Disturbance rejection using Model Predictive control for pneumatic actuator system

Abstract

`Disturbance' is one of the main reasons why the position control in pneumatic system is difficult to achieve. Disturbance can cause the process (controlled variable) to deviate from their respective set-point and at the same time tracking the set-point is quite impossible. The controller is said to be good if it has the ability to reject the disturbance, as this will ensure the stability of the controller itself. This paper presents the performance analysis of Model Predictive Controller (MPC) to reject the disturbance added into the system while controlling the position of cylinder's stroke for pneumatic actuator system. The disturbance was added into the system in order to test the performance of the controllers in rejecting the disturbance. The input constraint was also applied to the MPC algorithm to make the controller operates closer to the limits. In this paper, the performance of MPC in rejecting the disturbance as well as controlling the position of the pneumatic actuator system was compared with Proportional-Integral (PI) controller. Simulation results show the constrained MPC is more effective than unconstrained MPC and PI controller in producing better transient response and minimizing error.