User`s guide

Supported Nonlinearity Estima tors
Supported Nonlinearity Estimators
In this section...
“Types of No nline arity Estimat ors” on pag e 4-25
“Creating Custom Nonlinea r ities” on p age 4-26
TypesofNonlinearityEstimators
When configuring the nonlinear ARX and Hammerstein-Wiener models for
estimation, you must specify a mathematical structure for the nonlinear
portion of the model.
If you are working i n the System Identification Tool GUI, specify the
nonlinearity type by name when you con figure the n onlinear model structure.
If you are estimating or constructing a nonlinear model at the command
line instead, specify the nonlinearity as an argument in the
nlarx or nlhw
estimation command.
The following table summarizes supported nonlinearities in the System
Identification Toolbox product for each type of nonli near model. For a
description of each nonlinearity , see the references page for the corresponding
nonlinearity object.
Nonlinearity Object Name Supported Model Type Supports Multiple
Inputs?
Custom Network
(user-defi ned)
customnet
Hammerstein-Wiener and
Nonlinear ARX
Yes
Dead Zone
deadzone
Hammerstein-Wiener
No
Neural Net
work
neuralnet
Nonlinear ARX
Yes
Piecewise
Linear
pwlinear
Hammerste
in-Wi ener
No
One-Dimen
sional
Polynomi
al
poly1d
Hammerstein-Wiener
No
Saturati
on
saturati
on
Hammerstein-Wiener
No
Sigmoid N
etwork
sigmoidn
et
Hammerstein-Wiener and
Nonlinear ARX
Yes
4-25