Simple tuning of modified Smith predictor for unstable FOPTD processes

Abstract

A simple tuning for modified Smith predictor is reported here to control unstable first-order plus time delay (FOPTD) processes. In practice, a good number of chemical processes (like combustion, evaporation, distillation, etc.) are unstable as well as delay dominated in nature. Modified Smith predictor technique is a well-accepted methodology for its simplicity and efficacy to regulate delay dominated processes to ascertain desirable closed-loop response. Two-degree-of-freedom (2-DOF) control scheme of the proposed multi-loop controller provides performance enhancement during both set point tracking and load recovery phases. To mitigate tuning complexity of more than one controller involved in modified Smith predictor, the proposed simple guideline involves the same set of tuning parameters for all three controllers. Distinctive feature of the proposed tuning scheme is that there is no overshoot during set point tracking as well as smooth and improved recovery is observed subsequent to uncertain load changes. Superiority of the proposed methodology is established through performance evaluation and stability analysis with well-accepted modified Smith predictor-based tuning relations.