As planned, this is the second part of the MPC series. To dealing with multivariable, state-space model is most convenient. In this part, a tool to setup the state-space model based predictive controller is provided. The setup program returns a function handle for the online MPC controller.
The online controller is implemented as a nested function, hence the internal model, internal states, control parameters are all stored inside of the workspace of the setup program so that the interface of the online controller is significantly simplified.
A 2-CSTR example is included to demonstrate the usage of this tool.
The tool also provides an example for beginners to learn how to use nested function to replace persistent variables.
nice, im doing an mpc multivariable work on my work degree, i want to ask you about mpc tuning methodology, i mean, i tune weights just making system stable and simulating all operation points i´ll need, but i want to do a better job.
Well,its a good tutorial. Just need more example to make it perfect.
A bug removed. Now different output can follow different setpoint properly.
Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.