A genetic algorithm for generating initial parameter estimations for kinetic models of catalytic processes

The application of genetic algorithms (GA) for generating initial parameter estimations for the kinetic model of a catalytic process (oxidative methane dehydrodimerization) is described. The aim was to provide suitable starting points for the applied combination of an integration process and a locally converging nonlinear parameter estimation algorithm. The influence of the control parameters of this GA (number of individuals, the mutation rates and the selection methods) was studied. Additionally, studies of kinetic model parameters for two different PbO/alumina catalysts were carried out. The suggested method can be used to estimate suitable values for the model parameters of a complex mathematical model. (C) 1996 Elsevier Science Ltd