OceanFOAM 2015 abstract – Henrik Bredmose

“Wave-structure interaction in OpenFOAM. What is the role of the structural boundary layer? And how can we model it?”

The interfoam solver of OpenFOAM allows the modeling of free surface flows in 3D. At DTU it has been used for wave-structure interaction with large waves on monopiles [1-3]. For a steep, yet nonbreaking wave, these nonlinear wave forces are associated with a ‘secondary load cycle’ – a secondary force peak that occurs after the main force peak and before the time of minimum force. This is well-known from experiments. A closer inspection of the numerical results has shown that the secondary force peak is associated with two downstream vortices. They are created by the interaction of an upstream flow from the back of the cylinder and the outer wave flow which at that time is still in the main wave direction [1]. This leads to a moving separation point and production of vorticity.

Creation of vorticity requires viscosity. In the modelling setup, a simple free slip condition was used at the cylinder wall. Hereby, viscous effects can only be attributed to the internal fluid viscosity and numerical viscosity. Numerical experiments with grid refinement and (for preserved grid) inclusion of a no-slip condition did not change the results. This supports the suggestion that the vortices emerge due to separation in the outer flow.

Nevertheless, a full understanding of the flow phenomea and the role of the boundary layer requires a full solution of the viscous boundary layer flow. For wave motion at prototype scale, this is very ambitious as the Reynolds number is of magnitude 6×10^7. Further, the outer flow is unsteady. At a model scale of 1:40, the Reynolds number is 2.5×10^5 and here there might be a chance for a close numerical reproduction. While simplistic attempts has been made with LES modelling of steady boundary layers, detached eddy simulation (DES) seems a natural way to proceed. However, we have not yet defined a clear strategy for an accurate and eventually efficient way of computing the wave-structure interaction with inclusion of the boundary layer. Here fresh advice on the OpenFOAM capabilities for large Re, unsteady, free surface flows with a boundary layer will be welcome!

In the talk, the above results on steep wave flows will be presented and the modelling approach for the boundary layer will be put out for discussion.

[1] B.T. Paulsen, H. Bredmose, H.B. Bingham and N. G. Jacobsen (2014). ‘Forcing of a bottom mounted circular cylinder by steep regular water waves at finite depth’. J. Fluid Mech. 755 pp 1-34.

[2] S. Schløer, B.T. Paulsen and H. Bredmose (2014) ’Application of CFD-wave loads in aero-elastic computations’. Proc. 33rd Int. Conf. Offshore Mech. Arctic Engng. ASME.

[3] H. Bredmose and N. G. Jacobsen (2011) Vertical wave impacts on offshore wind turbine inspection platforms. Proc. 30th Int. Conf. Offshore Mech. Arctic Engng. ASME.