Use of genetic algorithms in the optimization of free radical polymerizations exhibiting the trommsdorff effect

Abstract

The genetic algorithm (GA) is adapted and used to obtain optimal temperature histories for methyl methacrylate polymerizations. The reaction time is minimized, while simultaneously requiring the attainment of design values of the final monomer conversion and number average chain length. The technique is robust, and gives near-global optimal solutions. As such, it can easily be used for on-line optimizing control of free radical polymerization reactors in which the reaction is associated with the Trommsdorff effect. The results obtained from GA can be improved further if these are provided as initial guesses to a computer code using the Pontryagin minimum principle with the first order control vector iteration method.