the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
On the influence of cross-sectional deformations on the aerodynamic performance of wind turbine rotor blades
Abstract. The aerodynamic performance of a wind turbine rotor blade depends on the geometry of the used airfoils. The airfoil shape can be affected by elastic deformations of the blade during operation due to structural loads. This paper provides an initial estimation of the extent to which cross-sectional deformations influence the aerodynamic load distribution along the rotor blade. The IEA 15 MW reference wind turbine model is used for this study. A constant wind field at the rated wind speed is applied as a test case. The resulting loads are calculated by an aero-servo-elastic simulation of the turbine. The loads are applied to a 3D finite element (FE) model of the rotor blade, which serves to calculate the cross-sectional deformations. For the individual cross-sections in the deformed configuration, the new lift and drag coefficients are calculated. These are then included in the aero-servo-elastic simulation and the obtained results are compared with those of the initial simulation that is based on the undeformed cross-sections. The cross-sectional deformations consist of a change in the chord length and the geometry of the trailing edge panels and depend largely on the azimuth position of the blade. The change in the airfoil geometries results in altered aerodynamic characteristics and therefore in a deviation of the blade root bending moments, the maximum change of which is -1.4 % in the in-plane direction and +0.71 % in the out-of-plane direction. Although these values are relatively small, the initial results imply that further investigations should be carried out with more complex wind fields and different rotor blade designs to identify aero-structural couplings that may be critical for the design of rotor blades or other wind turbine components.
- Preprint
(1527 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed (peer review stopped)
-
CC1: 'Comment on wes-2024-91', Baoxuan Wang, 05 Aug 2024
Dear all,
I took a quick look at this article, and I think it is a very interesting and valuable work.
Also, I have some comments for your consideration:
1) Line 123: In my understanding, "area moments of inertia" is not equal to "stiffness", is it right? Similar question is also occurred in the following content, such as Line 116: "the second moments of inertia"
2) Line 160, "Since the ElastoDyn module was used in the OpenFAST simulation, torsion was not accounted for", the statement is not right, Please see: Wang B.X., et al. 3D multiscale dynamic analysis of offshore wind turbine blade under fully coupled loads, Renew. Energy 223 (2024) 119985.
By the way, the cross-sectional deformations in the rotor blades are considered in the mentioned article (https://doi.org/10.1016/j.renene.2024.119985), by considering the change of nodal coordinates at each time step.
3) Line 163, "multi point constraints" can be done in many different ways in the software, which one were you using?
Aditionally, "The magnitude of the forces were calibrated so that the flapwise and edgewise bending moments from the loads simulations were well approximated", How to achieve it? Please clarified that in detail. Line 235, Are the bending moments are transferred to the shear forces? How to do this? It is not unimportant for readers.
4) The details need to be clarified all over the manuscript, including the detailed differences between the original model and your FEA model based on so-called MOCA, how about the eigen-frequencies?
5) For most of readers, they have few understanding about the MOCA. When it comes to that MOCA has been verified, I may think that it was only verified by small blade. Is there any limitation need to be mentioned?
6) 3D FEA results are expected to occur in the analysis for discussion.
Best regards,
Baoxuan Wang
Zhejiang University
wangbaoxuan@zju.edu.cn
Disclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-91-CC1 -
AC1: 'Reply on CC1', Julia Gebauer, 12 Aug 2024
Dear Baoxuan Wang,
Thank you for your interest and comments. In the name of all authors, I would like to respond to the discussion points you raised below.
- Line 123: In my understanding, "area moments of inertia" is not equal to "stiffness", is it right? Similar question is also occurred in the following content, such as Line 116: "the second moments of inertia"
Yes, true, the second moment of inertia (or area moment of inertia) is a purely geometric magnitude. Stiffness is the combination with material properties, e.g., the Young’s modulus. We will take care to thoroughly distinguish these terms in a revised version of the manuscript.
- Line 160, "Since the ElastoDyn module was used in the OpenFAST simulation, torsion was not accounted for", the statement is not right, Please see: Wang B.X., et al. 3D multiscale dynamic analysis of offshore wind turbine blade under fully coupled loads, Renew. Energy 223 (2024) 119985
As described in the OpenFAST Documentation, ElastoDyn uses only the 1st + 2nd flapwise and 1st edgewise modes to describe the beam behavior of the blades. There is no degree of freedom for blade torsion. Torsion is thus not considered in the kinematics of the blade, and can thus not be accounted for in the deformation. The torsional moments are consequently also not reliable, as they result purely from equilibrium, not from kinematics.
- By the way, the cross-sectional deformations in the rotor blades are considered in the mentioned article (https://doi.org/10.1016/j.renene.2024.119985), by considering the change of nodal coordinates at each time step.
Many thanks for letting us know. When I understand it correctly, you used the term “nodal coordinate” to describe the blade sensor positions along the blade span. Further, you applied the loads from the time series to generate a stress time series with FE. However, I did not find an analysis of cross-sectional deformations (change in airfoil geometry), which is the core of our manuscript.
- Line 163, "multi point constraints" can be done in many different ways in the software, which one were you using?
We used the contact element CONTA173 (3D 4-node surface-to-surface contact), with an extra node in the shear center as main node, and all cross-section nodes as helper nodes. The rigid option was used. We will add this information in a revised version of the manuscript.
- Additionally, "The magnitude of the forces were calibrated so that the flapwise and edgewise bending moments from the loads simulations were well approximated", How to achieve it? Please clarified that in detail. Line 235, Are the bending moments are transferred to the shear forces? How to do this? It is not unimportant for readers.
The flapwise and edgewise bending moment were approximated by applying concentrated forces as described in line 157 to 160. For more details see: https://windeurope.org/summit2016/conference/submit-an-abstract/pdf/35620222998.pdf
- The details need to be clarified all over the manuscript, including the detailed differences between the original model and your FEA model based on so-called MOCA, how about the eigen-frequencies?
As stated in section 2.1 the MoCA model showed differences in geometry and stiffness. Hence, from then on only the MoCA model was used.
Of course, we also checked the natural frequencies. We get frequencies of 0.533 Hz (1st flap) and 0.631 Hz (1st edge) with the MoCA blade model. Compared to the given values in the technical report of the IEA 15 MW RWT, the deviations are less than 5% and are thus considered negligible.
- For most of readers, they have few understanding about the MOCA. When it comes to that MoCA has been verified, I may think that it was only verified by small blade. Is there any limitation need to be mentioned?
Shell models are state of the art. MoCA is an in-house software with a parameterization that is very similar to WindIO. However, it is one possible implementation of how to create a shell model. To the opinion of the authors, it is not necessary to explain all details of the modeling strategy, as these are documented elsewhere (see https://wes.copernicus.org/articles/7/105/2022/). The reference is already included in the manuscript.
