Solar Power Prediction Modeling Based on Artificial Neural Networks under Partial Shading

Round 1

Reviewer 1 Report

 

 

line 67 - please reformulate the sentence...The concept is correct, but the terminology is a bit mixed up. Indeed...photovoltaic cells are the technology responsible for directly converting sunlight into electricity. Solar panels are the systems that house these photovoltaic cells and capture the sunlight to produce electrical energy.

line 83 - put a dot at the end of the sentence, for consistency.

line 152 - add a dot after "effectively"

line 201 - give more details about Simulink block or at least some references, for example, ref 6 or 8.

line 237 - The whole plots are unclear. I can not see the differences between the fitting and dot lines (Y=T). maybe you chose different colors for it.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

    In this paper, the authors utilize five parameters to improve the accuracy of power prediction using neural networks. The accuracy is improved by repeated training and compared with previous studies to confirm that the accuracy is up to the mark. Overall, this work is interesting and detailed. However, I strongly recommend the authors to revise the manuscript carefully as it still contains many obvious errors. Please note the following information.

 

1. Introduction: There is excessive emphasis on the background information about green energy, so it is recommended to incorporate some recent updates on the developments in solar photovoltaics.

2. It is suggested to move the part about the partial shading effects on line 42 of page 1 before “This study focus…” in line 40.

3. It is recommended that authors supplement the total number of iterations and the end accuracy to further refine the manuscript.

4. The authors make predictions for partial shading. I suggest further supplemental data at different shading ratios.

5. It is recommended to supplement the content related to the iterative approach used.

6. I suggest referencing literature from the past three years to demonstrate the novelty of the method proposed in this paper.

7. On line 58, it is suggested to introduce the reasons for selecting the five parameters used in this study. Why are these five parameters chosen instead of just three or four?

8. The data summary in Figure 3(a)(d) is too strong, it is recommended to add data in the middle part.

 

9. Description of Diode Saturation The current in line 249 is repeated from line 252. Yes, it is recommended to remove some of them.

 

10. Line 303: "When part of the solar shading conditions of the PV system modules in Figure 11" can be modified to "When the PV system operates under partial shade conditions as shown in Figure 11".

作者对这份稿件的描述比较清晰，描述内容通俗易懂，学术性强，但有些地方存在重复描述、用词不当的问题，建议予以纠正。

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Younghyun Lee et. al.  has studied a model that enables accurate power prediction under partial shading conditions using an artificial neural network. The author claim the improvement in  power  generation efficiency using bypass diodes in partial shading. But, I am not sure what new physics and advantage bring by this study to the community as a whole, for example higher accuracy predictions without any discrete quantity to support their claim.

 

I would like the author should address following comment:

 

1.     From line 58-63, I would like the author to clarify their claim with quantitative comparison otherwise not acceptable as its reflect lack of motivation to the scientific commitment. I think author should add their % increase in power production with their model with proper analysis of source of improvement.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Good explanations.

Some repetition occurs between line 248 and line 252.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 5 Report

The paper show a model that enables accurate power prediction under partial shading conditions using an artificial neural network and improves the power generation efficiency using bypass diodes.

In figure 10 is presented the artificial neural network SPICE model of PV module and in figure 11 the MATLAB/Simulink model. Why didn't you use the same software for both models?

To better understand how you made the predictive model of the PV module based on artificial neural networks the SPICE model in figure 10 can be detailed.

For solar power prediction the P–V characteristics curve of a photovoltaic system under partial shading conditions can be presented.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

1.     How is the training data generated? I would like the author to justify the choice of their training data and how much their analysis effect by this choice?

2.     Since the PN junction in photovoltaic is commonly used, I would suggest the author to confirm their claim by using some real data from manufacturer measurement parameter?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf