Simulation of wave impacts at BELWIND offshore wind farm and comparison with full-scale measurements
Bunnik, T.; Weijtjens, W.; Devriendt, C. (2017). Simulation of wave impacts at BELWIND offshore wind farm and comparison with full-scale measurements, in: Proceedings of the ASME 2017 36th International Conference on Ocean, Offshore and Arctic Engineering (OMAE2017): Volume 10: Ocean Renewable Energy. pp. 1-7 In: (2017). Proceedings of the ASME 2017 36th International Conference on Ocean, Offshore and Arctic Engineering (OMAE2017): Volume 10: Ocean Renewable Energy. ASME: New York. ISBN 978-0-7918-5778-6. , more |
| Authors | | Top | - Bunnik, T.
- Weijtjens, W., more
- Devriendt, C., more
| | |
| Abstract | The effects of operational wave loads and wind loads on offshore monopile wind turbines are well understood. For most sites, however, the water depth is such that steep and/or breaking waves will occur causing impulsive excitation of the monopile and consequently considerable stresses, displacements and accelerations in the monopile, tower and turbine. At Belwind offshore wind farm (offshore Zeebrugge, Belgium) the waves and accelerations of a Vestas V90 3MW wind turbine have been monitored since November 2013, using wave radar and several accelerometers. During this period the wind turbine was exposed to several storms and experienced several wave impacts, resulting in vibrations in the monopile. The measurements were compared with results from a numerical model for the flexible response of wind turbines due to steep waves. Previously this model was compared with scale model tests with satisfying results. The full-scale measurements provide an additional cross-check of the model. The numerical model consists of a one-way coupling between a CFD model for wave loads and a simplified structural model based on mode shapes. An iterative wave calibration technique has been developed in the CFD model to ensure a good match between the simulated and measured incoming wave profile, obtained with the wave radar. This makes a deterministic comparison between simulations and measurements possible. This iteration is carried out in a 2D CFD domain (assuming long-crested waves) and is therefore relatively cheap. The calibrated numerical wave is then simulated in a 3D CFD domain including a (fixed) wind turbine. The resulting wave pressures on the turbine have been used to compute the modal excitation and subsequently the modal response of the wind turbine. The mode shapes have been estimated from the measured accelerations at the Belwind turbine. A grid refinement study was done to verify the results from the numerical model. The horizontal accelerations resulting from this one-way coupling are in fair agreement with the measured accelerations. |
|
