User`s guide

3 Linear Model Identification
Extracting Dynamic Model and Noise Model Separately
You can extract the numerical data associated with a dynamic model and
the n o ise model separately.
For line ar models, the general symbolic model des cription is given by:
yGuHe=+
G is an operator that takes the measured inputs u to the outputs and captures
the system dynam ics. H is an operator that describes the properties of the
additive output disturbance and takes the hypothetical (unmeasured) noise
source inputs e to the outputs, also called the noise model. When you estimate
a noise model, the toolbox includes one noise channel e at the input for each
output in your system.
Thefollowingtablesummarizestheresultsof
ssdata, tfdata,andzpkdata
commands for extracting the numerical values of the dynamic model a nd
noise model separately.
fcn represents ssdata, tfdata,andzpkdata,and
m is a model object. L represents the covariance matrix e,asdened in
“Subreferencing Measured and Noise Models” on page 3-120.
For information about subreferencing noise channels or treating noise
channels as measured input, see “Subreferencing Model Objects” on page
3-119.
Note The s yntax fcn( m('noise')) is equivalent to fcn( m('n')).
Syntax for Extracting Transfer-Function Data
Command
Syntax
fcn(m)
Returns th
e properties of G for ny outputs and nu inputs.
fcn(m('no
ise'))
Returns th
e properties of H for ny outputs and ny inputs.
fcn(noise
cnv(m))
Returns th
e properties of [ GH] ny outputs and ny
+nu
inputs.
3-110