When it comes to validation, you are right. We validated the software and modeling strategy integrated in MoCA on a small blade up to now, as that was the only real blade where we had access to the detailed design information (geometry, layup, test setup, etc.). We are currently waiting for approval of a project where we will have access to the design data of a modern utility-scale wind turbine blade, and then we will see how MoCA performs with that one. As already mentioned, shell models are state of the art. It is known in science and industry that the bending behavior can be well approximated, whereas there may be issues with torsion. This limitation also holds for a MoCA model, which is solved in Ansys, so when it comes to FE implementation, it is an Ansys model. Industry is aware of the limitations in torsion.
- 3D FEA results are expected to occur in the analysis for discussion.
Can you please specify what FEA results you would like to see? The manuscript is about cross-sectional deformations, and these are presented in the paper. In the opinion of the authors, there is no need to present analysis results that are actually not needed for the core of the paper.
Best regards,
Julia Gebauer
Citation: https://doi.org/10.5194/wes-2024-91-AC1 -
CC2: 'Reply on AC1', Baoxuan Wang, 12 Aug 2024
Dear Julia Gebauer,
Thank you for your patient reply. Here are some points that I want to clarify, just for your reference only.
2. Although the torsion deformation cannot been considered in ElastoDyn, please note that the aerodynamic torque are considered. The torsional moments from kinematics seem does not seem to play a dominant role, from the perspective of magnitude.
3. In fact, as you said, analysis of cross-sectional deformations was indeed not conducted. I mentioned this only because your interesting research reminded me of it. In the mentioned work, the distributed nodal forces are calculated based on the time-varying nodal coordinates of each section, so the change or deformation in airfoil geometry was considered to some extent.
4. It seems better to illustrate with pictures.
5. To be more specific, has the lever arm change caused by large deformation been taken into account?
8. Sorry, the previous comments may have been misleading. Does the response of the 3D model match that of the 1D model? For reference only.
Thank you for your response and the effort you have put in for this, which is commendable.
Best regards,
Baoxuan Wang
Disclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-91-CC2 -
AC2: 'Reply on CC2', Julia Gebauer, 29 Aug 2024
Dear Baoxuan Wang,
Thank you for your feedback. We consider your comments for a revised version of our manuscript.
Best regards,
Julia Gebauer
Citation: https://doi.org/10.5194/wes-2024-91-AC2
-
AC2: 'Reply on CC2', Julia Gebauer, 29 Aug 2024
-
AC1: 'Reply on CC1', Julia Gebauer, 12 Aug 2024
-
CC3: 'Comment on wes-2024-91', Matthias Saathoff, 23 Aug 2024
Thank you for your insightful manuscript and the concise presentation of your approach and the results!
I would like to kindly bring to your attention a previously published paper that you may not have had the chance to read. The study concludes that with knowledge of blade distortions, rectified polars, twist, and sweep can be considered in turbine design to decrease possible gaps between design assumptions and field observations, particularly in terms of power performance and loads, cf. https://iopscience.iop.org/article/10.1088/1742-6596/1618/5/052011
Kindly reconsider your conclusion that the analysis of "aero-elastic simulations of a wind turbine under consideration of cross-sectional deformations in the rotor blades ... in has not been
carried out before by other groups".Many thanks,
Matthias Saathoff
Disclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-91-CC3 -
AC3: 'Reply on CC3', Julia Gebauer, 29 Aug 2024
Dear Matthias Saathoff,
Thank you for your comment on our manuscript. We are really happy to see your interest in our manuscript and appreciate the interaction. I would like to answer in the following on behalf of all authors.
In your paper, you address the change in blade geometry due to cool-down after manufacturing, the resulting deviation in lift and drag coefficients to the undistorted model, and the impact on aero-elastic loads during turbine simulations. I agree that we have overlaps when it comes to the impact of geometry changes on the aero-elastic performance. However, when we talk about cross-sectional deformations in our manuscript, we address the change in geometry due to mechanical loading during turbine operation and the corresponding deformation of the blade’s cross-sections. This is a dynamic process and was, to the best of our knowledge, not considered yet in literature. In this context, we also concentrated on the modeling of rotor blades and their deformations in the literature review, since it is the baseline for aero-servo-elastic turbine simulations.
However, we do see that changes in cross-sectional geometry cover a wide range of possible causes (as described, for example, in your paper). So, we consider adding a paragraph on this in a revised version of our manuscript.
Best regards,
Julia Gebauer
Citation: https://doi.org/10.5194/wes-2024-91-AC3
-
AC3: 'Reply on CC3', Julia Gebauer, 29 Aug 2024
-
RC1: 'Comment on wes-2024-91', Anonymous Referee #1, 30 Aug 2024
General commentsThis manuscript presents an analysis of cross-sectional deformations in wind turbine blades under operational conditions and their impact on aerodynamic loads. The added value of this work is to quantify the aero-structural couplings where the structural cross-sectional deformations impact the airfoil shape, which in turn affects the aerodynamic loads. This field of study is relevant for the wind energy community due to the increase in blade flexibility. Furthermore, quantifying couplings is particularly important for multi-disciplinary design optimization of wind turbines.The submitted manuscript presents weaknesses regarding (i) its contextualization with the literature, (ii) the strength of its methodology and (iii) its clarity and structure.The literature review presents relevant and related references. However, it does not provide sufficient argumentation to motivate the work, i.e. justify why one could expect cross-sectional deformations to have a significant impact on aerodynamic loads. The review can be improved in several ways described in the specific comments below, in order to contextualize the study better and highlight its scientific relevance.The methodology used in the article consists of a sequential process. Aerodynamic loads calculated from an undeformed turbine are applied to a finite element model of the blade to obtain a deformed configuration. This deformed configuration is in turn used to update the aerodynamic characteristic of the airfoil and enables the calculation of the aerodynamic loads for the deformed turbine. This analysis is applied to a single load case, with constant wind and at rated wind speed. Furthermore, the analysis is done for one turbine, the 15MW reference wind turbine. Finally, the torsional deformation of the blade is not taken into account in the aero-elastic simulations. As pointed out by the authors in the manuscript, this methodology has weaknesses and may justify the lack of significant results. This weakens significantly the relevance of the study. Consequently, the work would be significantly improved by either (i) providing a thorough justification for the chosen methodology, (ii) applying the analysis to other wind turbine designs (e.g. 10MW and 22MW reference turbines), or (iii) improving the methodology to obtain positive results.Finally, several parts of the manuscript lack structure, clarity and conciseness. Several suggestions to improve this aspect are described in the specific comments below.Specific comments
1. Literature review and contextualization.
- The literature review (l. 38-78) reads as a list of disconnected works, where the scientific contributions are not stated. Consider highlighting the problems that each cited work is addressing, and describing the associated results. Please highlight the similarities and differences between the cited work and the problem addressed in the manuscript. For the paragraph between l.38-56, linking the studies to the topic of blade cross-section deformation is critical to put the presented work in context.
- l.58: "Deformation of rotor blades ...": This paragraph is unclear. The first sentence mentions rotor blades and blade design, but it is not obvious how the cited works relate. Please make the link between the cited literature and the problem addressed by the manuscript more explicit.
