NONLINEAR DYNAMIC MATRIX CONTROL WITH DATA RECONCILIATION ALGORITHM TO A POLYMERIZATION SYSTEM

The autocatalytic nature and multivariability of polymeric processes m ake the mathematical model process highly nonlinear. In addition, it i s necessary to consider that the measured variable is subject to diffe rent degrees to error and so must be reconciliated aiming to avoid deg radation of the process and system control performances. This work pre sents a procedure to establish a dynamic data reconciliation into a fe edback control loop obtained by minimization of special designed objec tive function. The proposed methodology is based on the use of nonline ar predictive control techniques - Nonlinear Dynamic Matrix Control (N DMC) - associated to data reconciliation procedures, which is based on Nonlinear Programming (NLP) techniques. The algorithm developed will be implemented computationally to a styrene polymerization system. The predictive control techniques are robust and when applied to polymeri c system allows to make the system to have desirable dynamic character istics.