Model Predictive Control Toolbox™ 2.3.1

Product Description

• Introduction and Key Features
• Working with the Model Predictive Control Toolbox
• Defining Plant Models
• Designing Controllers
• Simulating Closed-Loop Behavior
• Deploying Model Predictive Controllers

Defining Plant Models

Model predictive controllers base their control actions on an internal plant model of the process. The internal model lets the controller forecast future process behavior and respect output constraints. The ability to update the internal model makes model predictive control easier to maintain than complex coupled PID loops that require individual tuning when system parameters change.

The Model Predictive Control Toolbox uses LTI models, enabling you to use transfer-function model structures common to all MathWorks control system design products. You can import multiple LTI models into the toolbox from the MATLAB workspace or as a .MAT file. The toolbox also lets you directly import multiple models estimated in the System Identification Toolbox.

Using Simulink and Simulink Control Design, you can extract a linearized form of the Simulink model that is automatically imported as the internal plant model of the controller. You can then refine your internal plant model in one step in Simulink.



The Plant Model Importer brings models into the toolbox. Click on image to see enlarged view.