open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi!

I work with T2D and T3D on the same study area in which there are bridges with complex foundations (piles from a concrete slab down to the river bottom, creating a pile net under the slab).

I model this in 2D using dragfo subroutine where FUDRAG and FVDRAG are computed.
As stated at the beginning of the subroutine, I understand that the source term added in the momentum equation is FUDRAG * U, U being the norm of velocity vector.

If I am right, I would then define my drag force in 3D (on all vertical nodes of a same 2D point) using subroutine source.f, in which I would define an implicit source term S1U inspired from 2D as S1U = FUDRAG * U = (-0,5*N/AIRE*DIAM*CD*U) * U, and idem for S1V...

Would that be correct?

Thanks for your help!

Best regards
PL

Hello,

Yes, the momentum equations are both DU/DT=... but in 3D we have :

DU/DT + S1U*U =...

and in 2D if I remember well:

DU/DT +S1U*U/H = ...

So both terms have not the same dimension in 2D and 3D.

Regards,

JMH

Hello,

Thank you for your reply.

But if I understand you right, what should be computed in source.f is ONLY:

S1U = -0,5*N/AIRE*DIAM*CD*U

and NOT:

S1U*U = -0,5*N/AIRE*DIAM*CD*U**2

Is that right? S1U should then be multiplied by U in a next step (which subroutine?).

Thanks in advance!

Best regards
PL

Hello,

Yes, there is no subroutine computing S1U*U, as it is an implicit term, S1U is just introduced in the matrix of the final linear system.

Regards,

JMH

Hello Mr Hervouet,

I am currently trying to implement tidal turbines in my 3D flow model. I have seen this successfully done using 2D depth averaged model, where the drag.fo file was edited to introduce the drag and thrust force.

My question is, why can't the same drag.fo subroutine be used to introduce the drag and thrust term in a 3D model?

And why source.f subroutine is needed to introduce the source term in the model?

Thank you for your insight

Regards
Anas

Hello,

I am not at work and can communicate only exceptionnally. In dragforce.f we deal with depth-averaged shallow water equations and the theory is slightly different because the equations are obtained with a division by the depth. In 3D there is no reference to the depth and the dragforce is a mere head loss, source.f is just the name of the subroutine were we add the explicit and implicit source terms, and a head loss is treated as an implicit source term.

With best regards,

Jean-Michel Hervouet