how to make it state space representation a dynamics microgrid model in Model Predictive Controller(MPC)

Hello,

so i have a microgrid model and MPC, like in the figure and I had en error "Plant Models cannot be a pure direct feed through system without additional dynamics " and I don't know how to make it dynamics.

[akshatsood]

Hi @Kawsar, could you please let me know if you did follow the procedure outlined in the document attached below

https://in.mathworks.com/help/mpc/gs/designing-a-model-predictive-controller-for-a-simulink-plant.html

In case you did and still facing the issue, please help me with your model and the data used to reproduce it on my end.

[Mounira]

Hello, sir, thank you for your response, I would love to send you the model, but maybe by email or something, because i have been working on it for a long time, I cannot just uploaded here, can I?

I tried using Mux to combine signals but it gave an error about the size of the plant model, different error than before so i deleted one input, then it gave me another error "Plant Models cannot be a pure direct feed through system without additional dynamics " and I don't know how to make it dynamics. so my question for you is is it the problem that I have used mux for signals, or what is the problem?

[Mounira]

this is the mux i was telling you about,

and the second figure is the state space representation

[akshatsood]

Hi @Kawsar, thank you for sharing the additional information. Please attach you model with the question by using the "paperclip 🔗" button. This will help me in debugging the exact cause of the error.

[Mounira]

i upload the file.mat for the microgrid model to run, but there is one variable which is not defined, Tm=26; you can write it on command window if you want,

so in the measurement block you will find:

1) subsystem for the signals input to the MPC

and 2)subsystem for the outputs signals to the MPC, that's how i was importing the signals to the MPC,

ps: i will upload for you another pic for the state space representation of the plant model, this plant is connected between the input of the MPC and the output,

and thanks a lot for your kind help

[akshatsood]

Hi @Kawsar,.

I investigated your issue and opened your model in MATLAB R2023b. While designing the MPC, I was able to get the following table. I seek your help in correctly identifying the issue you are facing, is it related to not able get the following table or after designing the MPC, you are not able to see the correct number of inputs and outputs as that in your microgrid model. Do help me with your workflow to take it further.

As a second thought, did you try using "Bus Creators" block for the purpose of combining the signals.

[Mounira]

hello, i did, i got the table and i imported the signals through the "selected signals" button, and i used mix signals to get them, all is good, the problem after it, I fixed the question that i wrote, but then i got an error, my question was about the error, maybe i should change the question now

Answer 1

[akshatsood]

Hi @Kawsar,

Referring to the discussion within the comment thread, it appears you have come across the error "Plant model cannot be a pure direct feed-through system without additional dynamics." To rectify this issue, I recommend consulting the comprehensive solution provided by the MathWorks Support team. Implementing the suggested workarounds should help resolve the problem.

https://in.mathworks.com/matlabcentral/answers/1957259-why-does-the-error-message-plant-model-cannot-be-a-pure-direct-feed-through-system-without-addition

Additionaly, do refer to the following answers if the above workarounds do not work for you

https://in.mathworks.com/support/search.html/answers/473358-mpc-controller-direct-feedthrough-from-manipulated-variables-to-any-output-in-plant-is-not-allowed.html?

https://in.mathworks.com/matlabcentral/answers/511317-mpc-cannot-be-designed-in-app

Should the problem continue to occur despite following the suggested solutions, please inform me as this will enable me to conduct a thorough examination of the problem and work with you to identify a more effective resolution.