User`s guide

4 Nonlinear Black-Box Model Identification
Extracting Parameter Values from Nonlinear Black-Box
Models
In this section...
“Nonlinear ARX Parameter V alu es” on pag e 4-30
“Hammerstein-Wiener Parameter values” on page 4-31
Nonlinear ARX Parameter Values
You can extract the numerical parameter values of a nonlinear ARX mo del by
accessing the properties o f the
idnlarx model o bject.
The no nlinear ARX model parameters you m ight want to explore include
model orders and delays, regressors, nonline arity estimator parameters, and
initial state values. To access the list of standard or custom regressors, you
can u se the
getreg command. For more informatio n, see the getreg reference
page.
You can view the parameters of the nonlinearity estimators using the
Nonlinearity property. Fo r example:
% Load sample data.
load iddata1
% Estimate a nonlinear ARX model that includes a
% treepartition nonlinearity.
m = nlarx(z1,[2 2 1],'tree');
NL = m.Nonlinearity;
get(NL)
NL.Parameters
For m ore information about the treep artition nonlinearity, see the
corresponding reference page.
To access any properties of a nonlinear ARX model object, use the
get
command. For m ore information about nonlinear ARX model properties, see
the
idnlarx reference page.
4-30