MATLAB: Nonlinear Model Predictive controller toolbox

nmpc

Hi,

Does any one knows if there exist a NMPC toolbox in Matlab?

I´ll appreciate your help.

Best Regards

Best Answer

Hi Cesar,

as I read through this topic, only linear MPC are available. Reason is that they are similar to a state feedback law and can be calculated fast. If you only look for nonlinear-MPC because your plant is nonlinear, Matlab helps by supplying a lot of awesome linearization tools. Recently, the MPC toolbox supports changing the linearization point during runtime so even though your model is linearized, it will be pretty close to the real solution. It is called multiple MPC:

 http://de.mathworks.com/help/mpc/ug/nonlinear-control-using-multiple-models.html

If you need a nonlinear MPC because of you needing a very special cost function or other nonlinear (cool, awesome) stuff, the Toolboxes don't help. Then you need to make your own nonlinear MPC. Be aware that such things run very slow!

Related Solutions

MATLAB: Do I get a “marshall.coerceToClass” error when simulating model containing Adaptive MPC block

In the Adaptive MPC block, the plant model is being specified by "mpcobj", which has 3 states, 1 input and 1 output. Therefore, the model provided to the Adaptive MPC block should also have 3 states, 1 input and 1 output. When the order of transfer function is increased within the 'Time Varying Predictive Model' block, the number of states after converting to state space becomes 4, and an error is thrown because mpcobj expects an order of 3. There are some ways to update the plant model and the predictive model during simulation. To explain this in more detail, I am including a couple of documentation links below:

https://www.mathworks.com/help/<https://www.mathworks.com/help/releases/R2018b/mpc/ref/mpc.mpcmoveadaptive.html#buik3vp-1-Plant releases/R2018b/mpc/ref/mpc.mpcmoveadaptive.html#buik3vp-1-Plant>

https://www.mathworks.com/help/<https://www.mathworks.com/help/mpc/examples.html?category=adaptive-mpc-design&s_tid=CRUX_gn_example releases/R2018b/><https://www.mathworks.com/help/mpc/ref/adaptivempccontroller.html#bugls7u-7 mpc/ref/adaptivempccontroller.html#bugls7u-7>

To avoid the error, please ensure that the Plant model for mpcobj and the model inports for Adaptive MPC controller are always consistent with each other. # Regarding resources on implementing Adaptive Control, please find below some examples on Adaptive MPC design:

https://www.mathworks.com/help/releases/R2018b/mpc/examples.html?category=adaptive-mpc-design&s_tid=CRUX_gn_example

The LPV (linear parameter varying) block in Control System Toolbox may also be helpful. See the link below:

https://www.mathworks.com/help/<https://www.mathworks.com/help/mpc/examples.html?category=adaptive-mpc-design&s_tid=CRUX_gn_example releases/R2018b/><https://www.mathworks.com/help/mpc/ug/adaptive-mpc-control-of-nonlinear-chemical-reactor-using-linear-parameter-varying-system.html mpc/ug/adaptive-mpc-control-of-nonlinear-chemical-reactor-using-linear-parameter-varying-system.html>

MATLAB: Are the nominal conditions of the nonlinear CSTR model set to [0 0 u] in the example code for successive linearization in the Model Predictive Control Toolbox Version 2.2 (R14SP2)

This enhancement has been incorporated in Release 2008a (R2008a). For previous product releases, read below for any possible workarounds:

The reason why we have "Model.Nominal.U = [0 0 u];" instead of "Model.Nominal.U = up;" is that the first 2 plant inputs are designated as unmeasured disturbance to the Model Predictive Control (MPC) Toolbox by the following structure in the example code:

Plant.InputGroup.UnmeasuredDisturbances = [1 2];

This implies that the Model Predictive Control (MPC) toolbox doesn’t know the values of those 2 plant inputs during its optimization calculation. Therefore, by design, nominal values for all the unmeasured disturbance variables to the Model Predictive Control (MPC) toolbox have to be 0. In other words, if we have

Plant.InputGroup.UnmeasuredDisturbances = [1];
Plant.InputGroup.MeasuredDisturbances = [2];
Plant.InputGroup.ManipulatedVariables = [3];

Then we will see:

Model.Nominal.U = [0 up(2) up(3)];

Note that those two values are used by the plant itself in the closed loop simulation.

Best Answer

Related Solutions

MATLAB: Do I get a “marshall.coerceToClass” error when simulating model containing Adaptive MPC block

MATLAB: Are the nominal conditions of the nonlinear CSTR model set to [0 0 u] in the example code for successive linearization in the Model Predictive Control Toolbox Version 2.2 (R14SP2)

Related Question