- l. 65-70: "Preliminary work on [...] and the deformed cross-sectional shape was obtained." This part of the literature review is critical for the motivation of the presented study. The authors cite two of their previous works that have quantified cross-sectional deformations in wind turbine blades. However, the interpretation of these works is weak. The authors only state the method used in these works, and not the results of the analysis. What is missing in the reasoning is a quantification of such deformations. In other words, were there enough deformations obtained in the cited works to suggest an impact on aerodynamic properties and justify the present work?
- Consider adding literature on the topic of multi-disciplinary analysis and optimization (MDAO), and aero-structural couplings (e.g. bend-twist coupling). The wind turbine design research community already includes such couplings in state-of-the-art design optimization frameworks. It would be relevant to highlight to which extent the state-of-the-art in MDAO takes into account cross-sectional deformations. See for example the following two references:
- Mangano, M., He, S., Liao, Y., Caprace, D. G., & Martins, J. R. (2022). Towards passive aeroelastic tailoring of large wind turbines using high-fidelity multidisciplinary design optimization. In AIAA SCITECH 2022 Forum (p. 1289).
- Bortolotti, P., Bottasso, C. L., Croce, A., & Sartori, L. (2019). Integration of multiple passive load mitigation technologies by automated design optimization—The case study of a medium-size onshore wind turbine. Wind Energy, 22(1), 65–79. https://doi.org/10.1002/we.2270
- Consider adding literature on the topic of cross-sectional deformation in the context of failure analysis, including experimental works. This can help contextualize your results and quantify the deformations expected for operational loads vs ultimate loads.
- Consider making it more explicit that the study focuses on cross-sectional deformations due to operational loads and not ultimate or failure loads.
2. The description of the problem, research objectives and research questions is not clear in the text. As such, the text does not convey efficiently the relevance of the work.- l. 36-37 "When optimising the design with the minimisation of the blade mass as an optimisation target, an increasingly elastic behaviour of the blade is expected, including the elasticity and flexibility of the cross-sections": This statement introduces the problem addressed by the manuscript. Consider describing why an increase in cross-section elasticity and flexibility is a problem.
- l. 72: "Hence, the main objective of this paper is to study the influence of cross-sectional deformations in the rotor blade on the aero-elastic behaviour of the wind turbine.". Consider making the research question of the study explicit. For example: "To what extent do cross-sectional deformations impact the aerodynamic performance of the rotor, under operational conditions?"
3. Choice of methodology- l.73 "A simple test case with respect to normal operation of the wind turbine and a constant wind field is selected for the initial investigation presented in this paper.": Please add a motivation for this choice of analysis.
- l. 84 "A two-stage process is applied": please justify this choice of analysis. The state-of-the-art in MDAO does fully-coupled aero-structural analysis. Why has a sequential approach been chosen here and not a fully coupled one.
- l. 160-161 "torsion was not accounted for": this is a strong assumption for the study. Please justify why this modeling assumption was chosen, and its impact on the results of the study. There are several state-of-the-art aero-elastic tools that implement torsional degree of freedom. This is particularly critical in the light of the following statement: l. 405-406 "It is expected that especially torsion can result in significant in-plane cross-sectional deformations". If that is the case, why not include torsion in the first place in the study?
- l.163 "The loads were applied via multi point constraints that represented a load introduction similar to load frames". Please justify this model choice and its impact on the results of the study. Why simulate a blade-testing environment and not an operational environment, i.e. loads continuously distributed along the blade? This is particularly important considering the results of Figure 11. It seems that the high deviation observed at the location r/R = 0.8 is due to the presence of a load frame.
- l.188 "High-fidelity computational fluid dynamics (CFD) was used to compare the polars and ensure the validity of XFOIL results." Why use XFOIL for the generation of airfoil polars and not CFD directly?
4. Concise description of the method. Section "2. Methods" starting at l.80 presents the methodology of the work. The different steps of the analysis are reported: (i) Generation of loads using aero-structural simulations for the undeformed turbine, (ii) Generation of cross-sectional deformations from the extracted loads using a FE model of the blade and (iii) Generation of loads for the deformed configuration. However, the text lacks structure and conciseness.- The text describes in length adjacent aspects of the analysis that are not central to the study: creation of a FE blade model from the reference turbine data (l.85-91, l.109-111), verification of the FE blade model (l.112-135), verification of the airfoil polars between XFOIL and CFD (l. 190-215), choice of the number and placement of load frames (l.170-173). Consider moving these parts of the text into appendices.
- Consider restructuring section 2. Due to the fact that several models are used twice in the analysis, I would recommend first describing the steps of analysis with the support of Fig.1 and then including one section for each model used. For example: "2.1. FE structural model", "2.2. Generation of airfoil polars", "2.3. Aero-elastic simulations". Then, each section can describe when the associated model is used in the global process. For example: "Aero-elastic simulations are conducted during steps 1 and 2 as shown in Fig. 1, using the undeformed and deformed configuration of the turbine, respectively."
- Having a flowchart like Fig. 1 to illustrate the different steps of the analysis is great to support the description of the methodology. However, there is a discrepancy between the terms used in the figure and the terms used in the text, which can be confusing for the reader. To increase the relevance of the figure, consider aligning such terms and referring to the headers of section 2 in the figure.
- Some steps of the analysis are repeated excessively in the manuscript. For example, the use of OpenFast for aero-elastic simulations is mentioned in l.94-95 ("an aero-servo-elastic simulation of the turbine was performed using OpenFAST"), l.157 ("an aero-servo-elastic turbine simulation with OpenFAST"), l.228 ("The aero-servo-elastic simulations were performed with OpenFAST"), l.256 ("Aero-servo-elastic simulations with a duration of 600 s were carried out in OpenFAST"). Consider making the description of the analysis more concise and referring to section 2. when needed.
- l.247-254 "Recall the blade element theory... have an impact on the lift and drag forces.": The equations and notations introduced in this paragraph are not used in the rest of the manuscript. In order to improve the conciseness of the text, consider removing this part or moving it to a part of the manuscript used to motivate the work.
4. The results of the study are described in sections "3. Cross-sectional deformations" and "4. Coupling effect on the turbine behaviour". The documentation of the results is thorough but lacks interpretation and conciseness.- l.246 "the cross-sectional deformations at two positions along the blade are presented and discussed" and l. 319 "The cross-sectional deformations were determined for all cross-sections along the blade. However, due to space limitations, they cannot be discussed in detail here.". The distribution of the cross-sectional deformation would be relevant for the scientific community and for reproducibility of the results. Consider shortening section 3.1. and adding one figure and one paragraph describing the deformations across the blade.
- Consider combining Figures 7 to 10 to help the reader see the differences between each case.
