Optimization of rate coefficients for simplified reaction mechanisms with genetic algorithms

A general procedure for determining optimum rate coefficients of simplified kinetic mechanisms is presented. The optimization's objective is to match heat release or net species production rates of the simplified and an underlying detailed kinetics mechanism. A genetic algorithm is employed to carry out the matching procedure with a minimum requirement of human effort and expertise. Applications of optimized two- and three-step schemes to lean- premixed laminar methane flames show very promising results: Profiles of temperature and main species and also the peak values of intermediate species match those obtained with the detailed kinetic mechanism very well; flame speeds have been reproduced with good accuracy. The optimized mechanisms have proven to be numerically robust and efficient; the flexibility and ease of use make the approach presented here particularly appropriate for the numerical modeling of combustion in situations of technical interest.