the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Oct 2024
Design of a two-bladed counterpart to the three-bladed INNWIND 20 MW offshore reference wind turbine
Abstract. It is still an ongoing discussion in the wind field whether a two-bladed or a three-bladed rotor is economically superior for large offshore wind applications. While the inherent question is clear, the answer has to tackle a multitude of challenges. One is the task of establishing an equal three- and two-bladed turbine design to fill the research gap on the engineering side. Due to the notorious difficulties of reaching an entirely equal design, the goal is to remain as similar as possible, always with the premise of reaching a fair comparability. For this purpose, the focus of the paper is on striving for the most similar aerodynamics, including an equal absolute power curve, as well as a most similar blade structure in terms of a comparable fiber composite layer stack and matching ultimate and fatigue material stresses and stability limits. The controller features the same architecture and is equipped with basic load mitigation algorithms. The gains are tuned by utilizing an objective control cost criterion. The generator is adapted to the changed torque and rotation speed, and the tower is adapted to fatigue loads. A complete load and cost comparison and further manufacturing, transportation, installation, and operation and maintenance aspects are outside of the scope herein. In turn, the paper presents a thought-through possibility of achieving a most similar two-bladed turbine based on a three-bladed reference. While this re-design could generally be performed with any detailed three-bladed horizontal-axis wind turbine, the steps are showcased on the three-bladed 20 MW INNWIND reference turbine. The final turbine models, controller, and blade models are made open source with the paper at hand.
- Preprint
(780 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2024-121', Anonymous Referee #1, 20 Dec 2024
The paper systematically redesigns a two-bladed wind turbine to make it as close as possible to the three-bladed Innwind 20MW RWT, taking aerodynamic, structural and control aspects into account.
Static, fatigue loads and Campbell diagram are compared for the two designs, with the design made open-access;
Even though these technical rigor, I miss the fundamental research questions. While the effort to achieve such a “fair comparison” is significant, it’s not sufficiently novel. Even though it tries to tackle the issue “"Clear-cut cost comparisons between two and three-bladed machines are notoriously difficult because of the impossibility of establishing equivalent designs" (Burton et al., 2021, p. 381).” However, the issue of “…impossibility of establishing equivalent design ” remains unresolved to me. Because even though the two design’s aeroelastic performance are close, the rotors are different at the end, with one rotor size larger than the other. It’s still difficult to do cost comparison of the two designs.
In my personal view, the paper doesn’t seem to introduce fundamental new concepts or methods. I think it can be published as a technical report rather than a scientific paper. Given these concerns, I find that the paper, in its current form, doesn’t meet the scientific significance as a manuscript published on WES “Does the manuscript represent a substantial contribution to scientific progress within the scope of WES (substantial new concepts, ideas, methods, analyses, or data)?” (WES - Review criteria). I recommend rejecting the manuscript. However, I encourage the authors to refine their research questions and objectives to leverage their current tools and design to provide deeper and more impactful insights to the community.
Citation: https://doi.org/10.5194/wes-2024-121-RC1 -
AC1: 'Reply on RC1', Fabian Anstock, 22 Dec 2024
Dear Referee #1,
Thank you for reading and commenting on our paper. We agree with the general description in the first two paragraphs. Still, we would like to react to the criticism provided afterward and open a constructive discussion on the novelty of our approach, which we, naturally biased, think is relatively high since we could not find comparable work elsewhere.
First of all, thanks for pointing out the “rigor” nature of our work.
What seems to be in question is the general novelty of the content because we did not completely solve the “… impossibility of establishing equivalent designs …” (Burton et al., 2021, p. 381) of two- and three-bladed wind turbines. This will, however, never be possible to its full extent because an equivalent design must have the same design issues, the same advantages – and in the end (paradoxical) the same number of blades (and costs). In that sense, there could never be a valid cost comparison of two- and three-bladed turbines, which would, in our view, be a pity for the research community. If our paper is not officially published, it will also unnecessarily complicate future work in this field due to less precedent work. Note that we also specifically addressed issues we observed during our work, which were out of scope and could potentially be tackled in upcoming projects.
We would like to open the discussion here on how a most similar design of two- and three-bladed turbines could be established. Our approach (more detailed described in the presented paper) includes that we:
- maintained the same relative chord layout and the same airfoils and airfoil positions along the blades and scaled the chord lengths with respect to the rotation speed. (not novel)
- Then, we adapted the blade twist to achieve the same angle of attack and, thus, the same airfoil characteristics at each cross-section. (relatively novel)
- Then, we increased the blade length to counteract the natural about 4% lower aerodynamic efficiency of a two-bladed rotor. Without this, a turbine with a lower energy yield would be compared against a turbine with a higher energy yield, including many implications. (novel and only mentioned by our work (mainly Anstock et al., 2019) that led to this paper)
- Then, we scaled the structure to establish equal material stresses for reasonably scaled loads, including load differences in edgewise and flapwise directions. (novel and only described by our work (mainly Schütt et al., 2020) that led to this paper)
- Then, we used a tip speed design that reached equal structural stability for the scaled loads and a similar material layup. (novel and only described by our work (mainly Schütt et al., 2021) that led to this paper)
- Then, we established an objective control design method, the control cost criterion, to obtain equally suitable controller gains. (using optimization functions is not novel, but utilizing these controller evaluation functions (Anstock et al., 2020) to tune two- and three-bladed turbines likewise is novel and led to this paper)
- We introduced non-standard control features that are, in the case of the speed exclusion zone, vital for such a two-bladed turbine. (novel in its equal application to both turbine types (Anstock et al., 2021))
- Then, we obtained equivalent blade fatigue stresses for the two- and three-bladed turbines for fatigue loads. (novel)
- Then, we performed the same design steps for a teetering hub version of the two-bladed turbine. (novel)
- Finally, we evaluated our procedure critically and addressed and described observed challenges. (novel in this detail)
In conclusion, our thorough summary on establishing the most similar two- and three-bladed designs combines known and new methods to achieve a novel outcome and provide scientific insights into issues of establishing equivalent designs for large 20MW turbines.
