Predictive PI versus Smith control for dead-time compensation

Abstract

In this paper, the design and analysis of a predictive PI controller, capable of dead-time compensation, and which overcome certain deficiencies of Smith control, is presented. The design is based on a first-order plus dead-time model of the actual process, which is representative of the majority of process dynamics encountered in the industry. Only simple classical closed-loop specifications are required from the user. Simulation examples, including one based on an analog process simulator, are provided to highlight the principles of the proposed scheme and to compare its performance with Smith control. © 2001 Elsevier Science Ltd.