System Identification Toolbox    

Defining Model Structures

Since the System Identification Toolbox handles a wide variety of different model structures, it is important that these can be defined in a flexible way. In the previous section you saw how models are automatically produced in the right form by the various estimation routines arx, iv4, oe, bj, armax, and pem, if you just specify model orders and delays.

This section describes how model structures and models can be directly defined. This may be required, for example, when creating a model for simulation. Also, it many be necessary to define model structures that are not of black-box type, but contain more detailed internal structure, reflecting some physical insights into how the system works.

The general way of representing models and model structures in the System Identification Toolbox is by various model objects. This section introduces the commands (apart from the parametric estimation functions themselves) that create these different models.

The model objects will contain a number of properties. For any model you can type

to see a list of the model's properties, and

to see what the assignable values are. See get and/or set in the "Command Reference" chapter. Each property value can easily also be retrieved by subreferencing as in

and set as in

See Function Reference for complete property lists. Here only examples are given. Note that it is sufficient to use any case insensitive, unambiguous abbreviation of the property names.


  Optional Variables Polynomial Black-Box Models: The idpoly Model