Design of Model Predictive Control to stabilize Two-Stage Inverted Pendulum

摘要

Model predictive control (MPC) defines as a controller algorithm utilizing optimal computation. The method of MPC has given numerous effect to the quality of some applications such as petrochemical plant, electronic devices, power electronics, robotics, and unmanned aerial vehicle. All of the MPC design in the literature has been successfully implemented and have given excellent performance for the systems. However, the MPC has been only designed for systems which have state variables under four variables. In this paper, an MPC method for the system that has state variables over 4 variables is proposed. The proposed MPC has been designed to stabilize two-stage inverted pendulum that has higher-order, nonlinear, very unstable, multivariable and 6 state variables. The MPC method is designed through the dynamic model of that pendulum employing Euler-Lagrange Equation. Then, the method is evaluated under several conditions to Figure out the performance using MATLAB Simulink software. The result shows that the three parameters of the system namely the parameter prediction horizon Np, the parameter tuning r <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">w</sub> and the parameter control horizon Nc can influence the system output. The modification of N <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</sub> has influenced the speed of the output system. The modification of N <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">c</sub> and r <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">w</sub> has influenced the maximum of pendulum angle deviation.