Dear referee #1 or interested reader, we would be happy if we added another perspective to your point of view. If you would like to maintain your standpoint of missing novelty, we would be glad to receive references to similar (or better) approaches for establishing a most comparable design of two- and three-bladed turbines, or gain your confirmation on its sufficient novelty.
Kind regards
Fabian Bölle né AnstockCitation: https://doi.org/10.5194/wes-2024-121-AC1
-
AC1: 'Reply on RC1', Fabian Anstock, 22 Dec 2024
-
RC2: 'Comment on wes-2024-121', Anonymous Referee #2, 13 Jan 2025
The paper outlines the framework and presents the results of redesigning the two-bladed version of an existing conceptual 20MW wind turbine initially designed with three blades.
It is clear that the authors have dedicated substantial effort to this work, both on the aerodynamic but most importantly on the aeroelastic and the structural dynamics side. Additionally, the contribution offers valuable information and insights into scaling procedures and principles.
The proposed design is not entirely realistic, as the authors make several assumptions during the process of converting the 3-bladed version into a 2-bladed one. Nevertheless, the scientific value of the contribution lies in the stepwise approach followed, which effectively highlights the differences between the two concepts. While the proposed design is not optimized, it can serve as a solid starting point for a design optimization study, with the primary objective being the minimization of a specific cost metric.
Given the above, I would recommend publication of the paper after some revision is made to the original text. In the accompanying pdf the authors can find some specific comments. Below, the key points necessitating further attention or rebuttal are outlined:
1) The authors reference several of their previous papers related to the same topic. It would be beneficial to explicitly outline the new contribution of the present work in the introduction section. Readers should clearly understand the starting point—what has already been accomplished and where it can be found—and how this work differs from and builds upon those earlier contributions.
2) Regarding the aerodynamic design, the authors apply a uniform scaling coefficient to the chord while altering the optimal tip speed ratio. However, it is well established that the optimal dimensionless chord distribution varies with the design lamda value (see, for example, the Wind Energy Handbook for details). Consequently, scaling the chord uniformly does not yield an equivalent aerodynamic design, as the new design will no longer be aerodynamically optimal (assuming the original design was an optimized one). This suggests that achieving the same power output could be possible with a smaller increase in radius. Please address this point in your discussion.
3) The Campbell diagram of Figure 8 is not sufficiently discussed. There is a lot of information in the plot that is not discussed in the text (eg. what is the stiff-stiff and what is the soft-soft design?)
4) The methodology for determining the ultimate loads is somewhat unclear. It appears that these loads are derived from the original 3-bladed design using an up-scaling law rather than being calculated through aeroelastic simulations of the Design Load Cases specified in the standards. Please clarify this point. If this is indeed the case, it should be explained in greater detail how this approach aligns with the new design featuring the teeter hub. In the teeter hub design, the ultimate loads might arise from a different Design Load Case compared to a fixed hub turbine (e.g., a shutdown scenario). This aspect requires further elaboration in the text.
-
AC2: 'Reply on RC2', Fabian Anstock, 17 Jan 2025
Dear Referee #2,
Thank you very much for your valuable and constructive review and the pdf comments. We will address your points 1-4 in an updated version of the paper. Beforehand, here are our brief comments on your points:
- It is true that previous papers have led to this scientific contribution. We will add a paragraph in the introduction describing earlier work's influence, overlap, and new additions to increase transparency and overview.
- Thanks for your valuable comment on the optimal chord length for changed tip speed ratios. Indeed, our approach is simplified by neglecting partial effects of the optimum chord length scaling in the root section of the blade from 0% to ~20—30% of the blade length. (At 22% blade length the relative chord length difference of both methods is ~2%. At 30% blade length, the difference is 1% and 0% up from 54% blade length.) We assumed that the outer 70-80% of the blade is the major contributor to aerodynamic loads and efficiency. This simplification then enables a more straightforward approach to structural scaling and design. However, it is true that the aerodynamic scaling laws, such as those described by Burton, lead to an even higher aerodynamic comparability of the blade root section. We will address this in the aerodynamic section and the discussion.
- Further descriptions of the Campbell diagram will be added. We are happy to do so because of its high importance in analyzing the challenges of the two-bladed turbines.
- Indeed, your guess is correct. We scaled the ultimate loads and received the fatigue loads from simulation time series. The conformity of scaled ULS and DLC simulations is based on an earlier paper (and on the observations made in the results of an upcoming PhD manuscript). Nonetheless, we will clarify this better in the paper at the end of the teeter blade section and the discussion and mark the potential risk of ULS that indeed could exceed the scaling rules.
Thanks again for all your constructive thoughts. Please let us know if you find the updates appropriate once a new version of the paper is uploaded.
Kind regards
Fabian Bölle né AnstockCitation: https://doi.org/10.5194/wes-2024-121-AC2
-
AC2: 'Reply on RC2', Fabian Anstock, 17 Jan 2025
Model code and software
Repository of x-Rotor – two-bladed wind turbines Fabian Anstock, Marcel Schütt, and Vera Schorbach https://doi.org/10.5281/zenodo.13269498
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 181 | 58 | 88 | 327 | 10 | 7 |
- HTML: 181
- PDF: 58
- XML: 88
- Total: 327
- BibTeX: 10
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1