User`s guide
8 Model Analysis
Using Impulse- and Step-Response Plots to Validate
Models
In this section...
“Supported Models” on page 8-24
“How Transient Response Helps to Validate Models” on page 8-24
“What D oes a Transie nt Respo nse Plot Show?” on page 8-25
“How to Plot Impulse and Step Response Using the GUI” on page 8-26
“Displaying the Confidence Interval” on page 8-29
“How to Plot Impulse and Step R esponse at the Command Line” on page
8-30
Supported Mod
els
You can plot th
e s imulated response of a model using impulse and step
signals as the
input for all linear parametric m odels and correlation analysis
(nonparamet
ric) models.
You can also c
reate step-respo n se plots for nonlinear m odels. These step
and impulse r
esponse plots, also cal led transient response plots, provide
insight into
the characteristics of model dynamics, inclu ding peak response
and settlin
gtime.
Note For freq
uency-response models, impulse- and step-response plots are
not availab
le. F or nonlinear models, only step-response plots are available.
How Transie
nt Response Helps to Validate Models
Transient r
esponse plots provide insight into the basic dynamic properties of
the model, s
uch as response times, static gains, and delays.
Transient r
esponse plo ts also help you validate h ow we ll a linear parametric
model, such
as a linear AR X model or a state-space model, captures the
dynamics.
For example, you can estimate an impulse o r step response from
the data us
ing correlation analysis (nonparametric model), and then plot the
8-24