Development of a numerical wave flume in FLOW-3D to model the interaction of a steady current with a monopile, and validation of the model with experimental results
Naeyaert, T. (2014). Development of a numerical wave flume in FLOW-3D to model the interaction of a steady current with a monopile, and validation of the model with experimental results. MSc Thesis. Ghent University. Faculty of Engineering and Architecture: Ghent. 143 pp.
|
| Beschikbaar in | Auteurs |
|
Documenttype: Doctoraat/Thesis/Eindwerk
|
| Trefwoorden |
Bed shear stress
Computational fluid dynamics
|
| Author keywords |
FLOW-3D; CFD; Steady currents; Turbulence modelling; Monopile |
| Auteurs | | Top |
- Naeyaert, T.
- Van Oyen, T., revisor, meer
|
|
|
| Abstract |
This master thesis presents a numerical study of the interaction of a steady current around a monopile. Especially velocity and shear stress results are investigated closely. An - turbulence model is developed in the CFD software package FLOW-3D and used to validate a specific experimental test case, acquired in the literature.
The subject is situated by performing a literature study. Further introduction contains the mathematical basics of fluid flow and turbulence modelling, both with the emphasis on their implementation in FLOW-3D. The classical model used to describe open channel hydraulics is presented, and linked to FLOW-3D, notably on the subject of boundary conditions.
A first basic model with periodic boundaries is developed with special attention to the effect of the cell size adjacent to the wall, and the effect of turbulence model parameters (type and maximum turbulent length scale).
This basic model is modified to correspond with the experimental test case. This is done step by step to trace possible simulation problems. Special attention is given to the modelling of a free surface (use of different VOF advection methods). A pile is introduced in the model and the velocity and shear stress results are validated. Poor agreement is met, since the velocity wake behind the pile reoccurs at the model inlet (due to periodic boundary conditions), causing an important underestimation of the actual velocity profile.
A second basic model is then developed, adopting a volume flow rate at the inlet and a hydrostatic pressure boundary at the outlet. Physical coherence is investigated, and the effect of turbulence model settings and the cell size at the bed is revised. A pile is introduced and again, velocity and shear stress results are validated against the experimental data. Good agreement is found qualitatively for the downflow and the horseshoe vortex in front of the pile. The wake field of the pile isn’t physically coherent due to a steady-state calculation of the flow field. Quantitatively, velocity and shear stress results are acceptable in the run-up to the pile, except right in front, where the limitations of the applied - turbulence model come to the surface. |
|
