Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, the authors present the results of a simulation study focusing on a method for predicting the mechanical performance of 110 kV transmission tower components under icing conditions. The proposed method is based on finite element simulation results, which serve as the basis of a learning dataset for five machine learning techniques. The accuracy of the prediction method has been verified through an actual tower collapse case.

The article is well written and I do not have many comments. The methods are well described, the authors have correctly presented the methodology of the study, especially with regard to the principles of parameter selection of the methods used. In the light of the above observations, I would like to make some general comments and one specific comment on the formatting of the manuscript.

General comments:

1.  It would be beneficial for the authors to consider a more transparent presentation of the summary and evaluation of the prediction results for the ML methods under study, potentially utilising graphical elements such as charts. So that the choice of the best method is clear and understandable.  

2.   A key omission from the summary is a clear articulation of the intended purpose for which the prediction is to be deployed and a comparison with alternative methodologies that are currently in use. In the event that such studies have not been conducted previously, it may be due to a specific reason (e.g. the prediction of these parameters is not applicable in the actual operation of electricity poles). In essence, none of the methodologies employed in the study can be considered a novel contribution. The methodologies employed are well-established and have been utilised in previous studies. It is my understanding that the value of this study lies in its combination of methods for a specific purpose. Furthermore, it would be beneficial to include a discussion of the rationale behind this approach and its contribution to the field of diagnostics or monitoring of structures. It would be beneficial to include suggestions for future research in the conclusions section. 

3. The authors should expand the technical description of the FEM model of the pole, cables and wind with information such as type and number of elements, number of degrees of freedom, type of connections between the elements pole-wires, wind wire, etc.

Specific comments:

1.      Error in reference to literature sources from item 31 (line number 200 onwards)

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks Authors. I have a few questions about the paper. FEM simulating model: any pictures the authors can use in the paper? If there is, use it in the paper.

(i.e., body deflection or member stress and so on).

What software was used? In the FEM modeling, how did the authors apply Wind Load on the structure?

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper addresses the effect of icing on transmission lines as potential cause of failure of four weak components of transmission towers.  Using data generated by a series of finite element analyses, five machine learning approaches were trained and tested to compare their ability to predict transmission tower component distress.  Before the paper can be accepted for publication, the authors need to address the following points.

1.      Is there only one specific geometry that all 110kV transmission towers have?  None of the seven parameters shown in Table 2 involve geometry variations.

2.      What member sections and member sizes were used in the finite element analyses?  Were the joints modeled as pinned or rigid?  Was the analysis elastic or inelastic?  How were the loads applied on the towers?

3.      On line 100, the authors indicated that 1500 sample data sets were constructed, but on line 170, the number changed to 500.  Where did the remaining 1000 data sets come from?

4.      More information needs to be provided in Section 2.7.  The three plots (Figures 5 – 7) shown are not sufficient to “validate” the finite element modeling and analyses unless the data on these plots are directly compared to those obtained by the other researchers.

5.      Explain why these particular five machine learning (ML) algorithms were selected for comparison.  And were all ML computations performed on commercial available software. If yes, what are they?  If the ML algorithms were developed in-house by the authors, some verification studies need to be presented.

6.      Throughout the paper, there are multiple occurrences of Error! Reference source not found.  This needs to be fixed!

7.   Table 10 needs to be redone or reformatted. The headings and numbers are all split into two or more rows!  This makes them very difficult to read.  The authors should consider switching the rows and columns and reducing the font size so the words and numbers will not get split randomly by the computer.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors responded the comments appropriately thus I recommend the paper be published in the journal. 

Comments on the Quality of English Language

 Minor editing of English language required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have addressed most of the comments from this reviewer, except one.  It is still not clear what member sizes were used in the finite element model.  A table of member sizes and where they were used on the transmission tower under investigation should be provided.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf