Constraint optimization of dead-time processes using Smith-Predictor

This paper presents a modified Smith-Predictor to optimize disturbance-rejection performance with constraints on robustness and reference-tracking. A 2-degree-of-freedom Smith-Predictor is used to deal with the set-point tracking and disturbances separately while ensuring a certain level of maximum sensitivity. So the paper covers three aspects of control system that are the set-point tracking, robustness and disturbance rejection. Constraint optimization algorithm is used to achieve the desired parameters of the controllers. A second order approximate model using Ziegler-Nichols dynamics characterization is used to get a generalized mathematical model for any process that includes dead time. Robustness is achieved by setting a maximum value of sensitivity function while taking optimized values of the controllers for the disturbance-rejection.