Re: ControlSystems Wolfram application. Does it support root locus for discrete time systems?
- To: mathgroup at smc.vnet.net
- Subject: [mg60015] Re: ControlSystems Wolfram application. Does it support root locus for discrete time systems?
- From: "Nasser Abbasi" <nma at 12000.org>
- Date: Sun, 28 Aug 2005 03:07:56 -0400 (EDT)
- References: <dek88k$aag$1@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
"Nasser Abbasi" <nma at 12000.org> wrote in message news:dek88k$aag$1 at smc.vnet.net... > > > > > Any ControlSystems package users out there? > > I can't seem to figure how to generate, using the ControlSystems > package, the root locus for a sampled system. It works OK for > continues-time system as you can here: > > http://12000.org/my_notes/mma_matlab_control/e45/HTML/e45.htm > > But when I try to sample the system, it gives the wrong root locus: > > http://12000.org/my_notes/mma_matlab_control/e68/HTML/e68.htm > > thanks, > Nasser The problem is resolved now. I needed to use the correct plot range to 'see' the root locus. It was there all along, but was too small to see. (If there was a good zoom option, I would probably have found this earlier). I did not notice this, since in that 'other' system, the plot range for the root locus is determined automatically, for example, for a discrete system, one is mostly interested in what the root locus looks like inside and close to the unit circle. Any way, Mathematica can for sure generate the correct root locus for a sampled time system, but one needs to be careful how to use the command. The updated plot is found here: http://www.12000.org/my_notes/mma_matlab_control/e68/HTML/e68.htm Nasser