- In section 3.1, the authors describe at length Figures 7 to 10. However, an interpretation of the results is missing. For example, l.292-294 "From the leading edge up to approximately c/8, the shell on the pressure side deforms into the cross-section. From there on up to approximately 5c/8, the pressure side shell deforms out of the cross-section.": This is a visual description of the results. Please provide some interpretation: can the deformation of the cross-section be justified by the type of loads or the stiffness of the components? Are these results similar to deformations described in the literature? Furthermore, consider making the description of these results more concise and to the point. This comment also applies to sections 3.2 and 3.3
- l.358-359 "Two regions along the blade span can be identified that show a change in aerodynamic properties. The first region is at around 20 % of the span, the second at around 80 % of the blade span.": This is an expected result considering the results shown in Figure 11. Please make the causality between the results of section 3.2 and section 3.3 clearer, i.e. the deformation shown in Figure 11 results in the change of aerodynamic characteristics in Figure 12.
- Figures 13 and 14 are particularly clear and efficient in representing the data.
- Consider linking the results of section 4 to the results of section 3 in a more explicit manner. Do the differences in lift and drag reported in section 3.2 justify the results of section 4?
5. Discussion of the results- l. 394 "This study revealed small changes ...": It could be relevant to have a separate section for the discussion of the results
- Consider contextualizing the results of section 4. Would the cross-sectional deformation have a significant impact on power production or other relevant metrics? What would be the impact for wind turbine design?
6. Conclusion of the work- l.14: "the initial results imply that further investigations should be carried out with more complex wind fields and different rotor blade designs" This statement needs to be backed up by either data or literature. Placing it in the abstract implies that the article supports this claim, which is not the case in this version of the manuscript. Please soften the language here, for example: "While these values are small, further investigations with ... could identify a stronger aero-structural coupling".
- l. 426-427 "The change in chord length was identified in the vicinity of 20 % and 80 % of the blade span.": Please put these results in context with the type of load applied, i.e. with discrete load frames and not a continuous loading.
- l. 432-439: "These relative deviations are quite small. However, ... We therefore expect the influence on the aero-elastic simulation to be greater when analysing extreme load cases.": This paragraph significantly weakens the results of the study. In essence, the author states that the relative deviations reported in the study are small because simple assumptions were taken, which negates the relevance of the work. This is done to justify the "negative results" obtained. However, it raises the question of why such assumptions were used. An alternative conclusion to the work would be to suggest that the aero-structural coupling between cross-sectional deformation and aerodynamic loads is negligible for operational loads. This would significantly increase the impact of the work and its relevance for the scientific community.
7. Description of the methodology- l. 6 "a 3D finite element (FE) model": please add details of the model in the abstract, for example stating the type of elements used.
- l. 93 "... and were used to calculate the aerodynamic loads via the blade element momentum theory (Jonkman et al., 2015). For the load calculation, an aero-servo-elastic simulation of the turbine was performed using OpenFAST (National Renewable Energy Laboratory, 2023).": Please clarify whether the "aerodynamic loads" and "loads" mentioned in the two sentences are calculated with different tools, and if loads due to gravity are included in the analysis.
Minor comments- For the cited work "Gebauer, J. and Balzani, C.: Cross-Sectional Deformation of Wind Turbine Rotor Blades, in: The 33rd International Ocean and Polar Engineering Conference, Ottawa, Canada, 2023.", the full paper (if there is any) couldn't be found from the information provided. Only a one-pager was found, without mention of relevant results (https://publications.isope.org/proceedings/ISOPE/ISOPE%202023/data/pdfs/160-2023-TPC-0416.pdf). Please provide the reference to the full paper or remove the reference.
- Consider adding an illustration of the cross-section topology in the manuscript.
- The term "MoCA model" is used repeatedly in the manuscript. The qualifier MoCA refers to a specific modeling tool, and would not be meaningful for readers not familiar with it. Consider using another term, such as "the 3D blade geometry model".
- The term "OpenFAST simulations" is used repeatedly in the manuscript. Consider using the term "aero-elastic simulations" instead to highlight the type of analysis conducted, instead of the tool used.
- The authors use the past tense repeatedly in the manuscript. Please make sure to use past tense when describing your numerical experiments, and use present tense when describing your results. For example, l.326 "The relative deviation was plotted against the normalised spanwise coordinate": instead of using "was plotted", consider using "is shown" instead.
- l. 11. "depend largely": the qualifier "largely" does not add extra meaning here. Consider rephrasing to keep the abstract concise and to the point.
- l. 26 "which is calibrated by the aerodynamic twist angle". Please consider replacing the term "is calibrated by" with "depends on" in order to increase the precision of the text.
- l. 31 "In the context of this paper, the structural composition of the blade defines its resistance against cross-sectional deformations": please clarify or rephrase. This sentence is unclear.
- l. 35 "structural loads increase." Consider aligning the term used here with the load types enumerated in l. 19-20. This will help the reader understand exactly what type of load is meant by "structural loads".
- l. 41 "The study further focused on bifurcation analysis, which is not the subject of this paper.": Consider removing this sentence.
- l. 59 "energy-based method for thin-walled ...": for modelling? analysing? designing? Consider adding a verb in this sentence to characterize the method.
- l. 66 "a three-dimensional finite element model": please use "3D" or "three-dimensional" consistently throughout the manuscript.
- l. 236 "The simulation was carried out until the rotor blade behaviour became periodic." and l. 256 "Aero-servo-elastic simulations with a duration of 600 s were carried out". These two sentences are not coherent with each other. Does the first sentence mean that the simulation was carried out until the end of the transient period? In addition, the 600s duration of aero-elastic simulation usually refers to a period of time after the transient period. Please precise the duration of the simulation, and whether it includes or not the transient period.
- l. 329 "The first one is below 30%": consider changing the unit in Figure 11 to a percentage to match the description of the results in the text.
- l. 403 "They are small, but they are finite." The term "finite" is generally used in opposition to "infinite". Consider using "non-negligible" instead.
Citation: https://doi.org/10.5194/wes-2024-91-RC1 -
RC2: 'Comment on wes-2024-91', Anonymous Referee #2, 31 Aug 2024
The authors investigated the effect of blade cross-sectional deformations on the aerodynamic behaviour of a large-scale wind turbine, i.e., the IEA 15 MW RWT. The reviewer believes that the topic and the activity are very interesting, innovative and worthy of investigation. The study is overall of good quality and both methodology and results are presented in a rigorous and exhaustive way. Nonetheless, two main flaws are preventing it from being published directly and need to be revised:
A. In the Introduction, the authors struggle in highlighting the scientific question and the novelty of their study. Lines 39-46 in the manuscript, in particular, are not functional to the narration and only make the introduction more vague;
B. While the adopted FE methodology is rigorous and adequate for the scope of the study, the CFD one is inefficient and presents many limitations. In the Reviewer experience, a 3D approach is not necessary when using a RANS formulation. Switching to 2D would save you a lot of computational effort and allow to use CFD to produce all the polars necessary for the analysis. In this perspective:
- the adopted domain dimensions are too small for the Reynolds numbers under consideration
- a proper mesh sensitivity analysis should be performed
- not considering laminar-to-turbulent transition effects might be detrimental for the prediction of both absolute and relative loads
- your CFD model should be validated against experiments. I suggest the database from the AVATAR project
Minor revisions:
1. Section 2 has a lot of information that is repeated in the following paragraphs. Please revise it to avoid repetition;
2. Figures 3-5: please indicate in the caption what the different frames are, using letters or "left" and "right"
3. It would be nice to add a figure representing the FE model
4. Figure 4: the right picture is redundant, I suggest adding some details of the mesh instead
Citation: https://doi.org/10.5194/wes-2024-91-RC2
Status: closed (peer review stopped)
-
CC1: 'Comment on wes-2024-91', Baoxuan Wang, 05 Aug 2024
Dear all,
I took a quick look at this article, and I think it is a very interesting and valuable work.
