Surrogate Modeling and Aeroelastic Analysis of a Wind Turbine with Down-Regulation, Power Boosting, and IBC Capabilities

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you to the authors for producing a good quality paper. Overall the topic is interesting and well presented and, with minor edits, should be published.  There are some aspects that I feel could be improved to add to the paper and/or improve the clarity. 

Comments are in order that I have noted them when reading, not in order of importance.

- On line 95 or so, it is written that "from the available information, a level of 5-10% above nominal" is typical for power boosting.  It is then mentioned immediately after that in this work it is assumed that the electrical system can handle a boost of up to 30%.  It should be clarified here as to what level of boosting is actually to be modelled.  The phrasing makes it sound like boosting of 30% will be modelled despite boosting typically being at 10%.

- The IBC approach is used, in whcih individual blades have their own separate controllers.  However, the advantages of IBC, namely that it can target blade loads directly rather than only hub loads, are not used in the work. As IBC is more rare in the literature it seems strange to have used it rather than IPC.  Of course, IBC can be just as effective for hub loads.  Is there a particular reason for choosing IBC over IPC?

- On line 127, the term "environmental contour" is used.  I was not sure what this term meant (I believe it is described later in the paper.  A clarification of the term on this line would make the paper more readable.

- Why is FAST v8 used now that OpenFAST is available and has been for some time?  Would any updates to OpenFAST alter the results?

- On line 199 it is commented that there are no damping loops used for the drivetrain.  This seems odd, as drive train frequencies can reduce the cross over frequency of a controller due to their high gain at high frequency.  Further to this, there is no mention of the cross over frequency, gain margin or phase margin of the controller used in the study.  These values may vary depending upon the operational strategy deployed (e.g. when down regulated or when power boosting).  Was the controller retuned for each operation mode to have similar cross over and/or margins?  If not then there is a question about how the controller could be retuned to perhaps improve performance under different strategies.  Cross over frequency and control design in general can have a large impact on performance so this should really be considered.  Perhaps there was not much difference in these values for the different implementations, but without evidence presented the reader does not know.  As rated power is met at different operating points the gain scheduling may need adjusting for each strategy too, though this is dependent upon the turbine dynamics.  Some comments around these issues and perhaps some quantification of the controller's performance would be useful.

- Similar to the above comment, the different strategies used will have a different impact on the "tightness" of the control of the rotor speed.  It would be interesting to see a comment on how the standard deviation of rotor speed in turbulent conditions varied with different strategies, particularly around the rated wind speed.

- In figure 3 some of the y axis values need clarifying (perhaps a nomenclature would help?)

- On line 294 it is written that "IPC can target mainly loads at the blade...".  This is not entirely accurate.  Whilst IPC can certainly have a positive impact on blade loads, it is the hub overturning and yawing moments that are used and "targetted". 

- On line 325 "previous studies" are mentioned but no reference presented.

- On line 330 a "manual fine tuning" of the controller is discussed, but not done as it is time consuming.  A comment in the paper on how much impact such a fine tuning would have would be useful.

- I assume GPR is Gaussian Process Regression but don't recall seeing the abbreviation explained (may have missed it).  Kriging is used earlier in the paper and, at the time, is not clearly explained.

- Table 3 has a lot of accronyms.  Again, perhaps a nomenclature would be useful?

- As a general point, is there any information that you have come across regarding how OEMs control to reduce power.  Whilst I have directly seen operational data that I cannot share, it may be useful to include a comment about which method you feel OEMs use (if known).

- On line 552 you mention the high loads around rated.  This is not surprising as this is where the controller switching point is. Perhaps add this.

- Second paragraph in section 5 mentions the tower frequencies and that one omethod brings the rotational speed nearer to the tower natural frequencies. What are these frequencies and how close does the controller come to them? 

- On line 664, could the increase in loads through IBC use be eliminated through improved tuning? Can you comment on this?

- In the discussion around lines 824 it would be interesting to have a brief comment on the impact of controller switching may have on the loads. Could different switching methods improve the loads?

- It would be nice to expand your thoughts on how different turbine designs may be impacted.  How much of this study is particular to the DTU 10MW machine?  How much is size a factor? How much is design of components a factor?

- It is great to see that the data and models for the surrogates are available openly online!

I hope these comments are useful.

 

Comments on the Quality of English Language

Good English throughout.  Minor typos that I spotted are given here.

1. Line 345 "allowing to request" should be "allowing requests"

2. Line 436 "smoothening" should be "smoothing"

3. "Smoothens" on line 439 should be "smoothes"

4. Line 787 "for the minimum......." could be rephrased for better clarity

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for sending me the manuscript “Surrogate modeling and aeroelastic analysis of a wind turbine with down-regulation, power boosting, and IBC capabilities” for revision. The Reviewer has the following comments about the manuscript:

 

1. The manuscript describes a complex wind turbine controller design and validation study. It is not entirely clear from the abstract whether the research is part of a larger project or a Ph.D. thesis.

2. It would be helpful if the authors defined the new scientific findings in the introduction and specified which models were adopted from previous studies.

3. The work seems too complex for a general audience. Some essential definitions, constants, and parameters (e.g., those from Fig. 1) were not explained. While the manuscript may be apparent to the authors, it could be challenging for external readers to understand.

4. Due to its complexity, there may be limited practical possibilities to validate the research. The reviewer must rely on the authors to ensure the accuracy of the results.

 

5. The conclusions are clear and straightforward, providing both qualitative and quantitative evaluations of the authors' study.

 

Comments on the Quality of English Language

Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf