User Forum of Software BASEMENT

DDem

User

Registered: 2023-05-05

Posts: 1

Water level difference in linked (internal) boundary with h-q relation

Hi all,

I have a model that we simulate flood propagation for different discharges. As inflow and outflow boundary conditions (BC), I have uniform_in, and zhydrograph, respectively. For uniform_in, I used a staircase-type discharge file. For outflow BC, I defined a certain water depth. With these BCs, I ran a simulation and it worked without an issue. Please note that the time the discharge was changed was sufficient, which allows to have stabile flow conditions for each step. According to these results obtained, I created h-q relations for the bridges I have in my model.

In the second sets of runs, the same in- & outflow BCs were used. In addition, I used the h-q relation (linked) BC for the bridges. In the evaluation part of the results, I checked if the results match with given h-q relations and saw that it works fine. However, I also checked the water depths for the in- and outflow parts of the internal BCs and saw that the water levels are different up to 2 meters. I should emphasize that the holes that we generated to set internal BCs have width less than 2 meters. Plus, in the first simulation that we used to generate h-q relation, we placed stringdefs in the same exact location we put our internal BCs. Therefore, the mesh discretization is the same for both runs (except the inner part of the holes. Namely, I have mesh elements in the non-hole run, whereas I do not have any mesh elements in case there is no holes).

Summing up, I do not think that the 2m difference between the inflow and outflow is realistic since the holes that were created have a width less than 2 meters in the flow direction. Also, there is no energy dissipator or another water structure that may cause such difference in water level.

I would be very glad if you could provide your opinion about the possible causes of this problem, and a possible solution method.

Thank you very much in advance!

Br

Matthias Bürgler

Developer

From: Zürich

Registered: 2019-04-04

Posts: 150

Re: Water level difference in linked (internal) boundary with h-q relation

Dear DDem,

With the LINKED hq-relation boundary condition, there is no momentum transfer from upstream to downstream, only mass. My best guess is, that the difference in water level you observed is the result of water mass being transferred to the downstream linked boundary, which sort of 'piles up' since it has no momentum in flow direction and then first needs to accelerate. This may result in higher water-levels at the downstream boundary.

We will evaluate the implementation of the linked hq-relation boundary condition including a momentum transfer.

Best regards,
Matthias

Lukas Vonwiller

Developer

Registered: 2014-09-04

Posts: 127

Re: Water level difference in linked (internal) boundary with h-q relation

Dear all,

That's an interesting topic and discussion. For the linked hq-relation BC it would be great if the user could choose the BC type at the linked inflow (from the mesh perspective), e.g. uniform_in or froude_in.

Br, Lukas