Model Type and Other Transformations

Convert model type for control design, reduce model order

System Identification Toolbox™ software allows you to create different types of dynamic system models. You can also carry out various transformations on the identified models such as merging and concatenating two or more models.

Functions

expand all

Model Types

`idfrd`	Frequency response data or model
`idpoly`	Polynomial model with identifiable parameters
`idtf`	Transfer function model with identifiable parameters
`idss`	State-space model with identifiable parameters

Model Transformations

`compreal`	Compute companion state-space realization (Since R2023b)
`modalreal`	Compute modal state-space realization (Since R2023b)
`noisecnv`	Transform identified linear model with noise channels to model with measured channels only
`translatecov`	Translate parameter covariance across model transformation operations
`merge`	Merge estimated models
`append`	Group models by appending their inputs and outputs
`noise2meas`	Noise component of linear identified model
`absorbDelay`	Replace time delays by poles at z = 0 or phase shift
`chgTimeUnit`	Change time units of dynamic system
`chgFreqUnit`	Change frequency units of frequency-response data model
`fdel`	Delete specified data from frequency response data (FRD) models
`stack`	Build model array by stacking models or model arrays along array dimensions
`ss2ss`	State coordinate transformation for state-space model

Topics

Using Identified Models for Control Design Applications
Using System Identification Toolbox models with Control System Toolbox™ software.
Subreferencing Models
Creating models with subsets of inputs and outputs from multivariable models at the command line.
State-Space Realizations
A state-space model can be expressed in an infinite number of realizations. Common forms, sometimes called canonical forms, include modal, companion, observable, and controllable forms.
Concatenating Models
Horizontal and vertical concatenation of model objects at the command line.
Merging Models
How to merge models to obtain a single model with parameters that are statistically weighed means of the parameters of the individual models.
Treating Noise Channels as Measured Inputs
Convert noise channels to measured channels and include the variance of the innovations.
Transforming Between Linear Model Representations
Converting between state-space, polynomial, and frequency-response representations.
Reducing Model Order Using Pole-Zero Plots
You can use pole-zero plots of linear identified models to evaluate whether it might be useful to reduce model order.
Create and Plot Identified Models Using Control System Toolbox Software
Identify models and use the Linear System Analyzer to plot the models.