[pmo]

I'm trying to model an oscillating flow (due to wave action) over a subsea structure at the seabed. The aim is to calculate the resulting hydrodynamic in-line/horizontal force and the vertical/lift force on the structure. The total forces will be depending on both fluid acceleration (inertia force) and velcoity. I'm running transient simulations where the horizontal velocity is set up to oscillate sinusoidally with time (oscillating period equal to wave period). The problem is single phase, and buoyancy is not included as the hydrostatic pressure can be assumed to be the same both over- and under the structure.

I've tried different boundary condition, and the one that seems to work best is opening with prescribed velocity (sine function) in both ends. My problem is that the amplitude of the oscillating pressure (and the hence the lift force) varies almost linearly with the length of the domain. Is this realistic? In that case; how should I set the boundary conditions to simulate my problem in a realistic way? I have attached a plot showing the pressure distribution at a time instant around the structure (simplified 2D-case). As can be seen in this plot, there's a pressure gradient in the flow direction, but the pressure is always constant (0 Pa) in one end.

Have anybody dealt with similar problems? Any advice would be appreciated.

[ghorrocks]

The linear pressure field is the outcome of your assumption of velocity boundary conditions.

I don't know much about sub-sea modeling, but I suspect the following:
1) The actual motion is a rotating motion to generate the wave. It will have a different pressure field to what you see there.
2) Whether the resulting flow on the sea bed is equivalent to your oscillating pressure is up to you to determine as it depends on how close the object is to the sea floor, water depth etc.
3) I suspect that a search of the literature will find a more accurate wave model. There has been some posts on the forum about wave models also, search for them, they might help.

[AliTr]

the hydrostatic pressure is varying across the domain(left->right) when a wave passing through the domain (left->right). I am not sure if this could be modelled in single phase.

[pmo]

Thanks for your comments.

1) I'm fully aware of the wave kinematics, but close to the bottom it's reasonalble to assume that the wave particle velocity is only horizontal. Since I'm only interested in the flow close to the bottom, I have modelled the flow as being horizontal and oscillating sinusoidally back and forth.
2)The hydrostatic pressure (buoyancy) is not included in my model (as explained above) since the hydrostatic pressure will be nearly the same on all sides of the structure.
3)I have seen a lot on wave modelling, but that's not really my concern.

[ghorrocks]

OK, so what is your question then? The pressure field you describe is a direct result of the velocity boundary you applied, and you say the velocity boundary is appropriate given your situation... Either the pressure field is appropriate or the assumption of an applied velocity field is not appropriate.