Also, I have some comments for your consideration:
1) Line 123: In my understanding, "area moments of inertia" is not equal to "stiffness", is it right? Similar question is also occurred in the following content, such as Line 116: "the second moments of inertia"
2) Line 160, "Since the ElastoDyn module was used in the OpenFAST simulation, torsion was not accounted for", the statement is not right, Please see: Wang B.X., et al. 3D multiscale dynamic analysis of offshore wind turbine blade under fully coupled loads, Renew. Energy 223 (2024) 119985.
By the way, the cross-sectional deformations in the rotor blades are considered in the mentioned article (https://doi.org/10.1016/j.renene.2024.119985), by considering the change of nodal coordinates at each time step.
3) Line 163, "multi point constraints" can be done in many different ways in the software, which one were you using?
Aditionally, "The magnitude of the forces were calibrated so that the flapwise and edgewise bending moments from the loads simulations were well approximated", How to achieve it? Please clarified that in detail. Line 235, Are the bending moments are transferred to the shear forces? How to do this? It is not unimportant for readers.
4) The details need to be clarified all over the manuscript, including the detailed differences between the original model and your FEA model based on so-called MOCA, how about the eigen-frequencies?
5) For most of readers, they have few understanding about the MOCA. When it comes to that MOCA has been verified, I may think that it was only verified by small blade. Is there any limitation need to be mentioned?
6) 3D FEA results are expected to occur in the analysis for discussion.
Best regards,
Baoxuan Wang
Zhejiang University
wangbaoxuan@zju.edu.cn
Disclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-91-CC1 -
AC1: 'Reply on CC1', Julia Gebauer, 12 Aug 2024
Dear Baoxuan Wang,
Thank you for your interest and comments. In the name of all authors, I would like to respond to the discussion points you raised below.
- Line 123: In my understanding, "area moments of inertia" is not equal to "stiffness", is it right? Similar question is also occurred in the following content, such as Line 116: "the second moments of inertia"
Yes, true, the second moment of inertia (or area moment of inertia) is a purely geometric magnitude. Stiffness is the combination with material properties, e.g., the Young’s modulus. We will take care to thoroughly distinguish these terms in a revised version of the manuscript.
- Line 160, "Since the ElastoDyn module was used in the OpenFAST simulation, torsion was not accounted for", the statement is not right, Please see: Wang B.X., et al. 3D multiscale dynamic analysis of offshore wind turbine blade under fully coupled loads, Renew. Energy 223 (2024) 119985
As described in the OpenFAST Documentation, ElastoDyn uses only the 1st + 2nd flapwise and 1st edgewise modes to describe the beam behavior of the blades. There is no degree of freedom for blade torsion. Torsion is thus not considered in the kinematics of the blade, and can thus not be accounted for in the deformation. The torsional moments are consequently also not reliable, as they result purely from equilibrium, not from kinematics.
- By the way, the cross-sectional deformations in the rotor blades are considered in the mentioned article (https://doi.org/10.1016/j.renene.2024.119985), by considering the change of nodal coordinates at each time step.
Many thanks for letting us know. When I understand it correctly, you used the term “nodal coordinate” to describe the blade sensor positions along the blade span. Further, you applied the loads from the time series to generate a stress time series with FE. However, I did not find an analysis of cross-sectional deformations (change in airfoil geometry), which is the core of our manuscript.
- Line 163, "multi point constraints" can be done in many different ways in the software, which one were you using?
We used the contact element CONTA173 (3D 4-node surface-to-surface contact), with an extra node in the shear center as main node, and all cross-section nodes as helper nodes. The rigid option was used. We will add this information in a revised version of the manuscript.
- Additionally, "The magnitude of the forces were calibrated so that the flapwise and edgewise bending moments from the loads simulations were well approximated", How to achieve it? Please clarified that in detail. Line 235, Are the bending moments are transferred to the shear forces? How to do this? It is not unimportant for readers.
The flapwise and edgewise bending moment were approximated by applying concentrated forces as described in line 157 to 160. For more details see: https://windeurope.org/summit2016/conference/submit-an-abstract/pdf/35620222998.pdf
- The details need to be clarified all over the manuscript, including the detailed differences between the original model and your FEA model based on so-called MOCA, how about the eigen-frequencies?
As stated in section 2.1 the MoCA model showed differences in geometry and stiffness. Hence, from then on only the MoCA model was used.
Of course, we also checked the natural frequencies. We get frequencies of 0.533 Hz (1st flap) and 0.631 Hz (1st edge) with the MoCA blade model. Compared to the given values in the technical report of the IEA 15 MW RWT, the deviations are less than 5% and are thus considered negligible.
- For most of readers, they have few understanding about the MOCA. When it comes to that MoCA has been verified, I may think that it was only verified by small blade. Is there any limitation need to be mentioned?
Shell models are state of the art. MoCA is an in-house software with a parameterization that is very similar to WindIO. However, it is one possible implementation of how to create a shell model. To the opinion of the authors, it is not necessary to explain all details of the modeling strategy, as these are documented elsewhere (see https://wes.copernicus.org/articles/7/105/2022/). The reference is already included in the manuscript.
When it comes to validation, you are right. We validated the software and modeling strategy integrated in MoCA on a small blade up to now, as that was the only real blade where we had access to the detailed design information (geometry, layup, test setup, etc.). We are currently waiting for approval of a project where we will have access to the design data of a modern utility-scale wind turbine blade, and then we will see how MoCA performs with that one. As already mentioned, shell models are state of the art. It is known in science and industry that the bending behavior can be well approximated, whereas there may be issues with torsion. This limitation also holds for a MoCA model, which is solved in Ansys, so when it comes to FE implementation, it is an Ansys model. Industry is aware of the limitations in torsion.
- 3D FEA results are expected to occur in the analysis for discussion.
Can you please specify what FEA results you would like to see? The manuscript is about cross-sectional deformations, and these are presented in the paper. In the opinion of the authors, there is no need to present analysis results that are actually not needed for the core of the paper.
