Genetic algorithms for optimisation of chemical kinetics reaction mechanisms

摘要

The aim of this paper is to review the techniques that exist in the literature for finding the optimal values of the reaction rates coefficients for a given reaction mechanism. Although traditional gradient based methods are reviewed as well, the paper focuses on recently developed self-adaptive evolutionary algorithms that outperform classical methods. Unlike the traditional, gradient-based methods one of the most important characteristics of computational intelligence techniques, such as genetic algorithms (GA), is the effectiveness and robustness in coping with uncertainty, insufficient information and noise. In this approach minimum human effort and little insight into the details of the chemical mechanism is required to generate the optimal values for the reaction rate coefficients. The use of the GA inversion procedure is illustrated on chemical systems of various complexities which govern the combustion of hydrogen, methane and kerosene.