Optimization of combustion chamber geometry for stoichiometric diesel combustion using a micro genetic algorithm

Abstract

This paper describes the optimization of combustion chamber geometry and engine operating conditions for stoichiometric diesel combustion, targeting lower gross indicated specific fuel consumption. The KIVA code, coupled with a micro genetic algorithm population of nine for each generation was used. The optimization variables were composed of ten variables related to the combustion chamber geometry and engine operating conditions. In addition, an auto mesh generator was developed for generating various kinds of combustion chambers, such as open-crater, re-entrant, deep, and shallow types. In addition, the computational models were validated against the experimental results for a stoichiometric process in terms of the combustion pressure history and emissions. Through the preset optimization, a 35% improvement in the gross indicated that specific fuel consumption was achieved. In addition, the optimization results showed that the optimum engine operating conditions employed a premixed charge compression ignition combustion regime with early injection and a narrow spray included angle. Furthermore, a higher boost pressure was used to prevent fuel film formation.