Best regards,
Julia Gebauer
Citation: https://doi.org/10.5194/wes-2024-91-AC1 -
CC2: 'Reply on AC1', Baoxuan Wang, 12 Aug 2024
Dear Julia Gebauer,
Thank you for your patient reply. Here are some points that I want to clarify, just for your reference only.
2. Although the torsion deformation cannot been considered in ElastoDyn, please note that the aerodynamic torque are considered. The torsional moments from kinematics seem does not seem to play a dominant role, from the perspective of magnitude.
3. In fact, as you said, analysis of cross-sectional deformations was indeed not conducted. I mentioned this only because your interesting research reminded me of it. In the mentioned work, the distributed nodal forces are calculated based on the time-varying nodal coordinates of each section, so the change or deformation in airfoil geometry was considered to some extent.
4. It seems better to illustrate with pictures.
5. To be more specific, has the lever arm change caused by large deformation been taken into account?
8. Sorry, the previous comments may have been misleading. Does the response of the 3D model match that of the 1D model? For reference only.
Thank you for your response and the effort you have put in for this, which is commendable.
Best regards,
Baoxuan Wang
Disclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-91-CC2 -
AC2: 'Reply on CC2', Julia Gebauer, 29 Aug 2024
Dear Baoxuan Wang,
Thank you for your feedback. We consider your comments for a revised version of our manuscript.
Best regards,
Julia Gebauer
Citation: https://doi.org/10.5194/wes-2024-91-AC2
-
AC2: 'Reply on CC2', Julia Gebauer, 29 Aug 2024
-
AC1: 'Reply on CC1', Julia Gebauer, 12 Aug 2024
-
CC3: 'Comment on wes-2024-91', Matthias Saathoff, 23 Aug 2024
Thank you for your insightful manuscript and the concise presentation of your approach and the results!
I would like to kindly bring to your attention a previously published paper that you may not have had the chance to read. The study concludes that with knowledge of blade distortions, rectified polars, twist, and sweep can be considered in turbine design to decrease possible gaps between design assumptions and field observations, particularly in terms of power performance and loads, cf. https://iopscience.iop.org/article/10.1088/1742-6596/1618/5/052011
Kindly reconsider your conclusion that the analysis of "aero-elastic simulations of a wind turbine under consideration of cross-sectional deformations in the rotor blades ... in has not been
carried out before by other groups".Many thanks,
Matthias Saathoff
Disclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-91-CC3 -
AC3: 'Reply on CC3', Julia Gebauer, 29 Aug 2024
Dear Matthias Saathoff,
Thank you for your comment on our manuscript. We are really happy to see your interest in our manuscript and appreciate the interaction. I would like to answer in the following on behalf of all authors.
In your paper, you address the change in blade geometry due to cool-down after manufacturing, the resulting deviation in lift and drag coefficients to the undistorted model, and the impact on aero-elastic loads during turbine simulations. I agree that we have overlaps when it comes to the impact of geometry changes on the aero-elastic performance. However, when we talk about cross-sectional deformations in our manuscript, we address the change in geometry due to mechanical loading during turbine operation and the corresponding deformation of the blade’s cross-sections. This is a dynamic process and was, to the best of our knowledge, not considered yet in literature. In this context, we also concentrated on the modeling of rotor blades and their deformations in the literature review, since it is the baseline for aero-servo-elastic turbine simulations.
However, we do see that changes in cross-sectional geometry cover a wide range of possible causes (as described, for example, in your paper). So, we consider adding a paragraph on this in a revised version of our manuscript.
Best regards,
Julia Gebauer
Citation: https://doi.org/10.5194/wes-2024-91-AC3
-
AC3: 'Reply on CC3', Julia Gebauer, 29 Aug 2024
-
RC1: 'Comment on wes-2024-91', Anonymous Referee #1, 30 Aug 2024
General commentsThis manuscript presents an analysis of cross-sectional deformations in wind turbine blades under operational conditions and their impact on aerodynamic loads. The added value of this work is to quantify the aero-structural couplings where the structural cross-sectional deformations impact the airfoil shape, which in turn affects the aerodynamic loads. This field of study is relevant for the wind energy community due to the increase in blade flexibility. Furthermore, quantifying couplings is particularly important for multi-disciplinary design optimization of wind turbines.The submitted manuscript presents weaknesses regarding (i) its contextualization with the literature, (ii) the strength of its methodology and (iii) its clarity and structure.The literature review presents relevant and related references. However, it does not provide sufficient argumentation to motivate the work, i.e. justify why one could expect cross-sectional deformations to have a significant impact on aerodynamic loads. The review can be improved in several ways described in the specific comments below, in order to contextualize the study better and highlight its scientific relevance.The methodology used in the article consists of a sequential process. Aerodynamic loads calculated from an undeformed turbine are applied to a finite element model of the blade to obtain a deformed configuration. This deformed configuration is in turn used to update the aerodynamic characteristic of the airfoil and enables the calculation of the aerodynamic loads for the deformed turbine. This analysis is applied to a single load case, with constant wind and at rated wind speed. Furthermore, the analysis is done for one turbine, the 15MW reference wind turbine. Finally, the torsional deformation of the blade is not taken into account in the aero-elastic simulations. As pointed out by the authors in the manuscript, this methodology has weaknesses and may justify the lack of significant results. This weakens significantly the relevance of the study. Consequently, the work would be significantly improved by either (i) providing a thorough justification for the chosen methodology, (ii) applying the analysis to other wind turbine designs (e.g. 10MW and 22MW reference turbines), or (iii) improving the methodology to obtain positive results.Finally, several parts of the manuscript lack structure, clarity and conciseness. Several suggestions to improve this aspect are described in the specific comments below.Specific comments
1. Literature review and contextualization.
- The literature review (l. 38-78) reads as a list of disconnected works, where the scientific contributions are not stated. Consider highlighting the problems that each cited work is addressing, and describing the associated results. Please highlight the similarities and differences between the cited work and the problem addressed in the manuscript. For the paragraph between l.38-56, linking the studies to the topic of blade cross-section deformation is critical to put the presented work in context.
- l.58: "Deformation of rotor blades ...": This paragraph is unclear. The first sentence mentions rotor blades and blade design, but it is not obvious how the cited works relate. Please make the link between the cited literature and the problem addressed by the manuscript more explicit.
- l. 65-70: "Preliminary work on [...] and the deformed cross-sectional shape was obtained." This part of the literature review is critical for the motivation of the presented study. The authors cite two of their previous works that have quantified cross-sectional deformations in wind turbine blades. However, the interpretation of these works is weak. The authors only state the method used in these works, and not the results of the analysis. What is missing in the reasoning is a quantification of such deformations. In other words, were there enough deformations obtained in the cited works to suggest an impact on aerodynamic properties and justify the present work?
