User`s guide

4 Nonlinear Black-Box Model Identification
For more information abo ut re gress ors, see “Using Regre ssors ” on page 4-6.
For a list of nonlinearity estim ators supported by nonlinear ARX models, see
“Nonlinearity Estimators for Nonlinear ARX Models” on page 4-9.
Using Regressors
You can use the follow ing types of regressors for no nline ar ARX models:
• Standard regressors—Pastinputu(t) and output signals y(t),computed
automatically as delay transformations f or specified model orders.
• Custom regressors — Products, powers, and ot her MATLAB expressions
of input and output variables that you specify.
Specifying Model Order and Delays
You must specify the following model orders for computing standard
regressors:
• n
a
— The number of past output terms used to predict the current output.
• n
b
— The number of past input terms used to pre dict the current output.
• n
k
— The delay f rom input to the output in terms of the n umber of samples.
This value defines the least delay ed input regressor.
The meaning of n
a
and n
b
is simila r to the linear-ARX model parameters in
the sense that n
a
represents the number of output terms and n
b
represents the
number of input terms. n
k
represents the minimum input delay from an input
to an output. For more information about the linear AR X m odel structure, see
“What Are Black-Box Polynomial Models?” on page 3-41.
Note The total number of regressors in the model must be greater than zero.
If you o nly need to use custom regressors, set n
a
=n
b
=n
k
=0 to omit crea ting
standard regressors.
4-6