Performance Analysis of Energy Production of Large-Scale Solar Plants Based on Artificial Intelligence (Machine Learning) Technique

Round 1

Reviewer 1 Report

Application of artificial intelligence and machine learning to predict the production of green electric energy is a relevant scientific direction. For this purpose, the authors applied a statistical forecasting method using the ARIMA model. For now, the ARIMA model is the most effective for assessing the performance of a solar power plant and the condition of the solar power plant in the long term. Employing the ARIMA model for similar purposes has given good results in well-known publications:

[1] Atique, S.; Noureen, S.; Roy, V.; Subburaj, V.; Bayne, S.; Macfie, J. Forecasting of Total Daily Solar Energy Generation Using ARIMA: A Case Study. 2019 IEEE 9th Annual Computing and Communication Workshop and Conference (CCWC), 2019. https://doi.org/10.1109/ccwc.2019.8666481.

[2] Alsharif, M.H.; Younes, M.K.; Kim, J. Time Series ARIMA Model for Prediction of Daily and Monthly Average Global Solar Radiation: The Case Study of Seoul, South Korea. Symmetry 2019, 11, 240. https://doi.org/10.3390/sym11020240

In this article, the authors use the ARIMA model to make a 10-year long-term forecast for a large-scale QASP solar power plant in Pakistan. The study is interesting and relevant. It contains new practical results. These results are useful for specialists on development and operation of solar power plants. At the same time, there are some remarks to the article.

1) Lines 102-103 state "Real time data is collected from Quaid-e-Azam Solar Power Park for the year 2018 to year 2019". Lines 100-101 state that "...using these predicted results again through the Machine Learning process by using ARIMA model, 10 years predictions can be generated from 2020 to 2029". Now is the second half of the year 2022. Is this not a current study but the results of studies from previous years?

2) In lines 117 and 118, after “Figure 2” the dots must be removed and the sentence must be continued by a lowercase letter.

3) The inscriptions in Figure 2 are hard to read. The authors should improve the quality of Figure 2.

4) In section 2. Materials and Methods there are subsections 2.2 and 2.3, but there is no subsection 2.1.

5)  In line 142, the sentence must start with an indent.

6) Equations (1)-(5) must be referenced if they are not proposed by the authors. Please check whether all notations in equations (1)-(5) were explained.

7) On page 6, the figure number should be 3, not Figure 1. At the top of page 7, the figure number should be 4, not Figure 2.

8) Put a full stop at the end of Captions under Figures 3-6, 8-9, and 13-15.

9) The inscriptions in figures 4-6 are too small and hard to see.

10) No numerical results are displayed in the Abstract and Conclusions.

11) Figures 4-6 show data for a semi-annual cycle (from October 2018 to April 2019), not an annual cycle.

12) The article does not contain information about the analysis of prediction errors.

13) The reference list is formatted with some mistakes. Check References [3, 5] and correct the list of References with the MDPI and ACS Style.

Overall, the article is good. It can be accepted after minor revision.

Author Response

Dear Reviewer,

Thank you for your recommendations. We have implemented them and I am attaching the answers to all the requested changes.

Best regards,

Authors team

Author Response File: Author Response.docx

Reviewer 2 Report

In the article, the authors describe the application of the ARIMA method to predict the parameters that determine the operation of an existing solar panel farm. I find the article interesting from the point of view of scientific interest. The form and content of the article are correct. Unfortunately, there were some minor errors that I allowed myself to present below. 

·      line 107: Each of the publications to which the authors refer should be described separately in 2-3 sentences. It is difficult to describe the five publications' purpose, methodology and research results in one sentence.

·      Please improve the readability and quality of Figure 2.

·      There is no reference to Table 2 anywhere in the text.

·      lines 144-154: explain the meaning of all symbols and indices used in the formulas (t, β, ɛ, α, ф).

·      line 156: the authors mention the parameter d. It does not appear in any of the previously presented formulas.

·      Lines 163 and 166 and 170 line: Each of the publications to which the authors refer should be described separately in 2-3 sentences. The authors did not support their conclusions with the description

·      line 186. Please renumber the drawing. should be 3. For the conditional block, add the description YES. In addition, it is not shown when it ends. No STOP block. There were also typos, "fitting thee model".

·      Line 198: Please renumber the drawing. Please improve the readability and quality.

·      Line 210and Line 220: Improve the readability and quality of Figure 5 and Figure 6.

·      Line 225: Please write "solar plants" instead of "plants". This remark applies to the rest of the article.

·      Line 231 and line 261, line 291 and line 301 and line 309: the date format must comply with the format shown in Figure 7.

·      Figure 7 and Figure 8: please remove the legends from the graphs. They do not contribute anything.

However, I would like to emphasize that they do not affect the substantive quality of the article.

Author Response

Dear Reviewer,

Thank you for your recommendations. We have implemented them and I am attaching the answers to all the requested changes.

Best regards,

Authors team

Author Response File: Author Response.docx