A simulation study of an adaptive Smith predictor

Smith predictor is an effective compensation mechanism for large pure time delay, However, its poor robustness limits its practical application. This article makes a study of the robustness of Smith predictor from the view of system simulation. Here an adaptive mechanism is used to overcome the weak point of Smith predictor, i.e. updating the process model that Smith prdictor needs from the on line system identification. Finally, correlative simulation results are given to show the control effects of the adaptive Smith predictor system.