A New Smith Predictor Controller Design Based on the Coefficient Diagram Method for Time-Delay Systems

Industrial/chemical processes usually involve significant time delays. The responses of systems/processes with long time delays can feature high overshoot and oscillation due to phase lag. Moreover, parameter variations and external disturbances make controlling such systems more difficult. The Smith Predictor (SP) Controller structure and the Coefficient Diagram Method (CDM) are commonly used in the literature to ensure robust control performance. This study introduces a novel design approach combining the strengths of SP and CDM. This method proposes using a second CDM-based controller for disturbance rejection, while using a CDM controller for setpoint tracking. The approach was tested on three high-order time-delay plant models, accounting for parameter variations and disturbance effects. Results show that this method can achieve low overshoot, quick rise time, and short settling time in set-point tracking. Furthermore, it delivers robust control performance under conditions of parameter changes and external disturbances.