Model predictive control based on Q-learning for magnetic levitation platform system

• DOI: 10.3233/jae-240003
• Publication Date: 2024-08-09T16:33:26Z
• Authors: Zhihao Ke; Huiyang Yi; Penghui Zhang; Yuexin Feng; Le Liang; Zigang Deng
• Keywords: Model Predictive Control

A model predictive control (MPC) method based on Q-learning algorithm, named QMPC, is proposed for weakly damped, nonlinear and open-loop unstable magnetic levitation platform (MLP) systems. In addition, the design of MPC controller MLP system, state space system airgap, action horizon horizon, reward punishment function are also included in this research. Based Simscape MATLAB/Simulink, joint simulation realized. Compared with PID traditional controller, results show that QMPC has better disturbance rejection ability tracking performance under six working conditions.