Bearing Fault Diagnosis of Split Attention Network Based on Deep Subdomain Adaptation
Round 1
Reviewer 1 Report
The explanation of the contributions of the paper should be improved and better highlighted.
The background of the paper should be extended and better explained, as the paper is addressed to researchers on PHM (or other related domains) and not to ML/DL researchers. For example, the acronyms MMD or LMMD are never explicited.
Author Response
Dear Reviewer and Editor,
We gratefully appreciate the editors and all reviewers for their time spend making positive and constructive comments. These comments are all valuable and helpful for revising and improving our manuscript entitled “Bearing fault diagnosis of split attention network based on deep subdomain adaptation”(ID: applsci-1994625), as well as the important guiding significance to our researches. We have revised the manuscript in accordance with the remarks made by the reviewer.
See the attachment for details of modification.
We hope that our reply could be satisfactory and the manuscript could be accepted for publication.
Thank you very much for your time spent on this manuscript!
Best Regards
Prof. Haitao Wang
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors investigate the bearing fault diagnosis accuracy of several methods. They also propose a new method to mitigate the problems, such as, insufficient learning ability and lack of learning labeled data for training. The proposed method presents good accuracy when compared with others using CWRU datasets. The manuscript is well written, the method is clearly presented. Overall, the manuscript present enough new results in order to be published in Applied Sciences and deserves publication. My recommendation is to publish as is.
Author Response
Dear Reviewer and Editor,
We gratefully appreciate the editors and all reviewers for their time spend making positive and constructive comments. These comments are all valuable and helpful for revising and improving our manuscrip entitled “Bearing fault diagnosis of split attention network based on deep subdomain adaptation”(ID: applsci-1994625), as well as the important guiding significance to our researches. We have revised the manuscript in accordance with the remarks made by the reviewer.
Thanks a lot for your direct recommendation of publication for our manuscript. And, thank you very much for your recognition of our works. We wish you much success in your future work.
Best Regards
Prof. Haitao Wang
Author Response File: Author Response.pdf
Reviewer 3 Report
Dear Authors,
I found your article very interesting, but the manuscript requires some changes to the present form before accepting it for publication. I hope that the list of following remarks will help the authors to improve the scientific quality of the paper:
1. There is no novelty coming from the research, which needs to be specified in the Abstract or in the Introduction. The mention about the Wavelet analysis and later signal decomposition is not enough.
2. Over the text, there are number of typos, however the manuscript is well formatted. Please revise the manuscript again.
3. In the introduction, I miss the mention about the another graphical representation of the time-series, which the recurrence analysis is. Please refer to the 2 following papers related with your work:
Analysis of dynamic response of a two degrees of freedom (2-DOF) ball bearing nonlinear model. Applied Sciences (2021), 11(2), pp. 1-23, 787.
Detection of cylinder misfire in an aircraft engine using linear and non-linear signal analysis. Measurement (2021), 174, 108982.
4. What kind of Wavelet method do you use? I mean, what is the length of window, what type of window and type of wavelets do you use?
5. What are the differences between the experimental time-series?
6. What types of bearing’s damages do you analyse? There is no mention in the manuscript.
7. There are no conclusion coming from research. What are the further steps of your work?
After improving above described issues in the paper I’d like to give my positive opinion on signing my review report.
Author Response
Dear Reviewer and Editor,
We gratefully appreciate the editors and all reviewers for their time spend making positive and constructive comments. These comments are all valuable and helpful for revising and improving our manuscript entitled “Bearing fault diagnosis of split attention network based on deep subdomain adaptation”(ID: applsci-1994625), as well as the important guiding significance to our researches. We have revised the manuscript in accordance with the remarks made by the reviewer.
See the attachment for details of modification.
We hope that our reply could be satisfactory and the manuscript could be accepted for publication.
Thank you very much for your time spent on this manuscript!
Best Regards
Prof. Haitao Wang
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have deployed a real effort to improve the paper's quality and respond to the issues I raised in the first review. Thus, I recommend the paper's acceptance.
Reviewer 3 Report
Dear Authors,
I recommend paper for its publishing in the present form. All remarks have been introduced.
Reviewer