Intelligent fault diagnosis method for mechanical equipment based on partial transfer convolutional network

Abstract

The present invention provides an intelligent fault diagnosis method for mechanical equipment based on a partial migration convolutional network, which collects operating data of mechanical equipment under different operating conditions to form a data set, and uses part of the data in the data set X as the source domain training Sample set and target domain test sample set, and perform data standardization on each sample data, and then use the source domain training sample set to train two one-dimensional convolutional neural network models with the same structure but different initialization parameters and based on the target domain test sample set The two trained convolutional neural network models are corrected to obtain the convolutional neural network mechanical equipment fault diagnosis model, and the fault diagnosis model is used to diagnose the mechanical equipment based on real-time operating data and output the fault type. The present invention can be effectively used in more realistic mechanical fault diagnosis, that is, considering the unlabeled nature of the target domain, the trained diagnostic model can better diagnose mechanical equipment faults.

Description

Technical field[0001]The present invention relates to the field of mechanical fault diagnosis, and in particular, to a mechanical equipment intelligent fault diagnosis method based on partial migration convolutionary network.Background technique[0002]With the rapid development of society, mechanical equipment has become an important pillar of modern industries. However, due to continuous operation and environmental interference, mechanical equipment inevitably degraded and malfunction. Once a fault occurs, it will cause equipment shutdown and economic losses. He has caused personnel casualties. Therefore, health monitoring and fault diagnosis of mechanical equipment has caused more and more attention in the industry and academic community.[0003]In the past many years, various diagnostic methods have been proposed successively. Among them, deep neural networks have gradually been increasingly studied as a dazzling new star. At present, the intelligent diagnostic method based on deep ...

AI Technical Summary

Problems solved by technology

[0004] In recent years, although methods based on deep neural networks have made some achievements, the above three problems are usually ignored and cannot be solved well.

The existence of these problems will not only cause misdiagnosis and missed diagnosis, but also make the accuracy of fault diagnosis lower, and may even hinder the development of intelligent diagnosis.

Images