[pashazanousi]

I'm working on the wave run-up along a vertical wall (A rigid wall).
I have two options to replicate a piston-type wavemaker movement and generate the waves. The first one performed by specifying the piston displacement or the piston location as a function of time and The second one implemented by specifying the piston velocity as a function of time.
I simulated a numerical wave flume by using the first option IN ANSYS CFX and validated the results with Linear wave theory and the stork 2nd order theory.
I used the wall velocity option for the wavemaker boundary and enter U(t) for the U(x-component of the velocity) and the mesh motion with an unspecified option selected. But the wall didn't move according to the results.
In fact, I don't know how I should assign a normal velocity component to a wall in CFX?
Sketch of the wave tank
Wave Flume.png
The wavemaker velocity
Wavemaker.jpg
All of the above information are corresponding to this paper: https://doi.org/10.3390/w10080986
Thanks for your attentions

[Opaque]

The velocity of the wall is the derivative of the displacement with respect to time, correct?

If you integrate your velocity with respect to time, you should obtain the location of the boundary as a function of time.

You can then write a CEL expression as a function of time for Displacement X, Y or Z.

[pashazanousi]

Thanks for your reply...
Yes, I can use integration to determine the piston displacement, but I need an easy and straight way if it exists!

[Opaque]

You are not imposing the fluid velocity, but the wall motion.

Using a wall motion velocity BC will introduce discretization error while the displacement specification will be exact, correct?

[pashazanousi]

As I explained before, I inserted U(t) as the x-component of the wall velocity and no error occurred during analysis but in post process the results showed that the wall didn't move to produce the waves.
I don't know is it possible to define the piston displacement as the bellow expression?
Piston displacement=U(t) *time step

[Opaque]

I think I understand where you got confused.

Let us think it through to understand the fluid velocity setting:
1 - Is the relative tangential velocity of fluid respect wall zero? -> No Slip, and the fluid does not slide over the surface
2 - Is the relative normal velocity of fluid respect wall zero? Better be; otherwise, the fluid will separate from the wall.

The wall is the one moving into the domain, and it is accounted for a mesh velocity.

Velocity of the fluid at the wall = Velocity of the wall + Velocity of the mesh

On the specification of the wall displacement, sure you can play with

Displacement = U(t) * dt

Just keep in mind that is a first-order integration.

[pashazanousi]

Thanks for your reply again...
I think it will be work if the time steps considered small enough.
In my case, the angular frequency is low about 0.145. This small frequency led to a big excitation period(T) near 43 sec.
According to the paper, the time step must be less than T/400. Therefore, I considered a time step equal to 0.1 sec.
I don't know the time step is small enough or not but The CFX-Solver is running now. I will share the results here to talk about it more.
Thanks for your attention...

[pashazanousi]

Hi, it worked perfectly!
Two additional tips to achieve better results:
1. Specified displacement equation should be written as below:
Wavemaker Displacement=x+U(t)*dt
It should be noted that x refers to the local X-coordinate of the wavemaker position but X refers to the x-coordinate of the wavemaker initial position.
2.The time step plays an important role to achieve an accurate result. You can use a numerical integration calculator to choose the optimal time step.
a comparison of my results and the paper is shown below Images
Rigid Wall-Freesurface.jpg
Rigid Wall-Run-Up.jpg

[Opaque]

Glad to hear you got it working!!

I should mention that there are two different sets of coordinates in CEL expressions with mesh motion,

x --> the current position

Initial X, Initial x, initcartcrd x --> initial position of the undeformed mesh

Similarly for y, and z.