SEMICONTINUOUS COPOLYMER COMPOSITION DISTRIBUTION PREDICTIVE CONTROL USING A DOUBLE ANN MODEL STRUCTURE

Copolymer composition distribution (CCD) is essential for product qual ity in copolymer manufacturing. In this work, we implement a double AN N structure for the on-line one shot control of MA-VAc semi-continuous latex copolymerization system. The control strategy is firstly assumi ng that the system is operating in a semi-starved condition. The feedi ng rate of MA can only be adjusted once in a single batch. Based on an intermediate measurement, a hybrid ANN model, that combines the infor mation provided by the experimental data and theoretical model simulta neously, is implemented to predict the product quality at the end of t he batch. However, it also has been found that because of the effects of measuring error, implementing a double ANN structure is better than implementing a single ANN. A critical parameter is identified by the first ANN. The parameter, in turn, is used as an input of the second A NN, that is a hybrid model. Both the experimental and simulation studi es show that the proposed double ANN is superior to a single ANN struc ture. Besides, the experimental studies also show that the ANN model p redictive control is promising for the CCD control of a semi-continuou s latex system.