- Consider adding literature on the topic of multi-disciplinary analysis and optimization (MDAO), and aero-structural couplings (e.g. bend-twist coupling). The wind turbine design research community already includes such couplings in state-of-the-art design optimization frameworks. It would be relevant to highlight to which extent the state-of-the-art in MDAO takes into account cross-sectional deformations. See for example the following two references:
- Mangano, M., He, S., Liao, Y., Caprace, D. G., & Martins, J. R. (2022). Towards passive aeroelastic tailoring of large wind turbines using high-fidelity multidisciplinary design optimization. In AIAA SCITECH 2022 Forum (p. 1289).
- Bortolotti, P., Bottasso, C. L., Croce, A., & Sartori, L. (2019). Integration of multiple passive load mitigation technologies by automated design optimization—The case study of a medium-size onshore wind turbine. Wind Energy, 22(1), 65–79. https://doi.org/10.1002/we.2270
- Consider adding literature on the topic of cross-sectional deformation in the context of failure analysis, including experimental works. This can help contextualize your results and quantify the deformations expected for operational loads vs ultimate loads.
- Consider making it more explicit that the study focuses on cross-sectional deformations due to operational loads and not ultimate or failure loads.
2. The description of the problem, research objectives and research questions is not clear in the text. As such, the text does not convey efficiently the relevance of the work.- l. 36-37 "When optimising the design with the minimisation of the blade mass as an optimisation target, an increasingly elastic behaviour of the blade is expected, including the elasticity and flexibility of the cross-sections": This statement introduces the problem addressed by the manuscript. Consider describing why an increase in cross-section elasticity and flexibility is a problem.
- l. 72: "Hence, the main objective of this paper is to study the influence of cross-sectional deformations in the rotor blade on the aero-elastic behaviour of the wind turbine.". Consider making the research question of the study explicit. For example: "To what extent do cross-sectional deformations impact the aerodynamic performance of the rotor, under operational conditions?"
3. Choice of methodology- l.73 "A simple test case with respect to normal operation of the wind turbine and a constant wind field is selected for the initial investigation presented in this paper.": Please add a motivation for this choice of analysis.
- l. 84 "A two-stage process is applied": please justify this choice of analysis. The state-of-the-art in MDAO does fully-coupled aero-structural analysis. Why has a sequential approach been chosen here and not a fully coupled one.
- l. 160-161 "torsion was not accounted for": this is a strong assumption for the study. Please justify why this modeling assumption was chosen, and its impact on the results of the study. There are several state-of-the-art aero-elastic tools that implement torsional degree of freedom. This is particularly critical in the light of the following statement: l. 405-406 "It is expected that especially torsion can result in significant in-plane cross-sectional deformations". If that is the case, why not include torsion in the first place in the study?
- l.163 "The loads were applied via multi point constraints that represented a load introduction similar to load frames". Please justify this model choice and its impact on the results of the study. Why simulate a blade-testing environment and not an operational environment, i.e. loads continuously distributed along the blade? This is particularly important considering the results of Figure 11. It seems that the high deviation observed at the location r/R = 0.8 is due to the presence of a load frame.
- l.188 "High-fidelity computational fluid dynamics (CFD) was used to compare the polars and ensure the validity of XFOIL results." Why use XFOIL for the generation of airfoil polars and not CFD directly?
4. Concise description of the method. Section "2. Methods" starting at l.80 presents the methodology of the work. The different steps of the analysis are reported: (i) Generation of loads using aero-structural simulations for the undeformed turbine, (ii) Generation of cross-sectional deformations from the extracted loads using a FE model of the blade and (iii) Generation of loads for the deformed configuration. However, the text lacks structure and conciseness.- The text describes in length adjacent aspects of the analysis that are not central to the study: creation of a FE blade model from the reference turbine data (l.85-91, l.109-111), verification of the FE blade model (l.112-135), verification of the airfoil polars between XFOIL and CFD (l. 190-215), choice of the number and placement of load frames (l.170-173). Consider moving these parts of the text into appendices.
- Consider restructuring section 2. Due to the fact that several models are used twice in the analysis, I would recommend first describing the steps of analysis with the support of Fig.1 and then including one section for each model used. For example: "2.1. FE structural model", "2.2. Generation of airfoil polars", "2.3. Aero-elastic simulations". Then, each section can describe when the associated model is used in the global process. For example: "Aero-elastic simulations are conducted during steps 1 and 2 as shown in Fig. 1, using the undeformed and deformed configuration of the turbine, respectively."
- Having a flowchart like Fig. 1 to illustrate the different steps of the analysis is great to support the description of the methodology. However, there is a discrepancy between the terms used in the figure and the terms used in the text, which can be confusing for the reader. To increase the relevance of the figure, consider aligning such terms and referring to the headers of section 2 in the figure.
- Some steps of the analysis are repeated excessively in the manuscript. For example, the use of OpenFast for aero-elastic simulations is mentioned in l.94-95 ("an aero-servo-elastic simulation of the turbine was performed using OpenFAST"), l.157 ("an aero-servo-elastic turbine simulation with OpenFAST"), l.228 ("The aero-servo-elastic simulations were performed with OpenFAST"), l.256 ("Aero-servo-elastic simulations with a duration of 600 s were carried out in OpenFAST"). Consider making the description of the analysis more concise and referring to section 2. when needed.
- l.247-254 "Recall the blade element theory... have an impact on the lift and drag forces.": The equations and notations introduced in this paragraph are not used in the rest of the manuscript. In order to improve the conciseness of the text, consider removing this part or moving it to a part of the manuscript used to motivate the work.
4. The results of the study are described in sections "3. Cross-sectional deformations" and "4. Coupling effect on the turbine behaviour". The documentation of the results is thorough but lacks interpretation and conciseness.- l.246 "the cross-sectional deformations at two positions along the blade are presented and discussed" and l. 319 "The cross-sectional deformations were determined for all cross-sections along the blade. However, due to space limitations, they cannot be discussed in detail here.". The distribution of the cross-sectional deformation would be relevant for the scientific community and for reproducibility of the results. Consider shortening section 3.1. and adding one figure and one paragraph describing the deformations across the blade.
- Consider combining Figures 7 to 10 to help the reader see the differences between each case.
- In section 3.1, the authors describe at length Figures 7 to 10. However, an interpretation of the results is missing. For example, l.292-294 "From the leading edge up to approximately c/8, the shell on the pressure side deforms into the cross-section. From there on up to approximately 5c/8, the pressure side shell deforms out of the cross-section.": This is a visual description of the results. Please provide some interpretation: can the deformation of the cross-section be justified by the type of loads or the stiffness of the components? Are these results similar to deformations described in the literature? Furthermore, consider making the description of these results more concise and to the point. This comment also applies to sections 3.2 and 3.3
- l.358-359 "Two regions along the blade span can be identified that show a change in aerodynamic properties. The first region is at around 20 % of the span, the second at around 80 % of the blade span.": This is an expected result considering the results shown in Figure 11. Please make the causality between the results of section 3.2 and section 3.3 clearer, i.e. the deformation shown in Figure 11 results in the change of aerodynamic characteristics in Figure 12.
- Figures 13 and 14 are particularly clear and efficient in representing the data.
