Control of stable, integrating and unstable processes by the Modified Smith Predictor

The Modified Smith Predictor (MSP) is designed, and in the literature classified, as a Dead-Time Compensator for integrating processes. In the present paper it is shown that the MSP is a PID controller in series with a second-order filter, defined by the dead-time and an adjustable parameter. Optimization of the regulatory performance of this controller is performed under constraints on the robustness and sensitivity to measurement noise. Excellent performance/robustness tradeoff is obtained for stable, integrating and unstable processes, including dead-time, as confirmed by simulations and by experimental result obtained on a laboratory thermal process.

► Modified Smith Predictor is a PID controller in series with a second-order filter.
► Anti-windup implementation is straightforward.
► Excellent performance/robustness tradeoff is obtained by constrained optimization.
► Simulations and experimental result on a laboratory thermal process are presented.