The Effect of Varying Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System Parameters on Wind Energy Prediction: A Comparative Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Your introduction and descriptive sections are quite good. Nevertheless, you should improve the technical part, i.e., sections 2.1 to 2.3. Your results section lists the values given in tables. You could improve it by better summarising your results and pinpointing the key elements. Finally, the layout of your paper is not optimal for some pages. You will find attached your paper with my question marks and comments.

With my kind regards

Comments for author File: Comments.pdf

Author Response

I thank you for your valuable comments.

• Your introduction and descriptive sections are quite good. Nevertheless, you should improve the technical part, i.e., sections 2.1 to 2.3.

Taking your valuable comments into consideration, the following additions have been made.

Section 2.1. has been improved with information about wind energy and wind turbines. Also, two figures have been added to this section. (Page 4, lines 164-196)

Section 2.2. has been improved by giving a brief explanation of activation functions. (Page 6, lines 220-235)

Section 2.3. has been improved by giving an information of determining the number of membership functions. (Page 7, lines 253-257)

• Your results section lists the values given in tables. You could improve it by better summarising your results and pinpointing the key elements.

Taking your valuable comments into consideration, discussion and conclusion sections have been written. (Page 16, lines 458 – 491)

• Finally, the layout of your paper is not optimal for some pages. You will find attached your paper with my question marks and comments.

Taking your valuable comments into consideration, the corrections determined by you have been made. Additionally, tables 1 and 2 have been updated to improve readability. (Page 2 line 93, Page 5 lines 197, 198, Page 6 lines 208, 212, 213, 215, 217, Page 7 lines 246, 263, 264, 265, Page 10 lines 320-322, Page 11 line 334, 335, Page 12 line 340, Page 13 line 357, Page 13 lines 361-365,  

Reviewer 2 Report

Comments and Suggestions for Authors

This article aims to address the efficiency and accuracy of wind energy prediction using artificial neural networks (ANNs) and adaptive neural fuzzy reasoning systems (ANFIS). The focus of the study is to compare the performance of these two methods in predicting wind energy to determine which model can more accurately predict the power output of wind turbines.

However, there are major flaws in the structure of the article and the methods of the research.

1.      The structure of the article is incomplete, and the discussion and conclusion are missing.

2.      The writing of symbols in the formulas that appear in the article is not standardized.

3.      The headings of Tables 3 and Table 4 are duplicated. Table 4 should be titled " multi hidden layer ANN model prediction values”

4.      Please describe the dataset used for the ANN or ANFIS model in the article. The article provides the average daily wind speed in the Velimese region and the power generated by the wind turbine at different wind speed, but doesn’t explain the relationship between the two values and how the average daily power or power generation of the wind turbine can be obtained from these two sets of data.

5.      This article lacks an introduction to the wind turbine and lacks the main parameters of the wind turbine. Wind turbine power is not only related to wind speed, but also closely related to wind direction.

6.      The two models adapted in this paper are single input and single output, and the data structure is relatively simple, so it is recommended to compare them with the linear fitting results.

 Overall, the article is not complete enough. The depth of research is insufficient, and there is a lack of rational analysis.

Author Response

I thank you for your valuable comments.

1. The structure of the article is incomplete, and the discussion and conclusion are missing.

Taking your valuable comments into consideration, discussion and conclusion sections have been written. (Page 16 lines 458-491)

2. The writing of symbols in the formulas that appear in the article is not standardized.

Taking your valuable comments into consideration, corrections have been made. (Page 5 lines 197, 198, Page 6 lines 208, 215, Page 7 lines 246, 263, 264, 265 )  

3. The headings of Tables 3 and Table 4 are duplicated. Table 4 should be titled " multi hidden layer ANN model prediction values”

Taking your valuable comments into consideration, a correction has been made for Table 4 heading. (Page 11 line 335)

4. Please describe the dataset used for the ANN or ANFIS model in the article. The article provides the average daily wind speed in the Velimese region and the power generated by the wind turbine at different wind speed, but doesn’t explain the relationship between the two values and how the average daily power or power generation of the wind turbine can be obtained from these two sets of data.

Taking your valuable comments into consideration, the section 2.1 has been revised. (Page 4 lines 164-196). In addition, the relationship between wind power and wind speed is specified in Equation 1. (Page 5 line 198)

5. This article lacks an introduction to the wind turbine and lacks the main parameters of the wind turbine. Wind turbine power is not only related to wind speed, but also closely related to wind direction.

Taking your valuable comments into consideration, Section 2.1. has been improved with information about wind energy and wind turbines. Also, two figures have been added to this section. (Page 4 lines 164-196)

6. The two models adapted in this paper are single input and single output, and the data structure is relatively simple, so it is recommended to compare them with the linear fitting results.

