Numerical Simulation of Roughness Effects of Ice Accretion on Wind Turbine Airfoils

Round 1

Reviewer 1 Report

- Avoid using symbols in the abstract; find an alternative way to refer to fine mesh required to capture turbulent boundary layer instead of y_plus<1

- "This work aims to investigate different state-of-the-art rough wall function models to investigate which of these models can provide the most accurate results with the lowest computational resources possible." can be replaced by something such as "This work aims to investigate the performance of state-of-the-art rough wall function models in terms of their accuracy and computational cost."

- Be consistent in using the verbs tenses; usually present tens is used for the abstract and third unknown for the rest of the article.

- It is quite unusual to use terms such as "after that".

- Don't include the future work in the abstract or introduction; this will be confusing.

- There are many studies on the subject of ice accretion on wind turbine blades, which can be found in the literature. Specially, there are research groups that are specialised in modelling the ice formation. The authors should demonstrate awareness of such studies, and that should be reflected in the introduction section.

- Consider enhancing the quality of the figures; example Fig 14

- "4.3 Agreement Analysis" is not an informative title; consider changing this to something such as "Accuracy analysis of the RWFs" or something similar

- use the same format and font for the symbols in the main text and the figures and tables

- you should provide the details of the ice profiles either in form of an appendix or by citing the data. This is needed for any future verification and/or reproduction of the results

- The RWF were developed for simulation of flow around rough surfaces. How the ice roughness is different from those of other studies? Are you trying here to demonstrate the applicability of these models for the particular problem of ice roughness for wind turbines? How the problem of ice roughness on a wind turbine differs, in context of using RWF, from for instance ice roughness on a wing section? More clarity is needed to justify the significance of the work and its contribution.

- Ref 24 is not easy to be found; use an alternative reference or modify the citation.

- The conclusion section should be re-written. Currently, it is more in the form of discussing the results, which has already been done. Outline the main finding of the research, comment on the applicability of the RWFs and provide some recommendations for future work.

 

Author Response

Dear Sir/Madam,

We would like to thank the reviewer for the valuable comments, questions, and suggestions. They helped us significantly. We agree that the manuscript needs improvement and have revised it as described below. All your comments have been addressed.

Author Response File: Author Response.pdf

Reviewer 2 Report

In this work, the authors conducted numerical simulation of flow over iced surface. Instead of using very fine grids near the wall, the authors used RWFs to model the flow behavior and compared the results to both fully resolved roughness simulation and experiments. The overall explanation of the methods is clear and the authors also pointed out some of the challenges in these methods. However, there still remains some key information missing, which the authors need to further explain as listed in the following:

1.       The biggest problem of the current manuscript is that, in the validation part, especially in figure 7, the authors show that the three methods predict Cp very differently for flow over rough cylinder at large θ. Indeed, such big different will result in strong variations of predicted drag force. Although the authors pointed out that good matching is challenging, they should at least provide some guidelines for readers to choose which RWF should be used in different roughness conditions, so that the method proposed in this paper can be beneficial to future scholars. A fitting for the overall trend of Cp over cylinder surface is far from good enough to convince people that RWF is a solid method, since the overall trend is already well known and people need to find out the difference of the curve which is induced by the roughness elements. If the flow over cylinder is a hard case to simulate with RWF, as pointed out by the authors, why not choose some geometries that RWF works better? In that case we can at least know in what condition RWF can be used.

2.       The author should provide more evidence that the grids used in the simulation is enough for the numerical simulation to converge, for both resolved and RWF methods.

3.       Since the authors are using RWF methods to predict the aerodynamics of flow over iced surface, then the validation of how such method can capture the trend of aerodynamic load change across different surface is very important. For example, in figure 7, the data in 1977 shows the clear trend that higher Ks decreases Cp at large θ. Is this trend also predicted by the RWF method?

Author Response

Dear Sir/Madam,

We would like to thank the reviewer for the valuable comments, questions, and suggestions. They helped us significantly. We agree that the manuscript needs improvement and have revised it as described below. All your comments have been addressed.

Author Response File: Author Response.pdf

Reviewer 3 Report

Please, see the attached review.

Comments for author File: Comments.pdf

Author Response

Dear Sir/Madam,

We would like to thank the reviewer for the valuable comments, questions, and suggestions. They helped us significantly. We agree that the manuscript needs improvement and have revised it as described below. All your comments have been addressed.

Author Response File: Author Response.pdf

Reviewer 4 Report

The present paper investigates the accuracy of different rough wall function models. The manuscript is certainly interesting for providing guidance over the model. However, the manuscript is more like a technical report rather than a paper. The reviewer suggests that the manuscript needs to be deeply revised before publishing.

1.  The abstract need to be rewritten. Some sentences are hard to read.

2.  There is no study on the computational cost of three RWF models in the Results part as mentioned in the Abstract. 

3. For the case design, it is better only to change one parameter instead of changing multiple parameters.

4. Although, the authors claim that the model, to a certain extent, can provide agreement with the experimental results. The reviewer thinks the current RWF model cannot describe the general physics of the flow passing by the iced flow, especially near the stall AOA. More persuasive evidence should be provided. 

Author Response

Dear Sir/Madam,

We would like to thank the reviewer for the valuable comments, questions, and suggestions. They helped us significantly. We agree that the manuscript needs improvement and have revised it as described below. All your comments have been addressed.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The paper can be accepted in its current form with some minor spell check, grammar check and writing style required

Reviewer 2 Report

In general, the authors addressed my comments and I can agree for the publication of the manuscript.