the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 May 2026
OC7 Project Phase II: Comparison of Global-to-Local Load Transfer Approaches in Floating Structures
Abstract. Global-to-local load transfer remains a critical – yet largely unstandardized – step in the structural assessment of floating structures. This paper presents the results of package WP2.2 from the OC7 project Phase II, establishing a cross-industry benchmark for the workflows connecting global performance analysis (based on integrated loads analysis, ILA) and the local structural assessment (based on finite-element analysis, FEA). The study evaluates a spectrum of industry practices, including sequential approaches with the FEA following the ILA, fully integrated time-domain approaches with hydro-structural coupling, and simplified ILA-only approaches. Using the VolturnUS-S reference semi-submersible, the models were first harmonized through mass/inertia, static, and modal verifications. Structural responses were then compared across three primary scenarios: topside-only excitation, irregular waves, and combined wind/wave loading. The results establish a structured comparison framework, highlighting how specific modelling choices and load transfer techniques directly influence confidence in design processes. The findings offer practical guidance to reduce uncertainty in "global-to-local" design workflows.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Wind Energy Science.
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.- Preprint
(9370 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2026-78', Anonymous Referee #1, 12 May 2026
-
AC1: 'Reply on RC1', Michael Karch, 05 Jun 2026
We thank the reviewer for the positive assessment of the manuscript and for the constructive suggestions. Our responses to the individual comments are provided below.
1. Literature review
We agree that a broader discussion of previous work strengthens the manuscript. Please take into consideration that the topic addressed in this paper is relatively new to the floating wind industry with previous work and existing approaches lacking general applicability and a broader benchmarking study, or involving substantial model simplifications. Accordingly, we expanded the introduction to include additional discussion of prior scientific work on global-to-local load transfer and hydroelastic modelling, including relevant recent studies.2. Clarification of Sect. 5.2.2 (“Modal FE analyses”)
Section 5.2.2 was revised to state explicitly that the reported modal FE analyses are dry modal analyses performed in the structural FE models without hydrodynamic added mass effects.3. RNA peak load frequency and relation to eigenfrequencies
In the revised manuscript in Section 7, we clarified the relevant RNA excitation characteristics and added a comparison between the 3P excitation frequency and the first fore-aft bending eigenfrequency of the hull-tower system. This was included to better explain the likelihood of resonance or near-resonance effects in the flexible models. We conclude that significant 3P excitation of the tower is not expected.4. Expansion of conclusions and future recommendations
We appreciate this suggestion. However, in order to keep the manuscript focused on the scope of the present benchmarking study and to avoid introducing recommendations that go beyond what is directly supported by the results presented, we have not expanded the conclusions further in this direction. For example, some modelling simplifications were introduced to allow more institutions to participate (e.g. exclusion of ballast, exclusion of second-order wave loads) which means some aspects were not considered in this study but some tools and workflows applied could model it in general. A dedicated experimental campaign as basis for more validation work is generally supported, though, developing it was not part of the scope for the WP2.2 working group within OC7.We hope that these revisions and clarifications adequately address the reviewer’s comments.
Citation: https://doi.org/10.5194/wes-2026-78-AC1
-
AC1: 'Reply on RC1', Michael Karch, 05 Jun 2026
-
RC2: 'Comment on wes-2026-78', Anonymous Referee #2, 20 May 2026
This manuscript presents a comparative study of numerical approaches for load-to-stress analysis of floating offshore wind turbines (FOWTs). Using the same configuration, different approaches were first calibrated in terms of model properties and rigid-body motions, and were subsequently applied to compare stress responses within the FOWT under various loading conditions.
The key findings include:
- Structural dynamics can influence stress responses and, subsequently, fatigue damage estimation. The influence varies under different dominant loading conditions, such as wind- or wave-dominated cases.
- The consideration of nonlinear wave pressure introduces additional loads into the floater depending on the floater geometry, thereby affecting the stress responses compared with the linear wave-pressure assumption based on the mean water level.
- The unit-load reconstruction approach demonstrates accuracy comparable to that of the direct calculation approach, indicating its potential for future application in industrial practice.
- In addition, the authors highlight the importance of global load calculation, as it provides a fundamental basis for global-to-local load transfer approaches. The accuracy of the final local stress calculation is highly influenced by the global load assumptions and their accuracy.
It is evident that the objective of this study is not to identify the most accurate approach, given the limited availability of validation data, but rather to investigate key factors that may affect load-to-stress analysis and help reduce associated uncertainties. The comparisons are carefully designed, clearly documented, and well presented. The observations and discussions are meaningful and provide a valuable basis for future research.
Therefore, this manuscript is recommended for publication. Nevertheless, some refinements to the discussion of the results may help readers better understand the key findings. The following aspects could be further considered:
- The stress comparison figures in Section 7, including Figures 11, 13, 14, 15, 16, and 17, contain a substantial amount of useful information. In the statistical plots shown in the middle of these figures, the use of colours could be reconsidered to distinguish different modelling assumptions or techniques, such as rigid versus flexible models, linear versus nonlinear wave pressure, or shell versus beam modelling. This may provide a clearer visual comparison of the effects of different modelling choices. At present, the colours represent different model IDs, which are already distinguished by their positions in the figures.
- The authors provide observations and discussions for each loading case. It would be beneficial to include a summary subsection, for example Section 7.4, to synthesize the main findings. In particular, the variations in fatigue damage index under different modelling techniques and loading conditions may indicate conservative or unconservative estimations arising from different sources. A concise table or summary paragraph outlining the trends associated with the selection of different modelling techniques would better guide future research and industrial applications.
