Renewable Electricity Management Cloud System for Smart Communities Using Advanced Machine Learning
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper is addressed to specialists in the energy industry. The paper does not have a high degree of scientific novelty but presents a very useful evaluation method for engineers working in the field of energy management. The software part is well explained and is exemplified by code subroutines. The conclusions of the paper are well supported by real data.
Author Response
I appreciate you taking the time to review the paper. Here is the attached file and I have attempted to make corrections with all of the reviews listed.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsAll in all the paper requires extensive changes, the purpose, aim and objective is lost in the middle of all the manuscript with no clear vision and research gap. The paper requires serious revision and more organization. Besides, the results section needs to be re-written as the errors are high.
1. The introduction is too short .. needs a paragraph to illustrate current challenges in the field of energy resources not just a sentence in the introduction. These challenges should be tackled by the recent most updated references.
2. Figure 1 should be Table 1 not a figure.
3. Proper citation of reference is required in Figure 1
4. What about time-based models? Nothing is mentioned in the paper about them
5. Section related work is poorly written and lacks coherence. No findings and solid conclusions could be drawn from this section.
6. Related work does not cover other techniques such as consumption forecast using times series classical time series (Arima for instance) when combined with ML, XGBoost, LightGBM, and CatBoost techniques for instance
7. The authors work in figure 1 shows that model performance has been evaluated using different error metrics such as MAE, RMSE, MAPE. There is no consistency in comparison between papers since the metrices are not unified for all. Short and medium forecast should not be included in the table as its of different time horizon. Besides, where the forecast made on daily, or hourly or minutely basis?
8. Similar comments could be drawn for figure 2 (remarks 3-8)
9. Define in the text what is meant by Y,N in figure 1.
10. Major Players in Solar Generation (figure 4) is irrelevant to the paper.
11. Figures 6-8 must be enhanced instead of just a screen shot, graphs could be demonstrated only. Figure 13 is a snapshot from a platform, data could be wither placed in table or removed and a reference could be added
12. Codes are not required to be demonstrated in the manuscript and should be removed
13. Data Engineering section requires a flow chart to show the framework used
14. What is frequency in figure15?
15. Errors in Table 1 are very high...indicate what level of threshold would determine how well is the model behaving.
16. Table 1 lacks values for R2
Comments for author File: 
Comments.pdf
The paper requires a comprehensive language review, including the revision of all sentences to adopt a third-person perspective.
Author Response
I appreciate you taking the time to review the paper. Here is the attached file and I have attempted to make corrections with all of the reviews listed.
Author Response File: Author Response.docx
Reviewer 3 Report
Comments and Suggestions for AuthorsThe reviewer invites the authors to reply and take these comments into consideration to modify the paper accordingly as follows:
1. The introduction section lacks from several important issues: a) The problem definition should be clearer more, b) the literature review and research gaps did not follow a straight path, it is advised that the authors give a deeper analysis on the ideas introduced in the literature c) contributions/novelty were not highlighted enough. The reviewer needs to see more to justify that this work gives new contributions. The motivation of the study should be further emphasized. In particular, the main contributions of the results in this paper should be clearly demonstrated.
Therefore, the introduction part should be improved and re-organized to cover three parts clearly and sequentially (To be easily understood for the reader) as follows: 1) motivation and incitement, and 2) literature review and research gaps, and 3) contributions and paper organization.
2. Although the paper evaluates the performance of different models, the computational costs and feasibility for real-time applications for different models is not sufficiently discussed.
3. There is limited information on how hyperparameters were tuned for each model, which could significantly impact the results.
4. The paper lacks integration of explainability techniques like SHAP values or feature importance analysis, which can yield profound insights into the model's reasoning.
5. The study lack of Comparative Breakdown on Feature Selection Impact. The authors discussed the impact of feature selection but do not provide a breakdown of how each feature removal impacts accuracy.
6. The study does not compare ML-based models to classical forecasting methods (i.e., ARIMA, Exponential Smoothing), thus missing the opportunity to demonstrate the superiority of ML over traditional techniques.
7. The abstract should be improved to include: 1) problem and motivation 2) the main objective of the study 3) the methodology 4) the main results and major conclusion.
8. The conclusions section lacks numerical results to support the findings and claims made throughout the paper. The absence of quantitative results makes it difficult to assess the actual impact of the different models.
9. The quality of the figures should be improved.
Author Response
I appreciate you taking the time to review the paper. Here is the attached file and I have attempted to make corrections with all of the reviews listed.
Author Response File: Author Response.docx
Reviewer 4 Report
Comments and Suggestions for AuthorsThe manuscript presents a renewable electricity management cloud system model for smart communities in the short term. The proposal is innovative and presents good contributions to research on electricity scarcity. However, many doubts and questions need answers, and many considerations must be made to improve the manuscript's clarity.
1 – Some references are mentioned in the text without formal citation, for example, when the manuscript cites a PRNewswire report. I ask that you check and include the references when this problem occurs.
2 – The manuscript presents 9 sections. However, in lines 70 to 76, where the organization of the paper is described, only 8 sections are listed.
3 – The captions for Tables 1, 2, 3, and 4 were written as figures. This led to an error in the description of the figures below, where Figure 5 is Figure 1, and so on.
4 – Furthermore, regarding Tables 1, 2, 3, and 4, I recommend inserting them as tables in the text and including the references according to the journal's guidelines. The current presentation of the tables appears as images, which is incorrect.
5 – In line 304, there is an error: “kWh/hh”.
6 – Figure 13 should be presented in Table form, the same applies to Figure 14, and Figure 18.
7 – In lines 295, 396, and 397, “W/m2” should be corrected to W/m². Check your manuscript and correct the same error when it occurs.
8 – What software was used to process the models? I recommend including this information in the manuscript, along with the machine specifications and the libraries used.
9 – Was any normalization or standardization performed on the data for processing the LSTM, CNN, and SVR models? Standardizing the data helps avoid scaling issues. If this was not done, I believe this is why these models presented bad MAE results for consumption and generation forecasting in sections 6.1.1, 6.1.2, and 6.1.3.
10 – Figure 19 and Figure 20 need to be enlarged.
11 – What is the unit of measurement for MAE and RMSE in the manuscript and Tables 1 and 3 on page 22? I recommend including this information.
12 – I also suggest including the processing times of the models, as this will allow for a more detailed analysis of the performance of each model.
Author Response
I appreciate you taking the time to review the paper. Here is the attached file and I have attempted to make corrections with all of the reviews listed.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsAuthors have not fully addressed points in the previous report including:
- a more structured research gap discussion
- Consistent use of third-person writing throughout the manuscript.
- Visualization is poor
In addition, the paper lacks coherency and explanation to enhance readability. The codes and snipets are inappropriate for journal publication.
Comments for author File: Comments.pdf
extensive language revision is required
Author Response
I appreciate you taking the time to review the paper. Here is the attached file and I have attempted to make corrections with all of the reviews listed.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsThank you for answering all the questions. I have no more questions.
Author Response
Thank you for taking the time to review the paper. I truly appreciate your feedback.
Round 3
Reviewer 2 Report
Comments and Suggestions for Authorsreviewer has no further comments
Comments for author File: Comments.pdf
minor English revision required
Author Response
Thanks for your time in reviewing our paper.
Minor English revision required: We made the required changes.
