Aeroelastic Instabilities of the IEA 15 MW Rotor During Extreme Yaw Maneuvers

Höning, Leo; Herráez, Iván; Stoevesandt, Bernhard; Peinke, Joachim

doi:10.5194/wes-2025-281

Preprints

https://doi.org/10.5194/wes-2025-281

Preprints

27 Dec 2025

| 27 Dec 2025

Status: this preprint is currently under review for the journal WES.

Aeroelastic Instabilities of the IEA 15 MW Rotor During Extreme Yaw Maneuvers

Leo Höning, Iván Herráez, Bernhard Stoevesandt, and Joachim Peinke

Abstract. This study explores the aeroelastic behavior of the IEA 15 MW wind turbine rotor during dynamic yaw maneuvers under storm conditions through high-fidelity computational fluid dynamics (CFD) simulations. The focus is on blade vibration responses to a variety of yaw misalignments while maintaining constant pitch and azimuth settings. Utilizing the Geometrically Exact Beam Theory (GEBT) and the OpenFOAM framework, the study reveals that certain yaw angles lead to significant edgewise blade vibrations, with distinct responses observed among the rotor's three blades. Detailed analysis of one blade at varying fixed yaw angles, employing Hilbert-Huang transformation, uncovers a lock-in effect where flow structures synchronize with the blade's eigenfrequencies, resulting in pronounced blade tip vibrations. Key findings indicate that the most substantial vibrations occur during specific yaw angles, suggesting that the rotor's structural integrity could be compromised under certain dynamic conditions. This work enhances the understanding of aeroelastic instabilities during off-design yaw maneuvers and highlights the need for operational strategies in managing rotor performance during extreme conditions.

Received: 15 Dec 2025 – Discussion started: 27 Dec 2025

Competing interests: Some authors are members of the editorial board of the Wind Energy Science Journal.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Leo Höning, Iván Herráez, Bernhard Stoevesandt, and Joachim Peinke

Status: final response (author comments only)

RC1:
'Comment on wes-2025-281', Anonymous Referee #1, 28 Jan 2026
This manuscript investigates the aeroelastic response of a standstill rotor undergoing a yaw maneuver in extreme wind conditions. Two particular conditions are additionally investigated with the rotor in a standstill. The analysis is rigorous and in-depth. Some remarks are listed below:
The motivation is not fully clear. The study aims to perform similar investigations as performed on smaller rotors on a larger one? If so I would expect more parallels to be drawn in the conclusions.

I’m not quite sure how inflow turbulence is accounted for in the simulations. A boundary condition on turbulent eddy viscosity is listed in table 1, but very little other detail is mentioned. This is relevant as the inflow turbulence may interact with the turbulent structures generated by the blades and alter the aeroelastic response.

One detail that is again not clear is whether drivetrain and yaw bearing flexibility are accounted for. Since edgewise vibrations are the main concern, drivetrain flexibility (and structural damping) may impact the observations herein. I would expect this aspect to be discussed.

The quality of writing is ok. Some paragraphs appear too long/convoluted. I’ve highlighted some inconsistencies below.

Detailed comments:
Table 2: tilt angle of the IEA 15 MW should be 6°.

Table 2: How are the pitch angles and azimuth positions chosen? It is obviously impossible to simulate lots of conditions with high fidelity methods but some motivation should be provided. Also, why is the pitch angle of blade #2 different from the others?

Section 2.3: The set-up is validated referencing simulations where the rotor is operating. The grid sensitivity study also appears to be performed in such conditions. These conditions are very different from those simulated herein, where the rotational speed is zero and the blades are vibrating. I’m not sure that given these differences the verification of the model is fully convincing.

Section 2.4: Blade torsion is mentioned when discussing damping factors but I haven’t seen it brought up again elsewhere. Are torsional vibrations irrelevant?

Figure 4: Is it possible to attribute the observed behavior to one or more modes? Based on the shape of the energy curve, is it possible that during ramp-up the first mode is dominant while during ramp down the second edgewise mode is dominant? Moreover, do torsional modes play a role?

L290-295: given the provided explanation, which seems logical, about how vibration amplitude can increase despite total power being negative, the poi of this section is not entirely clear. I.e.: if total power is not directly correlated to the amplitude what is the usefuleness of the analysis in Fig. 11?

Figure 13: Why was r/R = 0.9 chosen?

Section 3.2.4: the 37° case shows significant hysteresis on Cl, while the 60° case does not. Can the reason be reconducted to the different nature of the vibration (edgewise vs flapwise), or are they related to unsteady aerodynamic effects (vortex shedding for example) ?

L251: At this point..

L375: This phrase is unclear, please rephrase. In addition, the discussion in this and the previous paragraph is also unclear. Why lock-in happens in one case and not the other is not easy to understand, as in both cases one IMF has a peak at the edgewise natural frequency. Is it due to the shift of this peak in the flexible simulations? Pleas eexplain this more clearly.

