelligent technology, the number of variables that need to be monitored in industrial processes has increased significantly, and it is difficult for the traditional diagnostic mode using manual experience to cope with ...

AI Technical Summary

Problems solved by technology

In order to obtain better diagnostic results, most researches on fault monitoring and diagnosis based on deep neural networks focus on stacking deeper structures or optimizing model parameters, which leads to a large amount of computation in the process of structure adjustment and parameter optimization. resource

However, such expensive computing costs are often unbearable in actual industrial production, and factories pay more attention to the real-time performance, light weight and economy of the system

Images