open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello all.
I've set up a simple model using Matisse v2p1 and telemac2D v6p0. Everything seems to be fine, yet I keep getting the following message at each calculation time step:

---

ILL-POSED PROBLEM, ENTERING FREE VELOCITY
FOR EXAMPLE AT BOUNDARY POINT NUMBER 48

---

I'm attaching a zip file with the project in case someone can have a look.
Cheers.
Roberto

Hello,

This message only indicates that you have a local inflow at a boundary with free velocities. This can happen e.g. when von Karman eddies created behind a bridge pier get out of the model. This is also the case on boundaries where a tide is given by prescribed elevation and free velocities. If it works it is OK but in that case there may be an infinity of solutions, or very large velocities entering the domain (as you have free velocities this is also a solution of Saint-Venant equations).

When a computation crashes because the problem is not well posed, possible solutions are :

* using Thompson condition
* putting a small head-loss at the exit (it works well in 3D)

With best regards,

Jean-Michel Hervouet

Hello

I am also experiencing the same problem. I am modelling a 33km stretch of river for flooding with the banks set a solid boundary (sliding code 2,2,2,2 for H,U,V,T) in Matisse. An open boundary as been specified

Upstream: code 4 (H), code 5 (U,V,T)
Downstream: code 5(H), code 4(U,V,T)

The option for liquid boundaries has been specified as 1 (usual free or prescribed values) and with the required liquid boundary conditions (Q(1) m3/s , SL(2) m file provided.


It appears this notification disappears once the flux in both boundaries equalises.

I wanted to know if this can adversely affect my results.

thank you.

Meshack Keabetswe

Hello,

Generally in a river this message has no consequence, it only means some inflowing velocities at the exit, due e.g. to a von Karman eddy exiting. But such situations can lead to infinite inflowing velocities (that are still a solution since you say that the velocity is free). If the notification disappears there is no problem, if it stays and that you see strange recirculations at the exit it is a problem.

With best regards,

Jean-Michel Hervouet

HI JMH,
I have the same probleme but on a solid boundary (right river bank). It's realy strange and I do not see how we can explain a FREE ENTERING VELOCITY on a node from a hard shoreline.
Regards

Hello,

You certainly do not have such a message for a point on a solid boundary. If you see a velocity which is not exactly with U.N=0 on a solid boundary it is normal since the impermeability condition is enforced in weak form (so mass conservation strictly obeyed but some velocities may go across solid boundaries, and are not taken into account).

With best regards,

Jean-Michel Hervouet

YES JMH, you're right this ill-posed problem is just happened near the outline (solid boundary), not exactly on it. May be the DEM is not smooth enough and supercritical flow appears in such node. I have to do some corrections on the DEM and then see what will happen.
Thank you Mr JMH

Hello, I have the same issue with Mourad. The "ILL-POSED PROBLEM, ENTERING FREE VELOCITY" appears near my solid boundary as well. Are there any solutions about this issue?

Thank you,
Georgia

This message appear only on liquid boundary, not on solid boundaries.
Could you be more precise in the description of your case/problem in order to let us know how to give you appropriate advises
Regards

Thank you for your reply.
The problem appears in my liquid boundary (544). I have a northwest initial and boundary main direction 1 in tomawac and the problem appears in the last node of my liquid boundary (southeast) before the closed boundary
I attach the velocity on that node when the warning appears at the begginig of the simulation.
Below there are some of my choices from my steering files.
The simulation runs without problems but i keep getting this warning message and i dont know if my results are influenced.
I have tried other initial conditions such as constant elevation but the warning is always there.

Best regards,
Georgia