User`s guide

Using Models with Control System Toolbox™ Software
Using balred to R
educe Model Order
In some cases, th
e order of your identified model might be higher than
necessary to cap
ture the dynamics. If you have the Contro l Sys tem Toolbox
software, you ca
nuse
balred to compute a state-spate model approximation
with a reduced m
odel order for any
idmodel object, including idarx, id pol y,
idss,andidgre
y
.
For more inform
ation about using
balred, see the corresponding reference
page. To learn h
ow you can reduce model order us ing pole-zero plots, see
“Reducing Mod
el Order Usi ng Pole-Zero Plots” on page 8-51 .
Compensator D
esign Using Control System Toolbox
Software
After you esti
mate a plant model using System Identification To olbox
software, yo
u can use Control System Toolbox software to design a controller
for this plan
t.
System Ident
ification Toolbox m odels in the MATLA B workspace are
immediatel
y available to Control System Toolbox commands. However, if you
used the Sys
tem Identification Tool GUI to estimate models, you must first
export the m
odels to the MATLAB workspace. To export a model from the
GUI, drag th
e model icon to the To W o rkspace rectangle.
Control Sy
stem Toolbox software provides both the SISO Design Tool GUI
and comman
ds for working at the command line. You can import polynomial
and state-
space models d irectly into SISO Design Tool using the following
command:
sisotool(model('measured'))
where y ou u
se only the dynam ic model a nd not the noise m odel. For more
informati
on about subreferencing the dynami c or the no ise mod el, see
“Subrefe
rencing M easured and N oise Models” on page 3-120. To design a
controll
er using C ontrol System Toolbox co mmands and methods at the
command l
ine, you must convert the plant model to a n LT I object. For more
informat
ion, see “Converting Models to LTI O bjects” on page 1 0-4.
10-3