Automatic reduction of detailed combustion mechanisms using particle swarm optimization, differential evolution and angular modulation algorithms: application to Dimethyl Ether/air combustion

Mathematical modeling is used as a tool to predict the combustion behavior of fuels. The use of detailed chemical kinetics mechanisms in combustion models increases computational time. Metaheuristic optimization algorithms are one of the methods used to reduce the detailed mechanisms. The purpose of this paper is to investigate the possibility of using continuous metaheuristic algorithms in binary space to reduce the combustion mechanism of dimethyl ether fuel. For this purpose, particle swarm optimization and differential evolution algorithms have been used in a combination with angular modulation to map the continuous space to binary one and reduce the dimensions of problem from 351 to 6. Finally, a detailed mechanism with 79 species and 351 reactions is reduced to 17 species and 43 reactions. It has been showed that the reduced mechanism predicts the results of detailed mechanism in constant pressure and laminar premixed flame reactors very well with maximum error less than % 0.95.
 
.