[Luke92]

Hi,

i am trying to simulate ocean waves (Stokes 2nd Order) in a 2D wave tank. The method I tried out was based on specifying the water particel velocity and the volume fractions of water and air at the inlet. The horizontal and vertical velocities were given from the Stokes 2nd Theory defined with CEL expressions. At the outlet a static pressure Outlet BC was chosen with an expression that gives the relative pressure as a function of the volume of water in the fluid region and the vertical distance (Z direction value). The pressure decreases when the value of z increases and becomes 0 when z reaches the interface between the air and water.

As long as i use Outlet BC at the outlet, i get results very close tot he theory. However, during the solution after every timestep i get the warning message: "A wall has been placed at portion(s) of an OUTLET" until the Simulation crushes. So there is a supressed backflow at the outlet.

Then i tried to use a Pressure and Direction Opening at the Outlet and specified the relative pressure and the volume fraction for the air and water. In this case, I get no more error messages, but there is a backflow at the outlet, through wich my water level from the outlet to the inlet continues to rise.

On the uploaded images you can see my mesh and time charts of the surface elevation at the inlet for both cases. For the second case, I also uploaded a photo of the SurfaceElevation at the Outlet.
The last picture shows my BC's for the second case (Opening BC at the Outlet) and the Velocity in wave Propagation direction (X).

I have tried many different variations of my case but i dont know how to stop these flows at the outlet without using an outlet-BC. Are these flows at the outlet related to my velocity BC at the inlet?

Finally, my dimensions of the wave tank and my main wave parameters:
Wave Tank:
Length: 10 [m]
Hight: 2.5 [m]

Depth: 1 [m]
Wave Height: 0.044 [m]
Wave Period: 0.894 [s], (Timestep = Period/200)
Wave Length: 1.25 [m]
Still water Level: z= 0[m]

Thank you in advance
Lukas

[ghorrocks]

This is side-stepping your question a bit, but I know some researchers using CFX for ocean wave modelling put a region of coarse mesh at the outlet end, connected to the main mesh with a GGI. Then you can put your outlet on the coarse mesh (or even just make it a wall). The coarse mesh region causes dissipation in the wave and absorbs the wave, meaning that little wave energy is returned into the domain as a reflection and only a small wave is transmitted to the outlet boundary.