Next Article in Journal
A Message Passing Approach to Biomedical Relation Classification for Drug–Drug Interactions
Next Article in Special Issue
A Technical Device for Determining the Predispositions of Students—Air Traffic Controllers and Pilots during Multitasking Training
Previous Article in Journal
Temporal Logic Planning and Receding Horizon Control for Signal Source Localization
Previous Article in Special Issue
Comparison of Optimization Techniques and Objective Functions Using Gas Generator and Staged Combustion LPRE Cycles
 
 
Article
Peer-Review Record

Heat Conduction Plate Layout Optimization Using Physics-Driven Convolutional Neural Networks

Appl. Sci. 2022, 12(21), 10986; https://doi.org/10.3390/app122110986
by Yang Sun 1,*, Abdussalam Elhanashi 2, Hao Ma 3 and Mario Rosario Chiarelli 1
Reviewer 1:
Reviewer 2: Anonymous
Appl. Sci. 2022, 12(21), 10986; https://doi.org/10.3390/app122110986
Submission received: 31 August 2022 / Revised: 24 October 2022 / Accepted: 25 October 2022 / Published: 30 October 2022

Round 1

Reviewer 1 Report

1.     Authors propose a Physics-driven Convolutional Neural Networks (PD-CNN) method to infer the physical field solutions for randomly varied loading cases.

2.     More description is needed here. “. Therefore, the objective of layout optimization is to establish an available geometry that provides a significant heat distribution of the domain and thermal environment”

3.     Section 2 must be literature review (related work). Authors must discuss recent advancements in the field in this section.

4.     Major contributions of the work must be highlighted in a separate paragraph of the introduction section.

5.     Authors are advised to add suitable mathematical background (PSO) for the mentioned algorithms in order to enhance the overall quality of the manuscript.

6.     The considered simulation parameters must be presented before results and discussion.

7.     Authors fail to compare the proposed scheme with existing techniques. Why is the proposed scheme superior ? This is not justified.

8.     Authors must cite appropriate recent journal articles in order to enhance the overall readability of the manuscript.

Author Response

Dear Reviewer,

Please find the attachment.

 

Best regards,

Author Response File: Author Response.docx

Reviewer 2 Report

The subject is interesting. The article has been made effective with deep learning. But it has some shortcomings. These shortcomings are given. It would be appropriate to reevaluate after updates.

1)In the abstract section, the success of the proposed method should be expressed with numerical information.

2)Why is this study innovative? How does it contribute to the literature? It should be explained in the last paragraph of the Introduction Section

3)The research questions of the study should be added to the Introduction section.

4)As shown in ?? (Line 119) Please, correct it. There is a similar situation for other figures. All should be checked.

5) Provide reference for Physics-driven Training Approach (Line 152 and beyond). This section should be supported by reference.

6)Why was PSO chosen? It should be specified in the article.

7)The abbreviation of the FEM method is presented. The open form should be written in the first used place.

8) Why is the proposed approach only compared with the FEM method? It is also  recommended to compare with different approaches.

9) Unfortunately, the discussion is lacking. Please add a Discussion Section.

10)The Conclusion Section is very short. It should be expanded.

11) You should also present your future perspective. How will this study contribute to other future studies? In which works will it be used?

Author Response

Dear Reviewer,

Please find the attachment.

 

Best regards,

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Authors have not properly modified the manuscript. The review comments are not addressed appropriately. For example: No action is taken for the below point at all. 
Section 2 must be literature review (related work). Authors must discuss recent advancements in the field in this section.

Author Response

Dear sir,

 

Please find the attached comments.

 

Best regards

Author Response File: Author Response.docx

Reviewer 2 Report

I suggested 11 changes. 7 of them were carried out by the authors. However, my 4 criticisms still need to be supported in the article. So I repeat the 4 suggestions mentioned below. I expect more precise contributions from the authors.

1)Why was PSO chosen? It should be specified in the article.

2)Why is the proposed approach only compared with the FEM method? It is also  recommended to compare with different approaches.

3)The Discussion Section is too short. The study should be discussed in all its aspects.

4)The Conclusion Section is too short.

Author Response

Dear sir,

 

Please find the attached comments.

 

Best regards

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

The review comments of round 2 are addressed. However, some comments of round 1 is still not properly addressed. Some minor changes are needed.

Author Response

Dear Sir,

 

Please find the attached comments

 

Best

Author Response File: Author Response.docx

Reviewer 2 Report

The article is not sufficient in terms of innovation and contribution to the literature. Two of my suggestions have not been adequately evaluated. I repeat these suggestions:

1)Why was PSO chosen? It should be specified in the article. You mentioned here that PSO is an evolutionary algorithm. PSO is a swarm intelligence based algorithm. New information added about PSO should be checked again. There are many mistakes. Information should be given with reference to the literature. Also your answer doesn't meet  why the PSO question.

2) Why is the proposed approach only compared with the FEM method? It is also recommended to compare with different approaches. This suggestion was ignored. In order to increase the contribution of the study to the literature, you should have taken this suggestion into consideration. Just comparing two algorithms is not an innovative approach. The simulation results are far from showing the success of the proposed method.

Author Response

Dear Sir,

Please find the attached comments.

 

Best

Author Response File: Author Response.docx

Back to TopTop