- Consider linking the results of section 4 to the results of section 3 in a more explicit manner. Do the differences in lift and drag reported in section 3.2 justify the results of section 4?
5. Discussion of the results- l. 394 "This study revealed small changes ...": It could be relevant to have a separate section for the discussion of the results
- Consider contextualizing the results of section 4. Would the cross-sectional deformation have a significant impact on power production or other relevant metrics? What would be the impact for wind turbine design?
6. Conclusion of the work- l.14: "the initial results imply that further investigations should be carried out with more complex wind fields and different rotor blade designs" This statement needs to be backed up by either data or literature. Placing it in the abstract implies that the article supports this claim, which is not the case in this version of the manuscript. Please soften the language here, for example: "While these values are small, further investigations with ... could identify a stronger aero-structural coupling".
- l. 426-427 "The change in chord length was identified in the vicinity of 20 % and 80 % of the blade span.": Please put these results in context with the type of load applied, i.e. with discrete load frames and not a continuous loading.
- l. 432-439: "These relative deviations are quite small. However, ... We therefore expect the influence on the aero-elastic simulation to be greater when analysing extreme load cases.": This paragraph significantly weakens the results of the study. In essence, the author states that the relative deviations reported in the study are small because simple assumptions were taken, which negates the relevance of the work. This is done to justify the "negative results" obtained. However, it raises the question of why such assumptions were used. An alternative conclusion to the work would be to suggest that the aero-structural coupling between cross-sectional deformation and aerodynamic loads is negligible for operational loads. This would significantly increase the impact of the work and its relevance for the scientific community.
7. Description of the methodology- l. 6 "a 3D finite element (FE) model": please add details of the model in the abstract, for example stating the type of elements used.
- l. 93 "... and were used to calculate the aerodynamic loads via the blade element momentum theory (Jonkman et al., 2015). For the load calculation, an aero-servo-elastic simulation of the turbine was performed using OpenFAST (National Renewable Energy Laboratory, 2023).": Please clarify whether the "aerodynamic loads" and "loads" mentioned in the two sentences are calculated with different tools, and if loads due to gravity are included in the analysis.
Minor comments- For the cited work "Gebauer, J. and Balzani, C.: Cross-Sectional Deformation of Wind Turbine Rotor Blades, in: The 33rd International Ocean and Polar Engineering Conference, Ottawa, Canada, 2023.", the full paper (if there is any) couldn't be found from the information provided. Only a one-pager was found, without mention of relevant results (https://publications.isope.org/proceedings/ISOPE/ISOPE%202023/data/pdfs/160-2023-TPC-0416.pdf). Please provide the reference to the full paper or remove the reference.
- Consider adding an illustration of the cross-section topology in the manuscript.
- The term "MoCA model" is used repeatedly in the manuscript. The qualifier MoCA refers to a specific modeling tool, and would not be meaningful for readers not familiar with it. Consider using another term, such as "the 3D blade geometry model".
- The term "OpenFAST simulations" is used repeatedly in the manuscript. Consider using the term "aero-elastic simulations" instead to highlight the type of analysis conducted, instead of the tool used.
- The authors use the past tense repeatedly in the manuscript. Please make sure to use past tense when describing your numerical experiments, and use present tense when describing your results. For example, l.326 "The relative deviation was plotted against the normalised spanwise coordinate": instead of using "was plotted", consider using "is shown" instead.
- l. 11. "depend largely": the qualifier "largely" does not add extra meaning here. Consider rephrasing to keep the abstract concise and to the point.
- l. 26 "which is calibrated by the aerodynamic twist angle". Please consider replacing the term "is calibrated by" with "depends on" in order to increase the precision of the text.
- l. 31 "In the context of this paper, the structural composition of the blade defines its resistance against cross-sectional deformations": please clarify or rephrase. This sentence is unclear.
- l. 35 "structural loads increase." Consider aligning the term used here with the load types enumerated in l. 19-20. This will help the reader understand exactly what type of load is meant by "structural loads".
- l. 41 "The study further focused on bifurcation analysis, which is not the subject of this paper.": Consider removing this sentence.
- l. 59 "energy-based method for thin-walled ...": for modelling? analysing? designing? Consider adding a verb in this sentence to characterize the method.
- l. 66 "a three-dimensional finite element model": please use "3D" or "three-dimensional" consistently throughout the manuscript.
- l. 236 "The simulation was carried out until the rotor blade behaviour became periodic." and l. 256 "Aero-servo-elastic simulations with a duration of 600 s were carried out". These two sentences are not coherent with each other. Does the first sentence mean that the simulation was carried out until the end of the transient period? In addition, the 600s duration of aero-elastic simulation usually refers to a period of time after the transient period. Please precise the duration of the simulation, and whether it includes or not the transient period.
- l. 329 "The first one is below 30%": consider changing the unit in Figure 11 to a percentage to match the description of the results in the text.
- l. 403 "They are small, but they are finite." The term "finite" is generally used in opposition to "infinite". Consider using "non-negligible" instead.
Citation: https://doi.org/10.5194/wes-2024-91-RC1 -
RC2: 'Comment on wes-2024-91', Anonymous Referee #2, 31 Aug 2024
The authors investigated the effect of blade cross-sectional deformations on the aerodynamic behaviour of a large-scale wind turbine, i.e., the IEA 15 MW RWT. The reviewer believes that the topic and the activity are very interesting, innovative and worthy of investigation. The study is overall of good quality and both methodology and results are presented in a rigorous and exhaustive way. Nonetheless, two main flaws are preventing it from being published directly and need to be revised:
A. In the Introduction, the authors struggle in highlighting the scientific question and the novelty of their study. Lines 39-46 in the manuscript, in particular, are not functional to the narration and only make the introduction more vague;
B. While the adopted FE methodology is rigorous and adequate for the scope of the study, the CFD one is inefficient and presents many limitations. In the Reviewer experience, a 3D approach is not necessary when using a RANS formulation. Switching to 2D would save you a lot of computational effort and allow to use CFD to produce all the polars necessary for the analysis. In this perspective:
- the adopted domain dimensions are too small for the Reynolds numbers under consideration
- a proper mesh sensitivity analysis should be performed
- not considering laminar-to-turbulent transition effects might be detrimental for the prediction of both absolute and relative loads
- your CFD model should be validated against experiments. I suggest the database from the AVATAR project
Minor revisions:
1. Section 2 has a lot of information that is repeated in the following paragraphs. Please revise it to avoid repetition;
2. Figures 3-5: please indicate in the caption what the different frames are, using letters or "left" and "right"
3. It would be nice to add a figure representing the FE model
4. Figure 4: the right picture is redundant, I suggest adding some details of the mesh instead
Citation: https://doi.org/10.5194/wes-2024-91-RC2
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
346 | 139 | 83 | 568 | 10 | 12 |
- HTML: 346
- PDF: 139
- XML: 83
- Total: 568
- BibTeX: 10
- EndNote: 12
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1