How to reduce the effect of one of the manipulated variables in an MPC controller with a MISO model as the internal model?

I am using a MISO (2 inputs) model as the inernal plant model in an MPC controller and I want one of the manipulated variables to have maximum effect on the output.I tried to adjust the weights of the manipulated variables but still could not get a desired effect.How to fix this issue?I am using MPC toolbox from MathWorks for designing the MPC controller.

Answer 1

[Kothuri]

Hi Dawn Ivan,

To increase the effect of one manipulated variable on the output in the MISO model with MPC, you can follow the below steps:

• Ensure that the weights for the manipulated variables are set correctly. Make sure the weight for the MV you want to have a greater effect is higher relative to the other MVs.
• Ensure that both inputs are properly scaled to have similar magnitudes to allow for fair comparison and effective control. A significant difference in the magnitudes of the input variables could cause one variable to be less effective.
• Tune the prediction horizon (PH) and control horizon (CH) as they can significantly impact the controller’s performance. A longer PH allows the controller to anticipate future events better, while the CH determines how many future control moves are optimized at each step.
• Ensure that your internal plant model accurately represents the real system.
• Simulate the controller's performance under various scenarios using to test different configurations and observe the effects using the "mpcmove' function.

You can refer the below link for more info on “mpcmove”

https://www.mathworks.com/help/mpc/ref/mpc.mpcmove.html

You can refer the below Documentation link for more info on MPC controller for MISO system

https://www.mathworks.com/help/mpc/gs/control-of-a-multi-input-single-output-plant.html#d126e6352