open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello community
I have to model a lateral rectangular water intake in a river reach (diversion channel). The purpose is to ensure that the diverted discharge belong to the predifined range of discharge.
I have Q (t) upstream and h (t) downstream (without any lateral intake as indicated in the top figure of the attached).
The lateral water intake is modeled as a rectangular breach (see attached).
The downstream hydro condtions, formerly known, are modified and therefore they are unknown. Can I provide a free conditions (H or Q) at the breach and at the downstream boundary?
Can telemac 2d and 3d handle this type of problems?
Be sure that any kind of help will be appreciated
Regards
Mourad

Hello Mourad,

Problem 1: Yes TELEMAC-* handle this kind of problems.
Problem 2: Yes, but you can't set free H everywhere. You can try to set your downstream condition as far as possible from your 'lateral intake'.

best regards
J

Thank you for the replay,
However I can't understand the meaning of a prescribed variable (H or Q) at any dowstream end (in this case). I mean my goal is to check, with telemac 2d or 3d, wether a lateral breach (rectangular L * h) within the right river bank helps to derive more or less 2 m3/s or not. Physically I can't impose H or Q at the 2 downstream ends.
You said that TELEMAC resolve this type of problem. can you give more details about that, an example ?
Thanks again Jacques
Mourad

Hello,

You can't impose free depth everywhere. If the downstream (set H) condition is far from your breach you can assume that the influence of H at the downstream is negligible.

To illustrate with an extrem case: you can connect your chanel/river to the see and impose the see level.

Best regards,
J

Dear Jacques,
1- I did'n say "free H everywhere", but only downstream the channel and the water intake. This is physically true because you can't impose the water depth there (h is uknown : no measurments).
2- I know since 2009 that telemac needs two hydro conditions typically Q (upstream) and H (downstream) for free surface flows modeling in rivers. Typically we use the hydro gauges (measurments) tu provide data for boundary conditions.
3- This is not the case of a channel connected to a large basin "with constant level...", that's it!
4- Telemac deals with boudary problems : becare to force boundary conditions to adapt them to a specific cfd suite. you can juggle with initial conditions but not with the former
5- If you lack data downstream, you can then prescribe a free condition with Flow3D suite... I was just thinking that the newel versiosns of telemac (post V6p3) may handle this kind of problems.
Thanks again for your suggestion.
Mourad

Hello Mourad,

You said, H_intake and H_near_dowtsream are unknowns. That’s why extend the computation domain sounds good to me, to consider H_intake and H_near_downstream in the computational domain instead as liquid boundary conditions.
I assumed your flow is subcritical. If you do not have any elevation measurement at the downstream of your computational model, you lose the information going from downstream (outside the computational model) to upstream (in your computational model) but if you want, of course, in TELEMAC you can impose free depth, free velocity at your downstream liquid boundary condition (even if critical of subcritial).

Best regards,
J

Dear Jacques,

You're right !

- The water intake dosen't exists, neither the diversion channel.
With 1d computations I've obtained an approximative geometry of the artificial opening in the right bank (artificial breach) to derive a known q in the future diversion channel for a given Q upstream of the river.

- Now I've to check up the 1d computations with T2d and T3d.

- Since my first use of Telemac (2009), I've never prescribed free H and free Q (or U) at the same boundary. Worse, I'm not even aware that this is possible. Sorry for that.

Cheers
Mourad