User`s guide

4 Nonlinear Black-Box Model Identification
7 To plot the response of this m odel, select the appropriate check box in
the Model V iews area of the Sy stem Identification Tool GUI. For more
information about working with plots and validating m odels, see Chapter
8, “Model Analysis”.
If you g et an inaccurate fit, try estimating a new model with different o rders
or nonlinearity estimator.
You can export the model to the MATLAB w orkspace for further analysis
by dragging it to the To Workspace rectangle in the System Identification
Tool GUI.
How to Estimate Hammerstein-Wiener Models at the
Command Line
• “General nlhw Syntax” on page 4-20
• “Improving Estimation Results Using Initial States” on page 4-22
• “Example – Using nlhw to Estimate Hammerstein-Wiener Models” on page
4-23
General nlhw Syntax
You can estimate Hammerstein-Wiener models using the nlhw command. The
resulting models are stored as
idnlhw model objects.
Use the following gener al syntax to both configure and estimate
Hammerstein-Wiener models:
m = nlhw(data,'nb',nb,...
'nf',nf,...
'nk',nk,...
InputNonlinearity,..
OutputNonlinearity,...
'Property1',Value1,...,
'PropertyN',ValueN)
where data is the estimation dat a . nb, nf,andnk specify the orders and delays
of the linear model, which is similar to an Output-Error (OE) model. For more
4-20