Day-Ahead Wind Power Forecasting in Poland Based on Numerical Weather Prediction

Round 1

Reviewer 1 Report

The manuscript entitled “Day-ahead wind power forecasting in Poland based on numerical weather prediction” deals with a very interesting topic.

 

The manuscript is quite well organized and the motivations of the study are exposed clearly. In my opinion, the point of strength of this work regards the use of the data from the NWP ALARO model, developed by an international consortium. Three years of data (2018-2019-2020) have been used and this increases the value of the present study.

 

Nevertheless, the manuscript in my opinion should be substantially improved and I list here on my remarks.

 

The literature review is not adequate. The authors report a vast series of studies about wind power forecast in Poland and this is valuable in the context of their study, but as regards day-ahead wind power forecast techniques they cite only few review articles. In my opinion, the authors should analyze more in depth the literature about the topic. I list here on some suggested references: the authors are welcome to include them in the paper, as well as other ones at their choice.

 

 

Yang, M., Shi, C., & Liu, H. (2021). Day-ahead wind power forecasting based on the clustering of equivalent power curves. Energy, 218, 119515.

 

Zheng, D., Shi, M., Wang, Y., Eseye, A. T., & Zhang, J. (2017). Day-ahead wind power forecasting using a two-stage hybrid modeling approach based on scada and meteorological information, and evaluating the impact of input-data dependency on forecasting accuracy. Energies, 10(12), 1988.

 

Mana, M., Astolfi, D., Castellani, F., & Meißner, C. (2020). Day-ahead wind power forecast through high-resolution mesoscale model: Local computational fluid dynamics versus artificial neural network downscaling. Journal of Solar Energy Engineering, 142(3).

 

Devi, A. S., Maragatham, G., Boopathi, K., & Rangaraj, A. G. (2020). Hourly day-ahead wind power forecasting with the EEMD-CSO-LSTM-EFG deep learning technique. Soft Computing, 24(16), 12391-12411.

 

Shi, X., Lei, X., Huang, Q., Huang, S., Ren, K., & Hu, Y. (2018). Hourly day-ahead wind power prediction using the hybrid model of variational model decomposition and long short-term memory. Energies, 11(11), 3227.

 

Hong, Y. Y., & Rioflorido, C. L. P. P. (2019). A hybrid deep learning-based neural network for 24-h ahead wind power forecasting. Applied Energy, 250, 530-539.

 

Karakuş, O., Kuruoğlu, E. E., & Altınkaya, M. A. (2017). One-day ahead wind speed/power prediction based on polynomial autoregressive model. IET Renewable Power Generation, 11(11), 1430-1439.

 

Li, L., Yin, X. L., Jia, X. C., & Sobhani, B. (2020). Day ahead powerful probabilistic wind power forecast using combined intelligent structure and fuzzy clustering algorithm. Energy, 192, 116498.

 

Improving the literature review could in my opinion be useful for this study, because in my opinion the main flaw of this study regards the fact that it is not clear what is the scientific added value. The authors employ some data-driven methods for downscaling the forecast, but the rationale of the selection is not clear and is not justified in the context of the scientific literature about the topic.

 

Furthermore, the results section in my opinion is not much elaborative. Substantially, the authors report average error metrics for the forecast for each selected method. The discussion of these results is not elaborated: reading the paper, it is not possible to understand the advantages and drawbacks of each method and therefore the rationale of the methods selection is not justified.

 

The special cases study subsection should be improved, because it is difficult to understand the lesson that the reader should learn from this discussion.

Author Response

Thank you for your valuable comments. Please see attached document. 

Author Response File: Author Response.docx

Reviewer 2 Report

The study “Day-ahead Wind Power Forecasting in Poland Based on Numerical Weather Prediction” is interesting. The paper is well set, and the problem highlighted executed properly. However, attention should be given to the following highlighted points before resubmitting.

 

1. In Abstract what is mean by 26,7%? Please check all the relevant results.
2. What is the overall structure of the paper? Please introduce it in Introduction.
3. The current conclusion is also not acceptable. The authors needed to enhance the conclusion and discuss the findings of this work.
4. The format of the references needs to be checked.
5. Why not use GRNN model? Check the role of such similar techniques in master planning as in https://www.sciencedirect.com/science/article/pii/S0360544218321091and https://www.researchgate.net/publication/321580846_The_General_Regression_Neural_Network_Based_on_the_Fruit_Fly_Optimization_Algorithm_and_the_Data_Inconsistency_Rate_for_Transmission_Line_Icing_Prediction. 
6. How doyou calculate MAPE and RMSE? Please add a description of the corresponding calculation formula.
7. What is mean byRES? Define all abbreviations on their first appearance and then use them simultaneously.
8. What is the main contribution of the paper? Please add it in Introduction.
9. How these forecasting accuracies are computed based on training and testing sets or utilizing the whole data.

Author Response

Thank you for your valuable comments. Please see attached document. 

Author Response File: Author Response.docx

Reviewer 3 Report

This is a very interesting study, and it is worthy of publishing. I enjoyed reading the manuscript. Nevertheless, it needs some further improvements. In general, there are still some occasional grammar errors throughout the manuscript, especially the article "the," "a," and "an" is missing in many places; please make a spellchecking in addition to these minor issues. The reviewer has listed some specific comments that might help the authors further enhance the manuscript's quality.

• I would suggest rewriting the concluding remarks in the abstract more comprehensively rather than stating that ‘’ predicted error has the biggest mean absolute percentage error.’’
• Please justify why you selected the following period of data: 2018-2019?

• Methodology limitations should be mentioned.

• The full names of algorithms should be in capital each, e.g., Artificial Neural Network (ANN). Please make the necessary modifications.
• Figure 3 shows that the map’s coordinates' text size is too small; please increase it a bit.
• Please specify at which height from the ground was the wind speed measured?
• Please elaborate a bit more the Section 4. The discussion should summarize the main finding(s) of the manuscript in the context of the broader scientific literature and address any limitations of the study or results that conflict with other published work.
• Please consider writing a conclusion section, where you can conclude the main contribution of this study.
• Consider adding a list of acronyms.

 

 

Author Response

Thank you for your valuable comments. Please see attached document. 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have addressed my comments. I consider the manuscript adequate for publication.