Hi DFE
Indeed we haven't tested the GUI extensively with very large tables in
the Integrator tab. Two options are (1) to reduce the number of output
variables that you opt to 'observe' (don't set all of them with non-zero
'obs_id'), or (2) stop using the GUI for large models, and write your
own IntegratorReporter code in Python, and load and run your problem
using a Python script.
An example (which I haven't tested for a little while) is:
http://code.ascend4.org/viewvc/code/branches/leon/models/test/ida/leon/bballPlot.py?view=markup
(the integration 'driver') which in turn refers to
http://code.ascend4.org/viewvc/code/branches/leon/models/test/ida/leon/TReport.py?view=markup
(the IntegratorReporter implementation).
The '.a4s' approach is to use Tcl/Tk scripts, which is essentially the
'old way' of scripting ASCEND. You can try that too if you want, but you
may have to use an older version of ASCEND for complete success. Ben
Allan can give help with that approach better than I can.
Hope that helps
Cheers
JP
On 05/07/12 10:46, Daikokuten Eudaemon wrote:
JP,
Thank you.
One more polite request: my ASCEND models have by now become so big
that I am no longer able to navigate the PyGTK "Integrator" tab
comfortably. Ctr-C does not seem to work i.e. I am unable to copy data
from the PyGTK "Integrator" tab for further analysis in R. Could you
direct me to the best starting point to learn how to control ASCEND
integration? At this stage I am "only" trying to be able alter the
values of input signals over time and save the values of observed
variables to an ASCII file. I did notice files like
${HOME}/ascend/models/dyn_tank.a4s, for example, but how do I get them
(or something similar) executed?
Regards,
DFE
On Wed, Jul 4, 2012 at 7:32 PM, John Pye <[email protected]
<mailto:[email protected]>> wrote:
Hi DFE
We don't yet have any support for delay differential equations in
ASCEND. Two things would be needed: firstly, the language
constructs would have to be figured out for how to access a
historical (delayed) value of a variable. We don't have anything
like that yet. Secondly, a suitable solver would have to be connected.
Some information about these kinds of problems is here, including
a list of available software.
http://www.scholarpedia.org/article/Delay-differential_equations
Cheers
JP
On 04/07/12 10:10, Daikokuten Eudaemon wrote:
Ascendians,
Could you point me to any publication and/or code that would help
me use ASCEND to integrate a dynamic model with dead time, such as
tau * dy(t)/dt + y(t) = K * u(t - theta)
?
Regards,
DFE
--
####
Daikokuten Fukurokuju Eudaemon
mailto:[email protected] <mailto:[email protected]>
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats.http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Ascend-sim-users mailing list
[email protected]
<mailto:[email protected]>
https://lists.sourceforge.net/lists/listinfo/ascend-sim-users
--
####
Daikokuten Fukurokuju Eudaemon
mailto:[email protected] <mailto:[email protected]>
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Ascend-sim-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ascend-sim-users
