Abstract
Recent progress on intelligent fault diagnosis (IFD) has greatly depended on deep representation learning and plenty of labeled data. However, machines often operate with various working conditions or the target task has different distributions with the collected data used for training (the domain shift problem). Besides, the newly collected test data in the target domain are usually unlabeled, leading to unsupervised deep transfer learning based (UDTL-based) IFD problem. Although it has achieved huge development, a standard and open source code framework as well as a comparative study for UDTL-based IFD are not yet established. In this paper, we construct a new taxonomy and perform a comprehensive review of UDTL-based IFD according to different tasks. Comparative analysis of some typical methods and datasets reveals some open and essential issues in UDTL-based IFD which are rarely studied, including transferability of features, influence of backbones, negative transfer, physical priors, etc. To emphasize the importance and reproducibility of UDTL-based IFD, the whole test framework will be released to the research community to facilitate future research. In summary, the released framework and comparative study can serve as an extended interface and basic results to carry out new studies on UDTL-based IFD. The code framework is available at \url{https://github.com/ZhaoZhibin/UDTL}.
Highlights
W ITH the rapid development of industrial big data and the Internet of Things, prognostic and health management (PHM) for industrial equipment, such as aeroengine, helicopter, and high-speed train, is becoming increasingly popular, bringing out many intelligent maintenance systems
We can observe that CWRU and JNU can achieve an accuracy of over 95%, and other datasets can only achieve an accuracy of around 60%
The results of JMMD are better than those of MK-MMD, which indicates that the assumption of joint distribution in source and target domains is useful for improving the performance
Summary
W ITH the rapid development of industrial big data and the Internet of Things, prognostic and health management (PHM) for industrial equipment, such as aeroengine, helicopter, and high-speed train, is becoming increasingly popular, bringing out many intelligent maintenance systems. IFD based on traditional machine learning methods [1], including random forest [2] and support vector machine [3], has been widely applied in research and industry scenarios. These methods often need to Manuscript received June 26, 2021; revised August 30, 2021; accepted September 16, 2021. ZHAO et al.: APPLICATIONS OF UDTL TO IFD: SURVEY AND COMPARATIVE STUDY domain are all available, and the source domain can be defined as follows: Ds =. Where Dt represents the target domain, xit ∈ Rd is the i th sample, Xt is the union of all samples, and nt means the total number of target samples
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callMore From: IEEE Transactions on Instrumentation and Measurement
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
