Skillful Seasonal Prediction of Global Onshore Wind Resources in SIDRI-ESS V1.0
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe article discusses the issues of forecasting wind load, on which electricity production at wind power plants depends, using a dynamic forecasting system.
Electricity generation from wind power plants is a virtually inexhaustible generation option. Wind is a source of environmentally friendly energy, but at the same time, wind energy needs a lot of research to develop the scientific and technological foundations for predicting the volume of energy received. The environmental agenda and rising prices for fossil fuels are highlighting the use of renewable energy sources. Wind energy is one of the most attractive renewable energy technologies due to its relatively high efficiency and low pollution levels. However, since the power generated by wind turbines depends on meteorological data and wind speed, unexpected changes in power generation can have a negative impact on the operation of the power system. This circumstance requires high accuracy in forecasting wind speed. Forecasting the wind potential and, as a consequence, the generation of electrical energy at a wind power plant is the first question that practitioners and scientists face when starting to deal with the problems of operating wind farms, both within the framework of an individual project and on the scale of regions, conglomerations and entire countries. Various methods have been developed for wind forecasting, which can be classified according to time scales and methodology. Based on many studies, wind forecasting based on its temporal nature can be divided into three categories: forecasting eight hours ahead (short-term forecasting); day-ahead forecasting (medium-term forecasting); forecasting several days in advance (long-term forecasting). It is also worth noting that wind forecasting schemes can be divided based on their methodology: the physical method, based on the state of the lower atmosphere or numerical weather prediction using weather forecast data; a statistical method based on a large amount of historical data without taking into account meteorological conditions, while artificial intelligence (neural networks) and approaches to time series analysis are used to process “big data”; a hybrid method that combines physical and statistical methods.
The results presented in the work can be used in conducting similar studies.
However, it would be necessary to clarify a number of comments that are available to the article:
1. The introduction could provide a brief comparative description of the use of various renewable sources of electricity compared to wind energy in different regions of the world.
2. The work must provide a brief description of wind power plants used to supply energy to consumers, including in hard-to-reach conditions, and also show the connections between the wind power plant and the power supply system. The following works could be analyzed: https://doi.org/10.1016/j.engappai.2023.107036, https://doi.org/10.1109/URALCON.2019.8877674.
3. How were the most significant factors determined, the values of which are studied during the experimental studies? Was the method of expert assessments or cluster analysis used in this case (section “2.1 Observation-based data”)?
4. Have you considered, along with the two-tailed Student’s t-test, the use of other criteria to determine the statistical significance of correlation coefficients (formulas 1, 2)?
5. A generalized table should be presented in which the results of a comparative analysis of the spatial distribution of the temporal correlation coefficient shown in Figures 1-3 could be presented.
6. It is not entirely clear whether machine learning methods based on the use of artificial intelligence based on neural networks were used in the article. Currently, the use of these methods is the most relevant, since they allow the most accurate forecast of electricity generation with random variation of various factors.
7. Based on the data presented in the polynomial figures, it is necessary to provide specific mathematical models obtained during regression analysis, as well as the corresponding values of the coefficients of determination.
8. It is necessary to present forecast values of electricity generated at wind power plants in different countries of the world in the short, medium and long term based on the experimental studies conducted by the authors.
9. A generalized research methodology should be provided.
10. The conclusions are quite general. It is necessary to add testing of the results obtained at existing facilities and prospects for further research.
Author Response
Dear reviewer,
Thank you for your insightful comment. These comments have greatly improved the quality of our manuscript. We have carefully considered your concerns and have addressed them in the revised manuscript.
Thanks for your time.
Best regards,
Zixiang Yan
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper proposes a skillful seasonal prediction of global onshore wind resources in SIDRI-ESS V1.0, the idea of the paper is good. While I have some suggestions to help improve the quality of the paper.
1) More quantitative comparison with the state-of-the-art should be supplemented in the paper.
2) What about the advantages of the proposed technique compared to the nonlinear predictive control, such as [R1] as follows or artificial intelligence technique for the purpose?
[R1] H. Lin, H. S. -H. Chung, R. Shen and Y. Xiang, "Enhancing Stability of DC Cascaded Systems With CPLs Using MPC Combined With NI and Accounting for Parameter Uncertainties," in IEEE Transactions on Power Electronics, vol. 39, no. 5, pp. 5225-5238, May 2024, doi: 10.1109/TPEL.2024.3359672.
3) How about the stability of the overall system? The stability could be analyzed when the predictive technique is adopted.
Author Response
Dear reviewer,
We sincerely appreciate your time to review our manuscript and provide valuable feedback. The attachment is our response to your comments. Please refer to the attachment.
Thank you for your time.
Sincerely,
Zixiang Yan
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsIn their work, the authors explore seasonal changes in wind resources and their impact on wind energy generation. The authors note the importance of effective seasonal wind resource forecasting for the wind energy industry. Using the dynamic forecast system SIDRI-ESS V1.0, the seasonal forecast of global wind resources was assessed. The highest level of forecast skill for wind speed at 10 meters (ws10m) is shown to occur in several regions where there is already a significant wind energy industry or abundant wind resources. The level of forecasting skill is highest at a lead time of 1 month for most regions and decreases with increasing lead time. The authors also note that using an ensemble forecast can reduce uncertainty and improve forecasting skill. The overall conclusion is that the SIDRI-ESS V1.0 system has high potential to provide valuable seasonal climate forecasts for the wind energy industry.
The work does not cause any fundamental comments. However, the drawings provided by the authors at the end of the work. Very small and will not allow you to judge the study. They need to be recycled. Shades 1-4 are barely visible. The essence of the forecast is clear. But the result is not presented explicitly. I would also recommend including one drawing in the text of the work and making a link to the rest. There are also a number of places in the text where commas are missing (for example, 104, 210, etc.). In general, this does not reduce the quality of work. For this reason, I recommend it for publication, but with some improvements to the drawings!
Author Response
Dear reviewer,
Thank you for your suggestions and recognition of our manuscript. We further improved the figures in the manuscript.
Thanks for your time again.
Best regards,
Zixiang Yan
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have made the necessary changes. I recommend the article for publication.
