[Qunbin]

Dear all,

I'm trying to simulate a solitary wave propagating over a trapezoid bar with relaxation methods, but the solitary wave seems unstable when it leaves away from the relaxation zone.

I have added the control files and two instants wave profiles, please find in attachment files.

Any help on this problem is appreciated.

Cheers,

Qunbin

PS: REEF3D Version used here is 21.09

control.txt

ctrl.txt

1.png

2.png

[IADInocencio]

I'm still running something on my laptop (also still learning) so I can't do a parallel run, but my guess is your waves is much too large for your domain. The wave height is 0.13m but your domain is only 0.4m and a water depth of 0.2m. It's likely shoaling already in the 0.2m depth, causing the wave to hit to top of the domain as shown in your second screenshot.

If it doesn't shoal there, it will definitely will once it reaches the berm, hitting the top of the domain

Try either increasing your domain z axis or starting off with smaller waves.

[Qunbin]

Hi Incoencio,

Thanks for your comments.

Indeed, the length of z domain influences the results. Now the z domain has been increased to 1.2 m, i.e., 6 times the still water depth. The simulation results is better than the previous, but it is still has the phenomena as I shown before. Wave height of the solitary wave are also much larger than the input wave height (0.13 m), see the attachment figures.

Best,

Qunbin

[Qunbin]

I also tried to using Dirichlet boundary condition to generate solitary waves. The solitary waves can propagate well, but its wave height near the boundary of wave generation is much higher than the given value in the ctrl.txt file. I don't know what causes this problem.

I would really be grateful if someone could help me out with this issue.

Qunbin

[kamath]

Like Ismael said above, the water depth here is quite small. This will result in a strong shoaling effect on your very high wave, resulting in an almost breaking wave-like form.

You addressed one aspect: height of the tank. Here, I recommend water depth x 2 in general cases (eg. d=0.5, H=0.1 m, zmax= 1 m is ok).
In your case, d=H. This is quite high! Remember the breaking wave limit is Hb≈0.77 d.

I advice using inputs and domain setup accordingly.

[Qunbin]

Hi Arun,

Thanks for your quickly reply. Indeed, the present wave condition is a extreme wave condition, and it might a challenge to simulate it.

The reason why I still used these wave parameters is that I want to reproduce a published work on an experimental study of plunging solitary waves over a trapezoid bar and impacting on a vertical cylinder. Now I just test in a 2D situation (and without the vertical cylinder). I want to verify whether the solitary waves can be simulated well.

https://www.sciencedirect.com/scienc...29801820302493

Best

Qunbin

[Qunbin]

Hi Arun，

I did a simulation with a small wave height of solitary waves, where water depth d = 0.2 m and wave height H = 0.04 m. This solitary wave can be generated well although its wave height is a bit higher than the input wave height. But H = 0.13 m as used in the experiments is still not work.

best,

Qunbin

[IADInocencio]

Have you tried increasing the depth a bit more? 0.2m Still looks quite shallow for 0.13m wave height. Or was that also the depth of the study you are referring to?

(Side comment, wow 32 cores)

[valgrinda]

Hi Qunbin,

a couple of comments:

- wave height to water depth ratio is towards the extreme end and will result in an unstable soliton

- the mesh is not optimal: the wave tank is 1.2 m, whereas the water level is at 0.2m. Maybe try 0.5m tank height? Then you could run a finer grid to check convergence.

- I would turn off turbulence for this case

- if you have experimental data, maybe try the wavemaker signal as input

[Qunbin]

Originally Posted by IADInocencio

Have you tried increasing the depth a bit more? 0.2m Still looks quite shallow for 0.13m wave height. Or was that also the depth of the study you are referring to?

(Side comment, wow 32 cores)

Hi Ismael,

Thanks for your comments.

Yes, I have tried a more deeper water depth and with less wave height. The simulation result is ok, but that is not what I want. I want to reproduce the experiments and compare with the experimental data. The experimental setup as above mentioned is of water depth d = 0.2 m, and with three wave height, i.e., H = 0.11 m, 0.12 m, and 0.13 m.

Cheers,

Qunbin

[Qunbin]

Originally Posted by valgrinda

Hi Qunbin,

a couple of comments:

- wave height to water depth ratio is towards the extreme end and will result in an unstable soliton

- the mesh is not optimal: the wave tank is 1.2 m, whereas the water level is at 0.2m. Maybe try 0.5m tank height? Then you could run a finer grid to check convergence.

- I would turn off turbulence for this case

- if you have experimental data, maybe try the wavemaker signal as input

Hi Hans,

Many thanks for your comments.

--Indeed, the ratio of wave height to water depth (H/d = 0.65) is approaching to extreme conditions. So it is hard to simulate.

--For this cases, the length of z domain will influence the simulation results. I have tried to using z = 0.4 m, 0.6 m, 0.8 m, 1.2 m. I found z = 0.8 m is ok, but now I use z = 1.2 m just want to avoid the influence induced by the upper boundary. I have carried out a convergence test with dx = dz = 0.04 m, 0.02 m, 0.01 m, and 0.005 m. For a small or medium wave height of solitary waves, the simulation results is convergence when the grid size is less than 0.01 m. But for high wave height as I want to simulate, the solitary waves is still having some issues once it leaves away from the relaxation zone.

--Turning off/on the turbulence model almost has the same results.

--I don't have the experimental data, but I generate a wavemaker_eta.dat file as wavemaker signals through a 3-order theoretical solution of solitary waves (Grimshaw, 1971). It can also generate a solitary wave. And I found it is the best solution for the present wave parameters although wave elevation near wave generation boundary is less than 0.13 m, see the attachment figure.

Thanks for your help again.

Best

Qunbin

[Qunbin]

Sorry I forgot to including the wavemaker_eta.dat file in my previous reply.
Note that you need to change .txt to .dat as the *.dat file cannot be uploaded directly.