Several new designs for PID, IMC, Decoupling and Fuzzy Control

Abstract

This thesis is devoted to the development of new control design techniques for better performance or robustness with relatively simple controller or structure. Firstly, three new PID controller design approaches are developed. In the first method, a robust PID controller design to achieve gain and phase margins is proposed. In the second approach, a new criterion for robust stability and robust PID tuing are developed with both the gain and phase information. In the third work, algorithms based on the iterative LMI technique are developed to design PI controller for processes with state time delay. Moreover, two new approaches for robust IMC controller design, one in time domain and another in frequency domain, are proposed. Further, a simple and effective design is proposed based on our new modified Smith predictor structure for unstable process with dead time. Next, a new necessary and sufficient condition for the solvability of the decoupling problem with internal stability is deveoped and a new decoupling controller design via LMI is presented. Finally, a TS fuzzy modelling method and two fuzzy control strategy are discussed.