Hi Benjamin,
Thanks for your feedback on ANUGA - I'm glad that you have found it useful.
One potential way of dealing with this is to export the timeseries at a selection of points that are on your boundary and then convert them to the sts format and use the Field_boundary function and attach to your boundary as follows:
Bf = Field_boundary('file.sts',
domain, mean_stage=0.0,
time_thinning=1,
default_boundary=Dirichlet_boundary([0, 0, 0]),
boundary_polygon=boundary_polygon,
use_cache=True,
verbose=True)
domain.set_boundary({'back': Br,
'side': Bt,
'ocean': Bf}
Good luck!
Cheers, Jane.
________________________________
From: Benjamín Carrión [mailto:benja.carrion@...]
Sent: Thursday, 22 April 2010 12:41 AM
To: anuga-user@...
Subject: [Anuga-user] memory problem
Hello to all,
I've been working with Anuga for a while, and I have to say that it has proved to be an useful, powerful tool for tsunami floods.
Anyway, I've been running out of memory, since the files it produces are very large, due to the accuracy we're demanding of it.
Particularly, I run out of memory when I try to read an .sww file from a coarse grid to feed the boundary conditions for a finer one. Here's the question: is there any way to reduce the coarse .sww file only to the region of interest, so it would be a bit less larger?
Thanks in advance
--
Benjamín Carrión Aretxabala
--------------------------------------------------------------------
Por favor, considera el medio ambiente antes de imprimir este e-mail
|