Abstract
Current studies on intelligent bearing fault diagnosis based on transfer learning have been fruitful. However, these methods mainly focus on transfer fault diagnosis of bearings under different working conditions. In engineering practice, it is often difficult or even impossible to obtain a large amount of labeled data from some machines, and an intelligent diagnostic method trained by labeled data from one machine may not be able to classify unlabeled data from other machines, strongly hindering the application of these intelligent diagnostic methods in certain industries. In this study, a deep transfer learning method for bearing fault diagnosis, domain separation reconstruction adversarial networks (DSRAN), was proposed for the transfer fault diagnosis between machines. In DSRAN, domain-difference and domain-invariant feature extractors are used to extract and separate domain-difference and domain-invariant features, respectively Moreover, the idea of generative adversarial networks (GAN) was used to improve the network in learning domain-invariant features. By using domain-invariant features, DSRAN can adopt the distribution of the data in the source and target domains. Six transfer fault diagnosis experiments were performed to verify the effectiveness of the proposed method, and the average accuracy reached 89.68%. The results showed that the DSRAN method trained by labeled data obtained from one machine can be used to identify the health state of the unlabeled data obtained from other machines.
Highlights
Intelligent fault diagnosis can be successful only when two conditions are met [1]
The data are currently labeled manually in most cases, so that the labeling of a large amount of data is expensive and time-consuming. ird, the probability distributions of data obtained from different machines are different, and the classification performance of the intelligent fault diagnosis methods can be significantly weakened when the training and test data sets are obtained from different machines
Based on the abovementioned studies, currently available intelligent fault diagnosis methods based on transfer learning mainly focus on the conversion between different working and loading conditions
Summary
Intelligent fault diagnosis can be successful only when two conditions are met [1]. First, the model should be trained by a large amount of labeled fault data. Che et al [7] proposed a deep transfer learning method for rolling bearing fault diagnosis under variable operating conditions. Wen et al [8] developed a deep transfer learning method for fault diagnosis that was tested on bearing data sets collected under different loading conditions. Based on the abovementioned studies, currently available intelligent fault diagnosis methods based on transfer learning mainly focus on the conversion between different working and loading conditions. In engineering practice, it is difficult to obtain a large amount of labeled data from a machine for model training. A deep transfer learning method for bearing fault diagnosis based on domain separation and adversarial learning, domain separation reconstruction adversarial networks (DSRAN), was proposed. Traditional intelligent fault diagnosis methods collect the training and test data sets from the same machine.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
