You can simulate the response of an identified model to given
inputs in the System Identification app
sim. You can predict
the model response a certain time horizon into the future using past
measurements of inputs and outputs. Use
predict model response over the time span of the measured data, and
forecast to predict the
response over a future time span when no measured data is available.
You can also import identified models to Simulink, and simulate
model response using model simulation blocks.
|Simulate response of identified model|
|Option set for sim|
|Simulate linear models with uncertainty using Monte Carlo method|
|Option set for simsd|
|Predict K-step ahead model output|
|Option set for predict|
|Forecast identified model output|
|Option set for forecast|
|Generate input signals|
|Iddata Source||Import time-domain data stored in iddata object in MATLAB workspace|
|Iddata Sink||Export simulation data as iddata object to MATLAB workspace|
|Idmodel||Simulate identified linear model in Simulink software|
|Nonlinear ARX Model||Simulate nonlinear ARX model in Simulink software|
|Hammerstein-Wiener Model||Simulate Hammerstein-Wiener model in Simulink software|
|Nonlinear Grey-Box Model||Simulate nonlinear grey-box model in Simulink software|
Understand the difference between simulated and predicted output and when to use each.
Perform simulation and prediction in the System Identification app, and interpret results.
Perform simulation, prediction, and forecasting at the command line, specify initial conditions.
Use model blocks to import, initialize, and simulate models from the MATLAB® environment into a Simulink model.
Description of the System Identification Toolbox™ block library.
Understand the concept of forecasting data using linear and nonlinear models.
Workflow for forecasting time series data and input-output data using linear and nonlinear models.
This example shows how to perform multivariate time series forecasting of data measured from predator and prey populations in a prey crowding scenario.