Re: [Anuga-user] How to model a bridge

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

Hi Stephen,

due to your tip I changed two things:

First I have placed the riverwall in the middle of the culvert inlet
lines.

Second I have set the invert elevations explicitly.

This lets the results appear more realistic.

The problem that still remains is that I can't figure out the behavior
of the river wall.
Is there a timestepping_statistics function to check the
flux at the river wall?
It's not really clear to me how the riverwall is setup in the
shallow_water domain.

As far as the culvert operator is concerned: Maybe it tends
to produce more backwater as there really is?

Cheers
Stefan

Am Sat, 24 Apr 2021 09:42:55 +0000
schrieb Stephen Roberts <ste...@an...>:

>
> Hi Stefan,
>
> You should have a barrier of some sort, riverwall or raised elevation.
>
> The riverwall (or the raised elevation) should be a little distance
> from the inlet and outlet.
>
> You should also setup the size of the apron and enquiry_gap.
>
> See attached diagram.
>
> Cheers
> Steve
>
>
>
> ==============================
> Stephen Roberts
> Director
> Mathematical Sciences Institute
> Room 3.64 Hanna Neumann Building #145
> The Australian National University
> Canberra, ACT 2600 AUSTRALIA
> T: +61 2 61254445 | E: dir...@an...
>
>
>
> Executive Assistant: Brittany Joyce
>
> T: +61 2 6125 2897| E: ea...@an...
>
> CRICOS Provider: 00120C
>
> ________________________________
> From: S.Kummer <ste...@we...>
> Sent: Saturday, 24 April 2021 6:58 PM
> To: Stephen Roberts <ste...@an...>;
> anu...@li... <anu...@li...>
> Subject: Re: [Anuga-user] How to model a bridge
>
> Hi Stephen,
>
> I have tested the setup of the bridge in two ways:
>
> 1) no elevation barrier
>
> 2) as a river wall
>
> As 1) is concerned:
> What happens to the shallow water flux in the model
> at the location of the bridge?
>
> As 2) is concerned:
> My thougt was that this would be the most realistic setup because
> bridge overtopping can be modeled by the river wall formulas.
> I located the river wall at the inlet of the culvert. Should the river
> wall be located in the middle of the bridge?
>
> In all cases the culvert (width: 3.0 m) is given by exchange lines
> at the inlet and outlet of the culvert with an equiry gap that ist
> located in a reasonable distance (here: varying about 3 to 5 meters)
> to the exchange lines.
>
> My goal is to set up a realistic bridge behavior without the need of a
> time-consuming change of the existing elevation.
>
> Unfortunately I can't provide a working code of the example because it
> depends on to much self writen modules.
>
>
> Cheers
> Stefan
>
>
>
>
> Am Sat, 24 Apr 2021 05:23:12 +0000
> schrieb Stephen Roberts <ste...@an...>:
>
> > Hi Stefan,
> >
> > Sounds like you are doing a reasonable test.
> >
> > How have you set up the bridge? I assume as some sort of elevation
> > barrier in the channel.
> >
> > Make sure the inlets of the culvert are not interacting with the
> > barrier. Here a river wall can work well as it is infinitely thin.
> >
> > The culvert also works with enquiry_points which provide the data
> > used to calculate the flow through the culvert. Make sure they are
> > in reasonable locations.
> >
> > Also the culvert operator may have some difficulty in calculating
> > the correct input velocities. So some amount of backwater is
> > inevitable. There is an argument for the operator, use_momentum_jet
> > which you set to False which might provide more consistent results.
> >
> > Happy to check over your example.
> >
> > Cheers
> > Steve
> >
> >
> > ==============================
> > Stephen Roberts
> > Director
> > Mathematical Sciences Institute
> > Room 3.64 Hanna Neumann Building #145
> > The Australian National University
> > Canberra, ACT 2600 AUSTRALIA
> > T: +61 2 61254445 | E: dir...@an...
> >
> >
> >
> > Executive Assistant: Brittany Joyce
> >
> > T: +61 2 6125 2897| E: ea...@an...
> >
> > CRICOS Provider: 00120C
> >
> > ________________________________
> > From: S.Kummer <ste...@we...>
> > Sent: Saturday, 24 April 2021 2:22 AM
> > To: Stephen Roberts <ste...@an...>;
> > anu...@li... <anu...@li...>
> > Subject: Re: [Anuga-user] How to model a bridge
> >
> > Hi Stephen,
> >
> > I tried to setup a boyd-box culvert in a rectangular channel (width:
> > 3.0 meter, height: 1.0 meter) to model a bridge without concerning
> > submerging of the deck.
> >
> > I made some tests with a varying hight of the box culvert at a
> > constant width of 3.0 m.
> > It is not clear to me how ANUGA treats a culvert height greater than
> > the channel hight in this case (e.g. culvert height: 2.0 m) because
> > the backwater caused by the culvert doesn't vary significantly for
> > heights greater than 1.0 m although the culvert discharge varies
> > with the culvert hight.
> >
> > If the outlet controll of the culvert is "open channel flow" and the
> > inlet is not submerged: shouldn't there be nearly the same results
> > as the open channel flow without the culvert (e.g. no backwater
> > compared to the open channel flow without culvert)?
> >
> > Cheers
> > Stefan
> >
> >
> > Am Wed, 21 Apr 2021 11:49:41 +0000
> > schrieb Stephen Roberts <ste...@an...>:
> >
> > > Hi Stefan,
> > >
> > > I would recommend methods 1 or 3.
> > >
> > > There is a validation test in the anuga distribution
> > > (anuga_core/validation_tests/case_studies/towradgi) which uses
> > > method 1.
> > >
> > > Cheers Steve
> > >
> > > Get Outlook for Android<https://aka.ms/AAb9ysg>
> > >
> > > ________________________________
> > > From: S.Kummer <ste...@we...>
> > > Sent: Tuesday, April 20, 2021 9:26:39 PM
> > > To: anu...@li...
> > > <anu...@li...> Subject: [Anuga-user] How to
> > > model a bridge
> > >
> > > Dear ANUGA Users,
> > >
> > > what is the best way to model a bridge (including a possible
> > > bridge overtopping)?
> > >
> > > 1) Modeling a dam in the place of the bridge and using a culvert
> > > operator?
> > >
> > > 2) Modeling a dam and using rate operators (take water at a
> > > certain (limited) rate from the headwater and put water at the
> > > same rate at the tailwater)?
> > >
> > > 3) Using a culvert operator in combination with a river wall (if
> > > possible at all)?
> > >
> > > 4) ...?
> > >
> > >
> > > Best Wisches
> > > Stefan
> > >
> > >
> > > _______________________________________________
> > > Anuga-user mailing list
> > > Anu...@li...
> > > https://lists.sourceforge.net/lists/listinfo/anuga-user
> >
>