Dear list,

Recently I've been researching whether Ptolemy II is usable as a control system modeling and design tool in our research group. We do a lot of system modeling and controller design. A typical setup is to create a Continuous Time model that describes the dynamic behavior of the plant with a discrete time model that describes the controller behavior in a feedback loop.
Doing some preliminary tests I came across some unexpected behavior.
I created a simple top-level CT model that generates a sine-wave using the ContinuousSineWave actor. This signal is fed into a CompositeActor under the control of a DT director. Inside the CompositeActor the signal is multiplied by one and then sent to an output port and a TimedPlotter. Another TimedPlotter is connected to the output of the CompositeActor (in the CT domain).

What I'd expect is that the TimedPlotter inside the CompositeActor and the TimedPlotter connected to the output of the CompositeActor both plot the same sampled version (at sampling period 0.1 sec) of the input sinewave.

However, I get the following results:
- The TimedPlotter inside the CompositeActor plots the signal with the right number of periods, the right sampling interval, but on the wrong timescale (0 - 20 seconds while the simulation time is 0-10 seconds) - The TimedPlotter connected to the output of the CompositeActor plots the expected signal (right number of periods, correct sampling interval and correct timescale).

I've uploaded my test model at http://www.scintilla.utwente.nl/~keesv/DTtest.zip . I'm using a self compiled Ptolemy II 6.0.2 on Mac OS X 10.5.1, with the java version that came with my OS (Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_13-b05-237)).

I hope that someone can shed some light on the matter.

Kind regards,

Kees Verhaar
Control Engineering Group
University of Twente, The Netherlands

----------------------------------------------------------------------------
Posted to the ptolemy-hackers mailing list.  Please send administrative
mail for this list to: [EMAIL PROTECTED]

Reply via email to