System Identification Toolbox™

IDMODEL Model - Simulate `idmodel` object in Simulink software

Library

System Identification Toolbox

Description

The IDMODEL block simulates a linear model in the MATLAB workspace.

Note For simulating nonlinear models, use the IDNLGREY, IDNLARX, or IDNLHW Model blocks.

Input

Input signal to the model.

Output

Simulated output from the model.

Dialog Box

idmodel variable

Name of idmodel variable in the MATLAB workspace, which can be an idpoly, idss, idgrey, idarx or idproc model object that you estimated using System Identification Toolbox software.

This block supports both continuous-time and discrete-time models with or without input-output delays.

Initial state (state space only)

Initial state vector.

(Default) 'z' (in quotation marks) specifies zero, which corresponds to a simulation starting from a state of rest.

Enter 'm' to use the internal initial states of the model, stored by the idss or idgrey model property X0.

To specify your own initial states, enter one of the following:

Enter known initial states for a state-space model directly, as a numerical vector.
Compute the initial states using the findstates(idmodel) command to get the initial state values that produce the best fit between the model output and the measured output signal.
For example, if the data set z and model m are already in the MATLAB workspace, you can enter X0, such that:
```
X0 = findstates(m,z1)
```

Add noise

Select when you need to add noise, as specified by the model property model.NoiseVariance and the matrices or polynomials that determine the color of the additive noise.

For continuous-time models, the ideal variance of the noise term at any moment is infinite. In reality, you see a band-limited noise that takes into account the natural time constants of the system. You can interpret the resulting simulated output as filtered using a low-pass filter with a pass-band that does not distort the dynamics from the input.

Noise seed(s)

(Use when you select the Add noise check box.)

Enter an integer that specifies a seed that forces the simulation to add the same noise to the output every time you simulate the model. For more information about using seeds with functions that generate random number sequences, see the rand reference page in MATLAB documentation, for example.