Anuga-user Digest, Vol 42, Issue 5
anuga-use...@lists.sourceforge.net
anuga...@lists.sourceforge.net
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.sourceforge.net/lists/listinfo/anuga-user
or, via email, send a message with subject or body 'help' to
anuga-use...@lists.sourceforge.net
You can reach the person managing the list at
anuga-us...@lists.sourceforge.net
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Anuga-user digest..."
Today's Topics:
1. Re: HELP REQUIRED [SEC=UNCLASSIFIED] (Gareth...@ga.gov.au)
2. Simulation of a river flooding (Silvia Franceschi)
3. Anuga shallow water simulation (Silvia Franceschi)
----------------------------------------------------------------------
Message: 1
Date: Fri, 17 Feb 2012 13:11:47 +1100
From: <Gareth...@ga.gov.au>
Subject: Re: [Anuga-user] HELP REQUIRED [SEC=UNCLASSIFIED]
To: <scori...@gmail.com>
Cc: anuga...@lists.sourceforge.net,
anuga-us...@lists.sourceforge.net
Message-ID:
<60D02F9C526B7F48B2194...@EXCCR01.agso.gov.au>
Content-Type: text/plain; charset="us-ascii"
Hi Parul,
See my responses to your questions below,
Hope that helps,
Gareth.
'Is there a standard to determine the resolution for the domain'
The standard is to use convergence testing. Note that ANUGA is approximating the solution of the shallow water equations on the mesh you create. Theoretically, these algorithms should produce an approximation that gets closer and closer to the 'true' solution as the mesh is refined. For any given problem, it is hard to say what a suitable resolution is up-front.
However, you can determine if the error associated with the mesh is sufficiently small by comparing the outputs of runs with differing mesh resolutions, along the lines of what you are already doing (ideally I would suggest shrinking the triangle side length by 50% with each iteration). If the solution doesn't change significantly as you refine the mesh, then it is likely that you are using a fine enough mesh. One caveat to the latter statement is if you are still not resolving important features of the topography - e.g. a spatially extensive, 30cm thick vertical wall could strongly impact the results of a flood simulation, but would only be resolved if the triangle side length was fine enough.
The main challenge with this sort of approach is that you expect the run times to increase by a factor of 8 each time you halve the triangle side length. So convergence testing can quickly get computationally heavy. However, as far as I know it is the only rigorous approach to determining a suitable mesh size.
'Which resolution will be considered optimum?'
See above
'Opting for single resolution is advisable or differential?'
This is problem dependent. The main advantage of a variable resolution is a reduction in the computational time (because you have less triangles over the mesh, with a fine mesh only where you think it is needed). However, there may also be other errors introduced by having the mesh vary too strongly - ideally this sort of thing should also be subject to sensitivity tests.
Message: 1
Date: Tue, 14 Feb 2012 17:01:10 +0530
From: parul mogra <scori...@gmail.com>
Subject: [Anuga-user] HELP REQUIRED
To: anuga...@lists.sourceforge.net
Cc: anuga-us...@lists.sourceforge.net
Message-ID:
<CAPkZ3MTt7hK1p77=Nefa3QPvwE4fSC+Jd...@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"
Hello All,
I am using ANUGA for simulating the GLOF scenarios in hilly terrains.
I have taken into consideration all the parameters and inputs.
I am keeping certain standard parameters while varying few others and generating reports. For instance I have the following parameters as standard in the code: Yield step --> 60 sec, Duration --> 5 hours, the input data of time vs. Discharge...........
What i am varying as of now is the resolution. What i have opted for is "Differential" resolution for domain/lake/buffer.
I have tried varied combinations like 1,00,000/1000/1000 , 50,000/800/800, 1000000/10000/10000.
The problem i am facing is that i am getting too much variation in my outputs esp.... "Inundation Depth" which i got in a (.csv) file via the sww2csv_gauges( ).
Is there a standard to determine the resolution for the domain?
Which resolution will be considered optimum?
Opting for single resolution is advisable or differential?
An early response from your side will be appreciated.....
Regards
Scori...@gmail.com
Gareth Davies
Hydrodynamic Modeller
Regional Risk Section,
International Group
Geoscience Australia.
Ph: (02) 6249 9655
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
Message: 2
Date: Sun, 26 Feb 2012 21:35:32 +0100
From: Silvia Franceschi <silvia.f...@gmail.com>
Subject: [Anuga-user] Simulation of a river flooding
To: anuga...@lists.sourceforge.net
Message-ID:
<CABCykqLCMoOooV87nyHt9gLSO=ORD+F=pR6j08oDa...@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"
Dear list,
I am new to this list and to anuga software... I am trying to understand if
it is possible to use Anuga for simulation of flooding on rivers giving the
DTM and the discharge at a defined section.
I am sorry to ask a so general question, but I tried to find an example of
this in the manual but there are no example of rivers flooding with real
data, do you have any material or manual containing such an example?
Any help is really appreciated! :)
Best regards,
Silvia
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
Message: 3
Date: Mon, 27 Feb 2012 19:43:47 +0100
From: Silvia Franceschi <silvia.f...@gmail.com>
Subject: [Anuga-user] Anuga shallow water simulation
To: anuga...@lists.sourceforge.net
Message-ID: <4F4BCEE...@gmail.com>
Content-Type: text/plain; charset="iso-8859-1"
Hi all,
I am trying to run a shallow water simulation on a river with a defined Inflow_boundary.
I am having several issues right now and would appreciate some advice.
When I import:
from anuga.shallow_water.boundaries import Inflow_boundary
and then set the inflow rate to a fixed value:
Bin = Inflow_boundary(domain, rate=ref_flow)
I get and error that the global variable: epsilon
is not set.
Looking at the testcases I tried to import:
from anuga.config import epsilon
but that didn't solve the problem.
So I cheated and inside boundary.py I set a variable:
epsilon = 1.0e-12
at line 399 to make the condition:
if slope > epsilon and mannings_n > epsilon:
work! :)
At that point I have another error a couple of lines later in:
q = num.array([elevation + depth, xmomentum, ymomentum], num.Float)
which I changed to
q = num.array([elevation + depth, xmomentum, ymomentum], num.float)
At that point the simulation started. It is running and will take a while, but the last change was
too strange to me to be real :)
I apologize I am a python ignorant, so I am trying to do what I can.
Then I finally read this piece of documentation at the very start of the function:
# FIXME (Ole): This is work in progress and definitely not finished.
# The associated test has been disabled
So at that point I am wondering if it is supposed to work or not.
Could you please give me some advice on how such a kind of simulation should be best performed?
Thank you very much for any hint.
Best regards,
Silvia
PS: I am using today's latest svn checkout.
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
------------------------------
_______________________________________________
Anuga-user mailing list
Anuga...@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/anuga-user
End of Anuga-user Digest, Vol 42, Issue 5
*****************************************