Study on Improvement of Lightning Damage Detection Model for Wind Turbine Blade
Round 1
Reviewer 1 Report
The paper is focused on the study on Improvement of Lightning Damage Detection Model for Wind Turbine Blade. The damage detection model for wind turbine blade based on SCADA system data has been constructed and machine learning, thereby, considered about the technology that automatically confirms the soundness of wind turbine blade. Finally, anomaly detection accuracy of model is examined to verify the feasibility and reliability of the proposed model in practical application in the paper.
While finding the paper of potential interest, however, there are still some problems need to be clarified. If the authors wish to improve their paper’s contribution, then it requires clear information and analysis as below:
- It's hard to tell the reader from the paper what the new technologies or discover is in this work. The simulation studies for lightning damage detection model for wind turbine blade are well established so the authors need to emphasize why this work is important. What is its advantage?
- The operating settings of assessment result of the model in Figure 8 compared those as shown in Figure 5 are not very clear. Are they performed at the same working settings? They should be also explained and fully understood to verify high accuracy and effective computation of the proposed method.
The authors are recommended to revise their articles in order to enhance the paper’s contribution that can give some useful directions for the other researcher who will do the similar investigations.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper “Study on Improvement of Lightning Damage Detection Model for Wind Turbine Blade” is focused on a very interesting topic which is the SCADA based condition monitoring of wind turbine blades.
The paper is quite well developed but the English language need to be revised especially in the abstract and in the introduction; some sentences need to be rephrased and please check all the errors/typos.
Please at the beginning of the paper includes also a list of symbols, acronyms and abbreviations in order to improve the manuscript readability.
I think the potential of the paper can still be stressed before considering it for publication.
Here are some suggestions:
- Please include a wider discussion as regards the feature selection (have tried other solutions rather than wind and rotor speed?)
- Give more details as regards the minimum database required for training (one month, one year or more?)
- Also the time scale is a fundamental parameters; can your method be used also with the traditional 10 minutes averages?
- Please describe what is really new as regards your reference [15] (in the reference list the year of publication is missing) ….is the present work developed on the same database?
- Which is exactly the turbine technology considered in your study?… have you any updated/recent database to test your model with brand new technology?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Paper has been correctly revised.