Taking your valuable comments into consideration, the comparison has been made, and figure 7 has been added to the study in Section 2.7. (Page 14 lines 380-387)

Reviewer 3 Report

Comments and Suggestions for Authors

The authros studied adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN)  to predict wind energy. I have somm issues.

1.  The abstract should be revised to reflect the paper's contribution and the importance of the study. 

2. The authors need to discuss the impact of the number of hidden layers and neurons in ANN on algorithm performance.

3. The authors need to indicate the dimensions and size of the input data, and whether the data has been preprocessed.

4. The authors need to compare with the latest algorithm to demonstrate its superiority

Comments on the Quality of English Language

The authros studied adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN)  to predict wind energy. I have somm issues.

1.  The abstract should be revised to reflect the paper's contribution and the importance of the study. 

2. The authors need to discuss the impact of the number of hidden layers and neurons in ANN on algorithm performance.

3. The authors need to indicate the dimensions and size of the input data, and whether the data has been preprocessed.

4. The authors need to compare with the latest algorithm to demonstrate its superiority

Author Response

I thank you for your valuable comments.

1. The abstract should be revised to reflect the paper's contribution and the importance of the study.

Taking your valuable comments into consideration, the abstract section has been revised and the following additions have been made. (Page 1 lines 23-26)

2. The authors need to discuss the impact of the number of hidden layers and neurons in ANN on algorithm performance.

Taking your valuable comments into consideration, In the tables containing the results in section 2.6, the difference between ANNs with single and multi-hidden layers was calculated by looking at the average values and added to the study. Similarly, inferences showing the effect of the number of neurons were also made and added to the same paragraph. (Page 13 lines 346-356)

3. The authors need to indicate the dimensions and size of the input data, and whether the data has been preprocessed.

Taking your valuable comments into consideration, the section 2.5 has been revised and a paragraph has been added. (Page 7 lines 278-283)

4. The authors need to compare with the latest algorithm to demonstrate its superiority

Thank you for your suggestions. I agree with your recommendation and would like to emphasize that we will incorporate this comparison in our future work. However, it's important to note that the focus of our study is on Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS). We aim to contribute to the literature on wind energy predictions by diversifying the parameters of these models.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have made a lot of revisions to the article, and the article almost meets the requirements of the journal. There are still a few things to improve:

(1) Is the wind speed used in ANN and ANFIS  models instantaneous or daily average? Is the output parameter of the model instantaneous or daily average generator power? The input and output parameters of the models should be expressed clearly.

(2) Page 4 Line 182： what the meaning of "YERT turbines"?

(3) The equations presented in this paper are not in a standard format. For example, equation (1) the"." Should be replaced by"×".

Author Response

Taking your valuable comments into consideration, the following additions have been made. Any changes made are indicated in green font color.

(1) Is the wind speed used in ANN and ANFIS  models instantaneous or daily average? Is the output parameter of the model instantaneous or daily average generator power? The input and output parameters of the models should be expressed clearly.

In this study, daily average wind speeds were used as input. It is mentioned in lines 18, 19, 159, 267, 269, 277, 281, 389 and heading of Table 1. As the output, the daily average wind power was calculated based on the daily average wind speed. It is mentioned in lines 18, 24, 159, 270, 389, 468, 487.

(2) Page 4 Line 182： what the meaning of "YERT turbines"?

YERT expression has been changed to HAWT expression. Because the abbreviation stated as YERT (Yatay Eksenli Rüzgar Türbünü) is not in English. Page 4, line 188

(3) The equations presented in this paper are not in a standard format. For example, equation (1) the"." Should be replaced by"×".

All equations were reviewed and missing space characters were added. Additionally, the symbol 'x' was added in Eq 1 and Eq 13, and the symbol '}' was added in Eq 12.

Reviewer 3 Report

Comments and Suggestions for Authors

I am not satisfied with the author's revised version

Author Response

Taking your valuable comments into consideration, the following additions have been made. All changes made are marked with a green font color.

(1) Is the wind speed used in ANN and ANFIS  models instantaneous or daily average? Is the output parameter of the model instantaneous or daily average generator power? The input and output parameters of the models should be expressed clearly.

In this study, daily average wind speeds were used as input. It is mentioned in lines 18, 19, 159, 267, 269, 277, 281, 389 and heading of Table 1. As the output, the daily average wind power was calculated based on the daily average wind speed. It is mentioned in lines 18, 24, 159, 270, 389, 468, 487.

(2) Page 4 Line 182： what the meaning of "YERT turbines"?

YERT expression has been changed to HAWT expression. Because the abbreviation stated as YERT (Yatay Eksenli Rüzgar Türbünü) is not in English. Page 4, line 188

(3) The equations presented in this paper are not in a standard format. For example, equation (1) the"." Should be replaced by"×".

All equations were reviewed and missing space characters were added. Additionally, the symbol 'x' was added in Eq 1 and Eq 13, and the symbol '}' was added in Eq 12.