Application of model based predictive control to acrylnitryl polymerization reactor

An idea of long-range model-based predictive control is used to formulate an algorithm for locating the closed-loop system poles. The algorithm is modified in order to cope with nonlinear plants by means of a combination of nonlinear prediction with tuning parameters, calculated on the basis of a linear model. The resulting controllers are applied to a complex process namely an acryl-nitryl polymerization reactor. The performance is compared with the well-known general predictive control (GPC).<
>