Alternatives for Multiobjective Optimization of a Polymerization Process

The optimal control policies for a polymerization process, particularly for the batch free radical polymerization of methyl methacrylate, have been determined using a multiobjective optimization technique. The process objectives considered in the optimization include monomer conversion, polydispersity index, polymerization degree and total reaction time, weighted combined in a scalar objective function. The decision variables are the initial concentration of the initiator and the temperature represented by several isothermal steps. For solving the optimization problem, several methods based on sequential quadratic programming and genetic algorithm were used and compared. Combining them into a hybrid method (genetic algorithm provides initial values for the traditional iterative method) led to the best results. This paper aims to be an approach of the polymerization process optimization problem, describing alternatives of formulating and solving this problem and emphasizing the importance of user decision in choosing solutions based on technological criteria.