Optimum regulation of sampled-data processes

Publisher Website
Google Scholar
Abstract
An optimum regulator for a minimum-phase, stable, single-input single-output sampled-data process with delay and subject to disturbances is defined in terms of an inverse model and a predictor For the disturbances. It can be viewed as a modified Smith controller for dead-time processes. The optimum control system is discussed and criticized. It is shown that the use of exponential smoothing for prediction has the advantages of integral control, guaranteed controller stability and guaranteed system stability for appropriate choice of smoother parameter provided in the last case that the steady-state gains of the process and of an imperfect model are of the same sign.

This publication has 2 references indexed in Scilit: