The proportional-integral controller design based on a Smith-like predictor for a class of high order systems

Abstract

Heat transfer and fluid flow play important roles in industrial processes, especially in chemical and thermal processes. Their dynamics are often modelled as high order systems with the type of $K/{(\mathrm{Ts}+1)}^n$ , which have slow responses to the set point and the input disturbance. To enhance the tracking and disturbance rejection performance of these systems, a control structure combining the proportional-integral (PI) controller and Smith-like predictor is proposed in this paper. The tracking and disturbance rejection performance of the proposed control structure with the order mismatch are analyzed. In addition, the precondition where the proposed control structure can obtain satisfactory disturbance rejection performance is deduced. By analyzing the influence of PI parameters on control performance, an empirical tuning rule, which can balance the control performance and robustness constraint well, is summarized and the corresponding tuning toolbox is developed. Finally, the superiority of the proposed control structure is verified by simulations and comparative experiments based on Peltier temperature control platform.