The study of predictive control system based on coincidence points technology

Abstract

Predictive control is actually widely used for industrial process control. A big number of points in the predictive horizon and control horizon are often chosen not only to ensure the quality of the dynamic process for slow processes but also to get a better robustness for the control system. But it will raise the computational effort of the controller. Blocking and coincidence points technology can reduce the computation time while keeping the length of the horizon, which consider a reduced number of points in the predictive horizon (coincidence points) as well as in the control horizon (blocking point). Base on step response, a new method of selecting coincidence points is proposed in this paper. The simulation results show that this method could not only achieve the objective of simplifying the calculation, but also have certain advantages in the fast performance and stability. Also, this method is simpler and more practical.