Fitness Diversity Based Adaptive Memetic Algorithm for solving inverse problems of chemical kinetics

This paper proposes the fitness diversity based adaptive memetic algorithm (FIDAMA) for solving the problem of the inverse type consisting of retrieving chemical kinetics reaction rate coefficients in the generalised Arrhenius form based on the observed concentrations in a given range of temperatures of a limited set of species which describe the reaction mechanism. FIDAMA consists of the evolutionary framework and three local searchers adaptively governed by a novel fitness diversity based measure. Moreover, a certain simplification of the decision space was carried out without any deterioration in the result obtained. The numerical results preseted show the superiority of FIDAMA compared to the other published computational intelligence methods.