System Identification Toolbox™

nlhw - Estimate Hammerstein-Wiener models

Syntax

m=nlhw(data,[na nb nk],Nonlinearity)
m=nlhw(data,[na nb nk],P1,V1,...,PN,VN)
M = NLHW(DATA, ORDERS, InputNL, OutputNL, 'InitialState',INIT,..)

Arguments

data

Time-domain iddata model object.

[nb nf nk]

Model orders and input delays, where nb is the number of zeros plus 1, nf is the number of poles, and nk is the delay from input to output in terms of the number of samples.

For nu inputs and ny outputs, nb, nf and, nk are ny-by-nu matrices whose i-jth entry specifies the orders and delay of the transfer function from the jth input to the ith output.

InputNL and OutputNL

Specify the input and output nonlinearity estimator objects as one of the following: pwlinear (default), deadzone, wavenet, saturation, customnet, sigmoidnet, and unitgain. The nonlinearity estimator objects have properties that you can set in the constructor, as follows:

m=nlhw(data,[2 2 1],sigmoidnet('num',5),deadzone([-1,2]))

To use default nonlinearity properties, specify the nonlinearity object name as a string. For example:

m=nlhw(data,[2 2 1],'sigmoidnet','deadzone')
m=nlhw(data,[2 2 1],'sig','dead') % Abbreviated

The estimator unitgain (can also be entered as []) means no nonlinearity. Thus, m=nlhw(data,[2 2 1],'saturation','[]'). For more information about nonlinearity properties, see the corresponding reference pages.

InitialState

The value of the initial state. InitialState can have the following values:

• z: Zero (Default).

• e: Estimated initial state.

Description

m=nlhw(data,[na nb nk],Nonlinearity) constructs and estimates a Hammerstein-Wiener model with orders [na nb nk] and Nonlinearity. data is the estimation data set.

m=nlhw(data,[na nb nk],P1,V1,...,PN,VN) constructs and estimates the model with additional property-value pairs. For more information about model idnlhw model properties, see the corresponding reference pages.

M = NLHW(DATA, ORDERS, InputNL, OutputNL, 'InitialState',INIT,..) lets you indicate the status of the initial state.

Examples

The following commands construct and estimate a Hammerstein-Wiener model:

load iddata1
m1=nlhw(z1,[4 2 1],'sigmoidnet','deadzone')

See Also

nlarx

noisecnv