User Forum of Software BASEMENT

beroubt

User

Registered: 2015-01-05

Posts: 5

modelling outflow through gate with INNER_BOUNDARY

Hallo,

I want to simulate a computational domain with:
- Inflow through a segment of the boundary of the computational domain.
- outflow through a gate (not lying on the boundary of the computational domain)

Because the BOUNDARY Condition requires boundary edges I assumed solving my failure by using the INNER_BOUNDARY condition resulting in

-> The edge string 'gate_outflow' is not correct!
   Edges must be boundary edges!
   Please correct the edge string in the command file.
-> Domain: Simulation stopped due to errors.

Please help, what is may be going wrong.
It is a missunderstanding of INNER_BOUNDARY, which is described as "hydraulic structures like a weir or a gate within the computational domain"?

Best regards

Christian Volz

User

Registered: 2014-09-04

Posts: 31

Re: modelling outflow through gate with INNER_BOUNDARY

Hello,

in general it can be said, that outflows through gates/weirs can only be specified to boundary cells in BASEMENT. I think everything else does not make sense.
Yet, I suppose that this is no crucial limitation to your desired scenario. You can simply cut a "hole/gap" in your computational domain at any desired location. At the upstream front of the "hole/gap" you can place the string_def for a gate BOUNDARY. Then the water will flow out of your domain at this location. (At the downstream side of the "hole/gap" a wall boundary will be assumed if you do not specify anything else.)

The concept behind the INNER_BONDARY is slightly different, although it also bases on the idea to cut a "hole/gap" into the mesh. If you use a INNER_BOUNDARY, the water is expected to flow out of the domain at the upstream string_def and re-enter it again at a second downstream string_def. In your scenario, where the water shall not re-enter the domain(?), you can simply use a BOUNDARY condition as described above.

Hopefully this solves your problem?

Regards,
Christian

beroubt

User

Registered: 2015-01-05

Posts: 5

Re: modelling outflow through gate with INNER_BOUNDARY

Hello Christian,
thanks a lot - thats it.
Happy New Year!
beroubt.

Rossella

User

Registered: 2014-09-11

Posts: 8

Re: modelling outflow through gate with INNER_BOUNDARY

Hi,
I'm using BASEMENT V.2.5.2 R2107 and trying to use the HQ-inner-boundary condition. I defined two cross section (named BN1010 and BN1010_V) strings as STRINGDEFs; the two cross section are within the 2D grid. When I run the .bmc file I get the error (the same beiroubt wrote in the first post of the topic)

-> The edge string 'BN1010' is not correct!
   Edges must be boundary edges!
   Please correct the edge string in the command file.

What could I check to fix this error? Inner boundary conditions such as gate, weir, hqrelation have to be defined only on cross sections which are within the computational grid, isnt' it?
Thanks for helping

Rossella

Last edited by Rossella (2015-09-30 07:44:40)

Stephan Kammerer

Developer

Registered: 2015-03-31

Posts: 67

Re: modelling outflow through gate with INNER_BOUNDARY

Hi Rosella,

as Christian already mentioned above, in the current implementation, an inner boundary can only be specified along boundary edges. This means, you need a hole/gab in your mesh for an inner boundary structure (weir, gate or hqrelation). The figure below shows an example of a computational mesh with an inner boundary of type weir. The upstream stringdef 'weir_upstream' is defined as 'string_name1' in the BASEMENT command file and the downstream stringdef 'weir_downstream' as 'string_name2'.

Regards Stephan

Rossella

User

Registered: 2014-09-11

Posts: 8

Re: modelling outflow through gate with INNER_BOUNDARY

Thanks Stephan!
Now it's clear; reading the manual I didn't realize that the mesh has to have a hole to insert an inner boundary structure. I've just tried to run a simulation with an inner boundary condition of type "gate" and everything is ok.

Regards,

Rossella