Fault Diagnosis of Check Valve Based on KPLS Optimal Feature Selection and Kernel Extreme Learning Machine
Round 1
Reviewer 1 Report (Previous Reviewer 2)
Review to ”coatings-1855062”
1. Positive aspects
”coatings-1855062” manuscript is located as topic in the area of ​​interest of ”Coatings” journal.
This manuscript has an actual and interesting research idea. The title is clear and it is adequate to the content of the paper. The abstract presents objects, methods and results. The methods utilised are limpid explained. This manuscript has substantial merits in terms of the correctness of the calculations and the accuracy of the relevant representations in the figures. The conclusions are accurate and supported by the content.
2. Negative aspects
2.1. This manuscript does not include in the literature review references from 2022.
2.2. It is true that there are a few words about "future research", but considering the importance of the topic, it is natural to have a "Future research" section.
3. Conclusion:
”coatings-1855062” manuscript requires minor corrections and additions:
a) to complete the literature review with the contribution from 2022,
b) analyzing the possibility of introducing a section on future research.
Comments for author File: 
Comments.pdf
Author Response
Dear reviewer
Thank you very much for your recent email noticing us the required revision of the mentioned paper, which we submitted to the journal ‘Coatings. Thank you for your thoughtful, helpful, and most kind review of manuscript coatings-1855062. We are very grateful to the editor and all the reviewers for their time spending making valuable suggestions and allowing us to submit revisions to the manuscript. We have carefully revised the entire manuscript, and every comment made by the reviewer has been accurately considered and adopted. Respond to the comments point by point below, and point out the modified content. At the same time, the revised text is highlighted in the new manuscript.
We would again like to thank you for your support in advance, and look forward to hearing from you soon.
Kind regards,
Xuyi Yuan
Faculty of Information Engineering and Automation,
Kunming University of Science and Technology,
Kunming, 650500, China
Author Response File: 
Author Response.docx
Reviewer 2 Report (New Reviewer)
This work describes a fault diagnosis of check valve based on KPLS optimal feature selection and kernel extreme learning machine. However, you should answer and revise the paper following comments:
1. Why did you use the Wavelet Transform for frequency domain feature extraction? I think that Fast Fourier Transform(FFT)also can be used for frequency domain feature extraction. Could you compare the fault diagnosis performance of two method?
2. Organize the advantages and disadvantages of the proposed method and other methods into a table and add it in the chapter 1 or 3.
3. Why don’t you use CNN(Convolutional Neural Network) algorithm based on time domain row data? If CNN used for fault diagnosis, is it possible to improve the diagnosis accuracy, isn't it? The (The accuracy rates for the time domain are so low in Table 4, 5 etc)
4. Did you perform the on line experiments for diagnosis of fault check valves based on proposed algorithm? If so, Which is the result figure?
Author Response
Dear editor and reviewers
Thank you very much for your recent email noticing us the required revision of the mentioned paper, which we submitted to the journal ‘Coatings. Thank you for your thoughtful, helpful, and most kind review of manuscript coatings-1855062. We are very grateful to the editor and all the reviewers for their time spending making valuable suggestions and allowing us to submit revisions to the manuscript. We have carefully revised the entire manuscript, and every comment made by the reviewer has been accurately considered and adopted. Respond to the comments point by point below, and point out the modified content. At the same time, the revised text is highlighted in the new manuscript.
We would again like to thank you for your support in advance, and look forward to hearing from you soon.
Kind regards,
Xuyi Yuan
Faculty of Information Engineering and Automation,
Kunming University of Science and Technology,
Kunming, 650500, China
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report (New Reviewer)
The comments of the reviewer were well reflected and revised.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
file attached
Comments for author File: Comments.pdf
Author Response
Dear editor and reviewers
Thank you very much for your recent email noticing us the required revision of the mentioned paper, which we submitted to the journal ‘Coatings’. Thank you for your thoughtful, helpful, and most kind review of manuscript coatings-1587230. We are very grateful to the editor and all the reviewers for their time spending making valuable suggestions and allowing us to submit revisions to the manuscript. We have carefully revised the entire manuscript, and every comment made by the reviewer has been accurately considered and adopted. Respond to the comments point by point below, and point out the modified content. At the same time, the revised text is highlighted in the new manuscript.
We would again like to thank you for your support in advance, and look forward to hearing from you soon.
Please see the attachment.
Kind regards,
Xuyi Yuan
Faculty of Information Engineering and Automation,
Kunming University of Science and Technology,
Kunming, 650500, China
Author Response File: Author Response.pdf
Reviewer 2 Report
Review to Coatings-1587230 manuscript
- Positive aspects
Coatings-1587230 manuscript is located as topic in the area of ​​interest of Coatings journal.
Judging positively, this manuscript has a interesting research idea. The title is clear and it is adequate to the content of the paper. The abstract presents objects, methods and results. The study methods are explained clearly. The conclusions are accurate and supported by the content.
