Modelsko prediktivno upravljanje elektromehaničkim sustavom kolica na klackalici

Ovaj rad se bavi implementacijom MPC (engl. Model Predictive Control) regulatora nad nestabilnim elektromehaničkim sustavom kolica na klackalici. Matematičko modeliranje sustava je provedeno korištenjem Euler-Lagrangeove metode te prevođenjem dobivenih diferencijalnih jednadžbi u oblik prostora stanja. Zatim je kreiran MPC regulator koji stabilizira promatrani linearizirani model te, uz primjenu MATLAB-a, se formiraju odzivi sustava. Zatim slijedi prikaz simulacije u kojoj se ilustrira učinkovitost MPC-a tako što se odziv lineariziranog modela uspoređuje s nelinearnim. Dodatno, prikazani rezultati se također uspoređuju s još jednom metodom upravljanja procesa (LQR – engl. Linear Quadratic Regulator) kako bi se ilustrirala učinkovitost MPC regulatora.This paper deals with the implementation of the MPC (Model Predictive Control) controller over an unstable seesaw-cart electromechanical system. Mathematical modeling of the system was performed using the Euler-Lagrange method and translating the obtained differential equations into state-space form. Then, an MPC controller is created that stabilizes the observed linearized model, and, with the application of MATLAB, system responses are formed. This is followed by a simulation showing the effectiveness of MPC by comparing the response of the linearized model to the nonlinear. Additionally, the presented results are also compared with another method of process control (LQR – Linear Quadratic Regulator) to illustrate the efficiency of MPC controllers