the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Analyzing the Impact of Aeroelastic Model Fidelity on Control-Co Design Optimization of Floating Offshore Wind Turbines
Abstract. This work investigates the influence of aeroelastic modeling fidelity on design optimization of floating offshore wind turbines. To this end, the QBlade simulation environment was coupled to the Wind Energy with Integrated Servo-control wind turbine design and optimization framework. QBlade offers aerodynamic and structural models with varying levels of aeroelastic fidelity within a computationally efficient implementation. This enables time-domain optimization studies with levels of aeroelastic fidelity that are currently often deemed unfeasible for such purposes due to the computational expense involved. Five fidelity combinations are considered, ranging from blade element momentum aerodynamics with torsion-constrained Euler–Bernoulli beams to lifting-line free vortex wake aerodynamics with fully populated Timoshenko beams. To assess how aerodynamic and structural modeling fidelity influences optimization outcomes, the parameters of the floating wind turbine controller are co-designed together with the floating substructure, a system typically considered less sensitive to aeroelastic fidelity. The results show that controller tuning, structural load predictions and final design outcomes are all affected by the chosen fidelity level. Higher fidelity models broaden the design space through less conservative load estimates and variation in rotor operation, which in turn lead to more efficient platform designs. Increasing aeroelastic fidelity therefore improved the quality of the optimization results, albeit at the expense of higher computational cost.
- Preprint
(5356 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on wes-2025-174', Anonymous Referee #1, 04 Oct 2025
-
RC2: 'Comment on wes-2025-174', Anonymous Referee #2, 17 Oct 2025
This article presents a design optimization study using QBlade with multiple levels of modeling fidelity. The authors clearly demonstrate the influence of model fidelity on optimization outcomes and design results. While the findings are well presented and valuable to the community, the practical implications for design decision-making remain somewhat unclear. Addressing a few underlying assumptions and design philosophy questions would strengthen the paper.
Major Comments:
- The manuscript could more clearly guide readers on how to apply these findings in practice. For instance, should designers favor higher-fidelity models because they yield less conservative results, or rely on lower-cost models while accounting for uncertainty through safety factors? Given the current reliability challenges of offshore turbines, should the design community aim to reduce conservatism and safety margins? Alternatively, should model fidelity be driven by the need to accurately capture specific design features, like the ability to represent blade torsion in bend–twist coupled blades?
- Section 2.3, particularly its introduction, appears central to interpreting later results. A summary table or schematic comparing the effects and assumptions of the different model fidelities could improve clarity and accessibility.
- The study uses only DLC 1.1 for design optimization. Could the authors comment on why other design load cases (e.g., DLC 1.6 or 6.1) were excluded, given that prior work has shown their significant influence on common design constraints?
- It might be informative to compare the results of a low-fidelity optimized design when re-evaluated using a higher-fidelity model.
- Since models are ultimately approximations of reality, do the authors assume that higher-fidelity models are more accurate representations of real designs? If so, what evidence or validation supports that assumption?
Minor Comments:
- The tower mass in Table 1 appears inconsistent—perhaps a comma/decimal point issue. Please verify the value.
- Consider merging Tables 3 and 4 for improved readability.
- The manuscript states that the Frobenius norm is used to compute DELs, which differs from standard fatigue damage calculations and is typically applied to matrices. Please clarify or correct this methodology.
- Around line 412, the term “floater bandwidth” is described as ω_pc, though earlier ω_pc refers to the pitch control bandwidth. Are you instead referring to a coupled platform–pitch frequency? Please elaborate.
- In Figure 11, it would be helpful to mark or annotate the dominant platform, tower, and rotor natural frequencies.
- In Table 6, the signs of the initial and final values for k_float appear to differ. Please confirm whether this is intentional or a typographical error.
Citation: https://doi.org/10.5194/wes-2025-174-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 275 | 36 | 13 | 324 | 12 | 12 |
- HTML: 275
- PDF: 36
- XML: 13
- Total: 324
- BibTeX: 12
- EndNote: 12
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Very interesting and relevant research presented clearly and concisely. Although no new methods are proposed, state-of-the-art methods are used, with a new and valuable comparison of a few models and a demonstration of the impact of model choice on the optimisation process. Specific comments are included in the PDF attached. In general, the community will undoubtedly benefit from the new knowledge presented in this work.