- Negative aspects
2.1. This manuscript does not include in the literature review an interesting contribution in the field Yin, G., Zhang, Y. T., Li, Z. N., Ren, G. Q., & Fan, H. B. (2014). Online fault diagnosis method based on incremental support vector data description and extreme learning machine with incremental output structure. Neurocomputing, 128, 224-231.
2.2. The manuscript does not contain a section on future research.
- Conclusion
Coatings-1587230 manuscript requires minor corrections and additions:
- a) to complete the literature review with the contribution above mentioned,
- b) analyzing the possibility of introducing a section on future research.
Comments for author File: Comments.pdf
Author Response
Dear editor and reviewers
Thank you very much for your recent email noticing us the required revision of the mentioned paper, which we submitted to the journal ‘Coatings’. Thank you for your thoughtful, helpful, and most kind review of manuscript coatings-1587230. We are very grateful to the editor and all the reviewers for their time spending making valuable suggestions and allowing us to submit revisions to the manuscript. We have carefully revised the entire manuscript, and every comment made by the reviewer has been accurately considered and adopted. Respond to the comments point by point below, and point out the modified content. At the same time, the revised text is highlighted in the new manuscript.
We would again like to thank you for your support in advance, and look forward to hearing from you soon.
Please see the attachment
Kind regards,
Xuyi Yuan
Faculty of Information Engineering and Automation,
Kunming University of Science and Technology,
Kunming, 650500, China
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper presents a fault diagnosis approach to detect the current status of a check valve, which is one of the main components of high-pressure diaphragm pumps. This approach is based on the multi-feature dimensionality reduction method Kernel partial least squares (KPLS) and Kernel Extreme Learning Machine (KELM).
The paper deals with an interesting topic, although some minor and major observations must be addressed before a final decision can be taken.
1.-English grammar and style require a deep revision. In the current state, the paper is very hard to follow.
2.- Abstract. It requires some details about the physical variables that are measured in order to perform the fault detection approach.
3.- Introduction section: the novelties and contributions of the paper must be clearly and profusely stated, if necessary in the form of bullet points.
4.- The authors apply kernel extreme learning machine KELM. However, there are many alternative methods to solve this problem. The authors must clearly state the advantages of this method with respect the others.
5.-Section 4. The authors present results based on public datasets of bearings. However, the conditions in which the data was obtained, the measurements done, the measurement devices used and their characteristics (operational limits, uncertainty, accuracy, etc.) are poorly described.
6.- Section 4. The authors present results based on public datasets of bearings. However, the paper is related to detect the current status of a check valve. I cannot see any relationship. Please remove from the text.
7.- Table 5. Results presented are a little ambiguous because there is not a clear trend of the number of nodes in the hidden layer. Perhaps with 100 or 200 nodes or even less it is enough. How do you select the number? It is not possible an automatic selection?
8.- Section 4.2 describes the sensors and instrumentation used, but more information is required, as the reference, manufacturer, accuracy, etc.
9.- Section 4.2. It is not clear which types of faults are defined and how a given pattern is associated to a determined fault type. Which indicators are used? How the pattern is associated to a determined type of fault? This must be clearly described in the paper.
10.- Figs. 8 have low quality.
11.- The paper must be focused to analyze the health status of the check valve, so the analysis of the bearings is very confusing. The sample set, the type of faults analyzed, the division made to obtain the training and validation sets, etc. need a more detailed description. The results presented are also confusing. For example, Table 7 presents the classification accuracy, but the different classes (type of faults) of the problem are not clear. Therefore, I suggest to completely rework the paper.
The Reviewer believes that the questions/comments above can help to improve the quality and understandability of the paper.
Round 2
Reviewer 1 Report
The manuscript has been modified by the authors accordingly. Only Figure 8 is not very much up to the mark. There are six subfigures in one without markings and descriptions. I am sure authors can spare some more time in making the figure presentable so as to make it comprehensible to the readers. After all, it is the main experimental setup.
Author Response
Dear editor and reviewers
Thank you very much for your recent email noticing us the required revision of the mentioned paper, which we submitted to the journal ‘Coatings’. Thank you for your thoughtful, helpful, and most kind review of manuscript coatings-1587230. We are very grateful to the editor and all the reviewers for their time spending making valuable suggestions and allowing us to submit revisions to the manuscript. We have carefully revised the entire manuscript, and every comment made by the reviewer has been accurately considered and adopted. Respond to the comments point by point below, and point out the modified content. At the same time, the revised text is highlighted in the new manuscript.
We would again like to thank you for your support in advance, and look forward to hearing from you soon.
Please see the attachment.
Kind regards,
Xuyi Yuan
Faculty of Information Engineering and Automation,
Kunming University of Science and Technology,
Kunming, 650500, China
Author Response File: Author Response.pdf
