Towards Sustainable Energy Grids: A Machine Learning-Based Ensemble Methods Approach for Outages Estimation in Extreme Weather Events
Round 1
Reviewer 1 Report
The topic is interesting but Innovation is not outstanding enough. It is necessary to highlight the innovation to increase the novelty of the article. In its current form, the paper has no novelty, and does not include any significant scientific data. However, there are several critical issues found in the article. Reviewer does believe that by addressing these issues, the article's quality would be much improved.
• Abstract is not written in standard form. Also, the authors are advised to seek the answers to the following questions; (1) what is the objective of this article? (2) Why the authors wrote this article: gape? (3) What is the motivation? (4)How this article contributes to the research article? What are the main findings of the study? (5) How do you conclude your achievement in one sentence? The abstract section should contain more and more results
• The introduction needs to be improved by framing and structuring the subsequence of your novelty findings and results. The authors must present the general research area to unfamiliar readers and, at most to present the current state-of-the-art in order to show the contribution/novelty of their work. Authors must describe/analyze more the current mentioned references and must include many more related references.
• Various review articles are found in the recent literature covering a similar topic. It is suggested to authors to please identify how this study is different from the literature. What exactly are the missing areas this study is covering? What is the gap to show the novelty of the paper?
• Explain why different variants of artificial neural networks were selected in the development of the models in the study
• Does the initial sample size of the prediction method have a great impact on the accuracy of the artificial neural network models? For example, when the sample size increases, will the prediction accuracy of be better
• The authors need to explain that how they assure that their modelling is optimized? The results of performance indexes are enough?
• The section discussion is missing. In order to overcome the readers by the novelty of the paper, the obtained results should be compared to those reported in the literature using other modeling approaches, to our opinions this is highly important. A paper without discussion seems to be confused.
• Conclusions must summarize the work presented within the paper. Also, the future work should be discussed.
• It would be better to cancel paper citation in the conclusion section.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors have developed a machine learning based method for outages estimation in extreme weather events. Although the topic falls under the scope of Sustainability, I have the following concerns which need to be addressed carefully during the preparation of the revised version.
1) Research gaps and motivation of the paper should be clearly cited in the introduction.
2) Comparative study is missing. Results of the comparative study with alteast two existing methods should be depicted.
3) Conclusion section is not up to the mark. Future scopes are limited. So a COMPLETE revision of the conclusion section is required.
4) References section is too short. The following recent references related to "sustainability" can be cited. :
A) An integerated decision support framework using single-valued neutrosophic- MASWIP-COPRAS for sustainability assessment of bioenergy production technologies; Expert System With Applications, Elsevier, 2023.
B) A dual hesitant fuzzy sets based methodology for advantage prioritization of zero-mile delivery solutions for sustainable city logistics. IEEE Transactions on Fuzzy Systems, 2022.
Minor spell check required.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Using a machine learning model can give an effective implementation to predict the running situation of smart grid systems. However there, some confusion exists in this paper.
First, Using a machine learning model always needs to divide your dataset into training sets and test sets. However, the research only provides the accuracy of the experiment. It is difficult for a reader to know whether these models could predict other datasets.
Second, Table 1 demonstrates that the accuracy of ANN is 99.69%. It might be an obvious mistake. It is unclear to say the accuracy is 99.69% when you have 50% false negative and 66.66% false positive.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have addressed all my comments
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
The author revised the paper according to my comments. I recommend it for publication in your reputed journal.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This research provides a novel approach to forecasting the status of energy grids during extreme weather conditions, utilizing an ensemble method that yields impressive results. However, there are some areas of confusion in the study.
To begin with, the research utilized an ensemble method to enhance the performance of machine learning models. Nevertheless, this study fails to demonstrate a positive outcome when applying ANN to the ensemble method. It appears that the ensemble would perform better if the model disregarded the ANN model entirely.
Furthermore, while the research's findings are outstanding, it would be beneficial to include more details about the original classification methods employed by humans rather than relying solely on machine learning algorithms. This would enhance the credibility of the research by providing insights into how energy grids were classified in different training data scenarios. By doing so, the research's overall persuasiveness would be significantly improved.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
