User`s guide

Identifying State-Space Models
routines. Because the parameterization of A, B,andC is free, a basis for
the state-spa ce rea liz ation i s automatically selected to give well-cond iti on ed
calculations.
You can only estimate discrete-time state-space models with any
parameterization. Continuous state-space models are available for canonical
and structured parameterizations only.
To estimate the disturbance model K, you must use time domain data.
Suppose that you have no knowle dge ab ou t th e in tern a l structure o f the
discrete-time state-space model. To quickly get started, use the following
syntax:
m = pem(data)
where data is y our estimation data. This command estimates a state-s pace
model for an automatically selected order between 1 and 10.
To find a black-box model of a specificorder
n,usethefollowingsyntax:
m = pem(Data,n)
The i terative alg ori thm pem is initiali zed by the su b s p ace method n4sid.You
can use
n4sid directly, a s an alternative to pem:
m = n4sid(Data,n)
How to Estimate State-Space Models with Canonical
Parameterization
• “What Is Canonical Parameterization?” on page 3-91
• “Estima t in g Canonical State -Space Models” on page 3-92
What Is Canonical Parameterization?
Canonical parameterization represents a state-space system in its minimal
form,usingtheminimumnumberoffreeparameterstocapturethedynamics.
Thus, free parameters appear in only a few of the row s and columns in system
3-91