BASEMENTBasic Simulation Environment for computation of environmental flow and natural hazard simulationLaboratory of Hydraulics, Hydrology and Glaciology (VAW)ETH Zurich |
You are not logged in.
Hello everyone,
I am currently using BASEMENT (Version 2.4 R1692) for a flood Event Simulation. At some point in the model, I needed to simulate a few bridges and a pipe conduit, which I did by using the inner_Boundary module with the type 'gate'. Each String beeing one side of the bridge, respectively one side of the conduit. 'my' set to 0.6. I used QGIS, with the basemesh addon to create my model.
When running the simulation I am experiencing numerical oscillations (with velocities reaching from 1 m/s to more than 200 m/s) on some triangles at both 'gate' Strings as soon as water reaches the conduit. I tried to simplify the situation of the entry and exit String, in order to eliminate errors due to a too complicated geometry, but the problem remained.
Have you already had a similar problem, and/or do you have a solution? Or is there a better way to simulate such a situation?
Thank you in advance!
Attachement:
Situation:
http://img4web.com/view/26FYWT
conduit start:
http://img4web.com/view/GUDTLT
conduit end:
http://img4web.com/view/XF7VLX
additional Information:
http://img4web.com/view/RJKQJE
Offline
Hello,
actually, I am not aware of such problems concerning the inner gates in the 2D model. A pre-requisite for the inner gate is that your geometries of both string-defs should be very similar. The edge lengths should be about the same. Also, the z-heights of the corresponding edges should be similar or equal, because the average z-value of both sides is taken as the gate bottom in the gate formula. I suppose that you have larger deviations in the z-values and the gate formula therefore produces invalid results, leading to these oscillations. Please check if these conditions are met.
Offline
Indeed, the height difference seems to be a problem.
Any ideas, what I could use instead?
Thank you!
Offline
Hi kost-partner,
I think in your case it is important that you estimate the maximal discharge possible in your pipe conduit!
Then you can use an EXTERNAL_SOURCE in your river (with the estimated discharge of your pipe conduit).
If the flooding on the upper end of your pipe conduit is of interest, then you could use another EXTERNAL_SOURCE (with the excess water).
Hope this helps!
Offline
Hello,
in addition to the possible workaround described by Lukas, I think it would be nice to enhance the inner gate to be able to cope with such different z-elevations. This situation may be quite common for conduits and therefore should be handled by the program. If it is possible, it would be nice if you can send us your model data. Then we can try how to improve the behaviour of the inner gate for such scenarios.
best regards,
Christian
Offline
Hello once again
we have implemented a new type of inner boundary to be able to cope with such scenarios. This new 'hq-relation' inner boundary lets you specify a hq-table, which is used to determine the flow discharges through the culvert/conduit depending on the upstream water level. In contrast to the inner gate, the boundary does not depend upon the z-elevations and, hence, there should be no problems even with large differences in the elevations. It is a simple approach and still in a test phase, but if you want to use this feature, then let us know.
Regards,
Christian
Offline
Hi Christian
With the HQ relation option of the inner boundary (for modelling a culvert) do you still need to create the "hole" between the inlet and outlet strings. I want to be able to discharge the culvert over a long distance and it would be impractical to insert this many holes as it would affect the 2D model results.
Offline