Review Reports

Round 1

Reviewer 1 Report (New Reviewer)

Comments and Suggestions for Authors

Multi-Attention Mechanisms and Conditional GAN-Based Data Balancing are difficult to understand. Authors shall explain still simple way how collected data is processed despite framework is clearly defined in the manuscript.  The following issues should be addressed in the revision.

Comment: Please increase font size in the Fig.5 so that it is easily readable. The axis legend is missing for Fig. 5(a).

Comment: Highlight the limitations of the study.

Comment: In the conclusion which materials pipe posses more leakage and why it is so? shall be explained.

Comment: Where is the equation for leakage ratio? Why does this ratio decrease after 20 years of age?

Comment: There is no clarity how does equations 1 and 2 help data enhancement?

Comment: Authors shall clearly explain which connection type vulnerable to more leakage.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Brief summary

This paper is interesting for those related to water distribution network management.

Authors present a data-driven pipeline leakage risk assessment model for water distribution networks based on multiple attention mechanisms and a generative model-based data balancing method.

Authors apply the model to an interesting case study, B2 and B3 metering areas in Zhengzhou City, Central China, and draw some conclusions.

The paper follows a traditional structure, has a good set of references, including some recent ones, and presents interesting conclusions.

In my opinion the paper should be accepted for publication after taking into consideration the comments presented below.

 

Specific comments

After introducing acronyms (like WDN, SD, GA, NER, GDP) authors should use them in the remaining text. Acronyms like SMOTE should be explained before being used. Acronym TSNE-KDE was not explained.

Authors should avoid using “we”, “us” and “our”.

Line 122: 46%+52%=98%, there are 2% missing?

Line 214: suggest changing “number f features” to “number of features”.

Line 219: what is the meaning of “S” in “Multiple Attention Block (MSA Block)”? Figure 3 presents “MHSA Block” and in line 233 “multi-head attention”. Should it be “Multi-head self-attention block (MSA Block)”?

Line 265: suggest changing “analyses” to “analysis”.

Lines 344-346: isn’t this the same that was already presented in lines 319-321?

Figure 6: is there any explanation to the fact that pipes with age between 15 and 20 years present the biggest leakage ratios? Is this related to the pipe material or joint system?

Authors didn’t fill the Acknowledgments section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

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

Comments and Suggestions for Authors

The authors need to be more well versed on the hydrology terms in English.  Proofreading and formatting prior to the submission is required.

Comments on the Quality of English Language

I can appreciate the originality of the work in terms of data collection and case study in the pipeline water supply leakage assessment. 

However, the paper has many errors in the terminology used. The authors need to be more well versed on the hydrology terms. 

My first impression is that the title and type of paper looks like a review paper; however, the number of references is too small. If it is a technical paper contribution, the methodology is not very clear, and result analysis is not depth enough to be presented. 

 

Author Response

See annex for details.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

See comments on attached documents

Comments for author File: Comments.pdf

Comments on the Quality of English Language

 Good. Occasional issues with a choice of words

Author Response

See annex for details.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript explained the pipeline data of the water supply pipe network in a city in central China. The application cases of the data-driven pipe network health assessment model based on the whole process and multiple models in the actual project were presented in the study. The following points are suggested to improve the article quality.

1. The abstract needs to revise and numerical data should present in this section.

2. The study area didn't explain and pipe network should present in pictorial view. 

3. All the figures need to improve the graphics.

4. The literature should be added to support the models used for this study 

5. Overall English editing is required in all sections

Comments on the Quality of English Language

English editing required to improve the article

Author Response

See annex for details.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

All Figures need to improve the graphics and text in readable form. 

Comments on the Quality of English Language

NA

Author Response

Comments 1: All Figures need to improve the graphics and text in readable form. 

Response 1: Thank you very much for your valuable advice. After our review, the images do still have some issues. We have recreated the Figure 1 frame image and have made some adjustments to the content of the other images as well as the formatting.

-L112 We have reworked Figure 1.

-L200 We have adjusted the font and resolution of the image content in Figure 2.

-L231 We have adjusted the font and resolution of the image content in Figure 3.

-L313 We have reorganized Figure 4.

-L394 We have processed and reorganized the content of Figure 5.

-L428 We have processed and reorganized the content of Figure 6.

-L459 We have processed and reorganized the content of Figure 7.

Author Response File: Author Response.docx