open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear all,

I'm building a 2D coupled model between Telemac2D and Tomawac. I'd like to model the dike permeability in a port.

I've read that we can either model a dike through porosity and a drag force or with ground friction. Does this seem correct? What is the final difference between both approaches? The second one seems quite easy to use. However, in both cases, I have to calibrate a coefficient that I have no idea about (roughness or drag force).

Based on the drag force example, I've already implemented the CORPOR subroutine. However, I have a question about the DRAGFO subroutine. I've understood that the forces FUDRAG and FVDRAG are computed along the X and Y axes. However, I think I have to adapt the Fortran file to modify the direction of these forces. I've found some examples for a TEC application, like in the attached paper. I was wondering if anybody has already done this with a simpler example ?

Best regard

Adrien

Hello Adrien,

Yes indeed you need to change your FUDRAG and FVDRAG depending on what you want to apply on X and Y. On Joly et al. article they wrote down properlly FUDRAG and FVDRAG the way we usually do it for a turbine like (with upstream velocity etc). You probably wants to do something similar (i.e. hand writing ).

I attached a user_dragfo.f file that I use (a very simple modification of the one from the examples) where I can specify the orientation of a turbine relative to north (DIRN). There is 4 turbines (hence 4 INPOLY, 4 DIRN). The velocity used by dragfo is the velocity of the cell/node.

Have a nice day

Pierre-Henri

Hi Pierre-Henri,

Thank you very much for your answer. I think I can adapt and reuse your script, and specify the wave direction (it's a Tomawac model) for the ANGLE variable. However, in an ideal world, it would be better to specify the current direction at each node. I'm not sure if that's possible...

Have a nice day

Hello Adrien,

you can get the direction of the current with something like:
ATAN2( VN%R(I), UN%R(I) )
be careful it's velocity-in-y-direction (VN) first - double check doing stuff like: write(lu,*) "atan2(0,1)=", atan2(0.d0,1.d0)*180/PI

Have a nice day

Pierre-Henri

Hello Pierre-Henri,

Thank you again for your reply, I have implemented this equation and it seems to work perfectly ! I ran a test case on the dragforce example and everything is going as the original example.

However I'm performing additionals checks to make sure that everything is working properly.

I tried a case with only corpo subroutine (without dragforce) and I've set the porosity at 0.
I'm quite surprised by this result, I would have expect that any velocity goes through the dike. I observe some velocities and scalar flow rates through the dike even though out of the dike the velocity field seems correct…
Do you have any idea about this behaviour? I've attached the screenshot of the velocity field that I get and my corpo code.

Best regards

Glad it helps,

sorry but no I have never used corpo subroutine.