DESIGN OF PID CONTROLLER FOR UNSTABLE TIME DELAYED SYSTEMS

Abstract

Proportional integral derivative (PID) controller forms an indispensable part in various industrial applications. In this thesis, the proposed control scheme is based on an alternative configuration of PID control structure and its optimal tuning, designed to deal with the problems of unstable systems, close-to-the-origin zero, time-delayed systems and non-minimum phase functions. The configuration has used PD part, placed in the inner feedback loop of the process while the, I (integral) part is placed in cascade. This I-PD configuration does not introduce unwanted zeroes and stabilizes the process. To get the tuning parameters of I-PD configuration two different tuning methods are proposed, i.e., linear quadratic regulator (LQR) and improved continuous cycling method. LQR is used as an optimal tuning approach, to tune the parameters of the I-PD control structure via determination of positive symmetric matrices P & Q in algebraic Riccati equation (ARE). Further, an improved continuous cycling tuning method is formulated for control of the unstable and time delayed systems. It involves the determination of controller parameters by solving phase angle and magnitude criteria for the system. Subsequently, I-PD controller is cascaded with system model and solved via magnitude and phase angle criteria to tune its parameters. An extensive comparative analysis has been undertaken with the existing techniques from literature to demonstrate the effectiveness and superiority of the proposed technique. The simulation results are a testimony to the efficacy of the proposed control techniques.