Section 3.2.5: The analysis of the velocity probe highlights some interesting findings. It is not clear what phenomenon is triggering the vortex shedding which leads to the critical lock-in phenomenon in the FLEX37 case. Authors motivate that it is not Strouhal vortex shedding, but I haven’t seen an alternative explanation.

Conclusions: the premise in the introduction is that somewhat similar studies have been performed previously on smaller rotors. The conclusions would strongly benefit from a critical comparison to such studies. Although I imagine they cannot be directly compared, are there any indications that size may play a critical role? Can any differences are similarities be noted? This would go great legths in demonstrating the importance of these approaches.
Citation: https://doi.org/10.5194/wes-2025-281-RC1
RC2: 'Comment on wes-2025-281', Anonymous Referee #2, 29 Jan 2026

General Remarks:
This work presents a study on vortex induced vibrations on the IEA15MW wind turbine in extreme conditions, where the turbine is in stand still. An instability is found around 40deg yaw-misalignment on one blade with small inclination angles. Subsequently an in-depth analysis of the this isolated blade is carried out.
The content of the article is well presented but the language and flow could be improved in several parts. Furthermore, care should be given to the bibliography and the references as at times references are not found in the bibliography or not rendered correctly in the text. Abbreviations are not always introduced properly.
The implications summarized in the conclusion (Line 439 onwards) are very relevant but are only superficially addressed in the work itself. I recommend detailing specifically how this work helps contribute to, for example, controller design.
Major Comments
The authors mention Grinderslev (2023) and that numerical results are very sensitive to low-inclination angles. It is of my understanding that the core analysis of this paper is also based on a low-inclination case (blade 2). Could the authors comment on the uncertainty related to the results based on the sensitivity that Grinderslev found?
The authors also refer to Pirrung (2024) to highlight the importance of simulating a fully coupled rotor. However, as far as I understand, an isolated blade is analyzed for the in-depth analysis, while the coupled approach is mentioned in the conclusion as future research. Could the authors comment on their decision to focus on an isolated blade and the possible implications for the findings of this paper?
Introduction:

In general, the introduction would benefit from more literature references. Specifically, whenever a new phenomenon is introduced or a statement is presented (some examples are highlighted in the minor comments)
Numerical methods and setup:

The edgewise frequency amplitude was set to 5%, which is assumed to be reasonable. Could the authors comment on the basis for this assumption? Perhaps an additional source would be helpful.
Results and discussion

Some figures (especially in Section 3.2) are misplaced relative to the text making it hard for a reader to focus on the analysis. It is recommended to place the figures as close as possible to the analyzing text.
The results section is structured so that the case described in Sect 3.1 identifies a critical situation in which VIV occurs on blade 2. Sect 3.2 isolates this case and provides an in-depth analysis of the driving mechanism behind this instability. This context could be communicated more clearly as currently a reader expects two test cases to be analyzed that are not necessarily connected based on the paragraph between Lines 86-92. This leaves one with open questions after reading Section 3.1, such as:
“An instability that occurs between 33° and 55° yaw misalignment is shown and the energy exchange between the fluid and structure is analyzed, however the underlying reason for this instability is not discussed.”
These questions are addressed largely in Section 3.2. It is recommended to outline the arc of the paper more clearly before the results section.
Conclusion:

Line 441:

“Furthermore, it could contribute to incorporating critical rotor-inflow conditions into the controller design process. Actively pitching the blades can readily circumvent these vibrations, because of a fast change in angle of attack. In the event of a pitch motor failure, these critical inflow-turbine scenarios would need to be avoided.” While an interesting aspect, the control aspect is only raised in the conclusion. It is not fully clear how this work might contribute to controller design. Do the authors suggest to identify as many cases as possible where VIV occurs of a specific rotor following the approach presented in this work and letting this inform control design?
Minor Comments:
Line 14: (IEA (2023)) reference not formated correctly, should be (IEA, 2023)
Line 14-16: It is stated that the deployment of wind turbines neccessitates a deeper understanding of the complex dynamic whereas the reason is more rooted in the changing nature of the turbines that are driven by the economic pressure (as is stated afterwards by the authors), consider restructuring. Also, some references could be added here.
Line 16-17: Please provide some background for why larger rotors are beneficial in terms of LCOE, and why this leads to more slender blades. Consider adding sources.
Lines 20ff: SIV and VIV are introduced. While the latter is defined afterwards, a definition of SIV is missing. Please consider adding foundational references for both.
Line 31: … turbulence modeling and grid characteristics “in CFD analyzes” of a 10 MW wind turbine
Line 34: For clarity it would help if aerodynamic power was defined by the authors e.g., power injected into the structural system by aerodynamic loads.
Line 36: Horcas 2020 is cited with a description of the work done in the paper but no conclusion concerning the influence of blade tip geometry with regards to changes in the wake is given.
Line 39; FSI was not introduced yet
Line 42: “In this configuration, substantial blade deflections considerably affect power injection near the blade tip.” What is the driving difference between isolated blade set-ups and fully coupled rotors. I would assume that for an isolated blade, blade deflections also affect power injection to a similar degree near the blade tip.
Line 44-55: The gap in existing literature is not fully clear to an independent reader at this point and is only detailed afterwards. The paragraph could be restructured for clarity, first pointing towards the current gap in the literature and then mentioning in which way this work addresses this gap. Moreover, the necessity of this analysis compared to previous analyses of the 10MW rotor could be further explained. Drawing a comparison to the IEA10MW could also be an interesting discussion point for the conclusion.
Line 57: “To investigate the blade deformation response of a large wind turbine to yaw misalignment during storm conditions,” Is blade deformation resonse the correct expression here?
Line 58: “generic” is not necessary
Line 60: TURBINIA, reference is not correct. I can’t find the entry in the bibliography. I assume this should be Scherpers et al. 2025.
Line 60: It is mentioned that several numerical studies have been carried out but only one reference to TURBINIA is provided. Is it that Turbinia contains multiple studies or might references be missing here? Please clarify.
Line 90: IEC not introduced
Table 3. Absolute values for thrust and torque of the reference (fine) might be helpful here. Also, could the authors comment on the computational cost difference between the different mesh configurations?
Line 124: IEAWindTask37b (2020), no entry found int the bibliography. Also, the formating in the text is not correct.
Line 139: IEAWindTask37a (2020), no entry found int the bibliography. Also, the formating in the text is not correct.
Figure 2d: Could the authors comment on why around r/R 0.6 the phase offset reduces to zero and back to 180° again for further outboard regions. For this region of the span the maximum deflection in edgewise and flapwise DOF seem to be in phase.
Line 150: "Empirical Mode Decomposition” usually, no quotation marks are used when an abbreviation is introduced. Same for IMF.
Line 186-188: The description of the the instability in 3e could be rephrased for improved clarity. The word "lead" might be confusing in an instability context.
Table 4: The blade position should be added, I assume this would be 330° matching blade 2 of the previous subsection.
Line 251-252: “A slight shift of inherent oscillation frequencies in comparison to the theoretical eigenfrequencies of the blade towards smaller values can be noted”. Were the structural frequencies matched (validated) in a modal analysis of the blade structure previously? Could it be related to how the drive train is modeled or the lack of a flexible tower?
Figure 8: It might be helpful to add the energy of the yawing case as a reference (as a dotted line perhaps). Making it easier to compare the two.
Line 258: “blade tip of the blade” . remove one “blade” from that sentence.
Line 260: Could the authors comment on why the outer blade sections contribute to energy injection into the structure for the case presented in Figure 8 in contrast to Figure 4? Would this be enough to trigger a flapwise instability as well?
Line 393 “by the by the”
Line 440: “The study presented describes an approach for predicting blade vibrations of multi-megawatt wind turbines under storm conditions. It is directly relevant to the simulation of extreme load cases, enabling more accurate prediction of peak aeroelastic loads and lock-in behavior under yawed inflow.”

This study seems to present an analysis of the instabilities that can occur while applying tools and methods that are referenced. If so, I am not sure the statement is totally accurate. Otherwise, the approach itself should probably take on a larger role in the paper. Please clarify
Line 444: “Although being slower than pitching, yawing the turbine proved being effective in guiding the system to a stable state while reducing vibration amplitudes to roughly half its magnitude and thus offers a mitigation possibility when a fast pitch movement is restricted”

To my understanding, VIV is especially a problem in situations when the yaw motor is broken or deactivated for maintenance. Hence, this control option seems to have limited applicability.

Citation: https://doi.org/10.5194/wes-2025-281-RC2

Leo Höning, Iván Herráez, Bernhard Stoevesandt, and Joachim Peinke

Viewed

Total article views: 367 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
220	127	20	367	17	13

HTML: 220
PDF: 127
XML: 20
Total: 367
BibTeX: 17
EndNote: 13

Views and downloads (calculated since 27 Dec 2025)

Month	HTML	PDF	XML	Total
Dec 2025	72	35	6	113
Jan 2026	136	84	14	234
Feb 2026	12	8	0	20

Cumulative views and downloads (calculated since 27 Dec 2025)

Month	HTML	PDF	XML	Total
Dec 2025	72	35	6	113
Jan 2026	136	84	14	234
Feb 2026	12	8	0	20

Viewed (geographical distribution)

Total article views: 326 (including HTML, PDF, and XML) Thereof 326 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 06 Feb 2026

Short summary

High-fidelity fluid-structure coupled simulations of the IEA 15 MW rotor under storm and yaw misalignment shows that certain misalignments trigger strong edgewise vibrations. Growth surges when effective power turns positive near −35° and fades near −43° yaw. Single-blade analysis finds lock-in at −37° with large tip motion and stability at −60° due to off-resonant Strouhal shedding. It is concluded that aeroelastic response is inflow-specific and operational mitigation strategies are needed.


Total:	0
HTML:	0
PDF:	0
XML:	0