- The authors observe that the rigid and flexible models exhibit different trends in fatigue damage index when aerodynamic loads are considered in addition to wave loads. Furthermore, when the wave height increases from LC6.X to LC7.X, the trend for the flexible model appears to change. This behaviour may be related to nonlinear hydrodynamic effects. Considering the increasing size of FOWTs and the growing need to account for floater flexibility, this observation is highly relevant to future studies. Further discussion or additional information of this point would be valuable.
Overall, the manuscript addresses an important topic in the numerical analysis of floating offshore wind turbines. The work is well organized and provides meaningful insights into the influence of modelling choices on load-to-stress analysis.
Citation: https://doi.org/10.5194/wes-2026-78-RC2 -
AC2: 'Reply on RC2', Michael Karch, 05 Jun 2026
We thank the reviewer for the positive assessment of the manuscript and for the constructive suggestions. Our responses to the individual comments are provided below.
1. Colour coding in the statistical plots of Figs. 11, 13–17
We appreciate this suggestion and carefully reconsidered the presentation of the statistical plots in Section 7. However, because the compared models differ simultaneously in several modelling aspects (e.g. rigid vs. flexible hull, linear vs. non-linear hydrostatics, software/toolchain, and load transfer methodology), a revised colour scheme intended to highlight these categories did not lead to a clearer overall presentation. On the contrary, applying multiple colour groupings within the same figure reduced readability, while changing colour logic between figures depending on the aspect discussed was found to be potentially confusing for the reader. We therefore decided to retain the original colour scheme, in which colours identify the individual model IDs consistently across all figures. To make it easier for the reader to find the key conclusions, also involving the statistical plots, we added a summary in Section 7.4.2. Summary subsection synthesizing the main findings
We agree that a synthesis of the main observed trends improves the readability of the results section. Accordingly, we added a new summary subsection, Section 7.4, in which the main findings from Sections 7.1–7.3 are consolidated. This section highlights the observed trends associated with the different modelling techniques and loading conditions and is supported by a summary table to provide a concise overview for the reader.3. Further discussion of the differing trends in fatigue damage index and possible non-linear hydrodynamic effects
We agree that this is an important observation. To better investigate the underlying root causes, we extended the study by incorporating additional Ramboll model variants with non-linear hydrostatics, namely RAM1a (rigid hull) and RAM25a (flexible hull). A short description of how hydrostatic non-linearity is considered in these new models was added to Section 2. In addition, the naming of the Ramboll models was revised (see Table 1) to improve readability and facilitate comparison. These additional results allowed us to refine the discussion in Section 7 and to draw further conclusions regarding the isolated influence of hydrostatic modelling assumptions and hull flexibility. At the same time, the extended comparison still did not allow us to clearly isolate or confirm the influence of non-linear hydrodynamic effects (i.e. the Froude–Krylov pressures applied to the instantaneous wave surface) in LC7.X. We therefore revised the discussion to reflect this limitation explicitly. We also added that one remaining possible explanation in Section 7.3 for the inconsistencies observed in the calculated FDIs in LC7.2/7.3 is the difference in FEA load equilibration techniques. While this effect could not be proven conclusively within the present benchmark, we believe it may also contribute to some of the observed differences.We hope that these revisions and clarifications adequately address the reviewer’s comments.
Citation: https://doi.org/10.5194/wes-2026-78-AC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 492 | 161 | 28 | 681 | 19 | 17 |
- HTML: 492
- PDF: 161
- XML: 28
- Total: 681
- BibTeX: 19
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The paper represents an important step toward addressing a critical gap in the literature concerning global-to-local load transfer methodologies for floating offshore structures. The authors tackle this problem through a comprehensive comparison of the main approaches currently adopted in academia and industry.
The study analyses three principal modelling strategies. The first is the sequential workflow, which is generally considered current industrial practice, where the global integrated load analysis (ILA) is performed first and the resulting loads are subsequently transferred to a structural finite element model (FEA). The second is the integrated workflow where the global analysis and the structural FEA are solved through a strongly coupled hydro-structural interaction. The third is a simplified global analysis approach based on beam-element modelling.
The comparison is carried out using stresses evaluated on selected structural panels as the primary metric. The assessment includes comparisons of time histories, mean values, standard deviations, and fatigue-related indicators obtained through rainflow cycle counting and simulated fatigue damage evaluation.
The numerical results reinforce several important conclusions: (i) they highlight the critical importance of accurately predicting the global dynamic response, since any inaccuracy at the global level inevitably propagates into the local structural assessment; (ii) the study demonstrates the relevance of hydroelastic coupling effects: hull elasticity contributes significantly to the structural response, and rigid-body assumptions may lead to non-negligible deviations; (iii) the simplified beam-element approaches appear to have limited applicability for detailed structural assessment, especially in high stress-concentration regions like junctions; (iv) load superposition methods based on pressure linearisation provide results reasonably consistent with those obtained from fully coupled time-domain structural simulations.
The main limitation of the paper, however, is the absence of experimental validation. The conclusions are based entirely on numerical cross-comparisons, without comparison against experimental measurements of panel stresses or other physical reference data. As also acknowledged by the authors, this prevents a direct assessment of the accuracy of the investigated methodologies and partially limits the scientific strength of the conclusions.
Nevertheless, despite this limitation, the paper provides a valuable and timely contribution to the field. The breadth of the benchmark exercise, the quality of the comparisons, and the practical relevance of the findings make the work scientifically significant and deserving of publication.
The paper may be accepted in its current form. However, I would like to offer the following considerations that, if deemed relevant by the authors, could further strengthen the manuscript: