Optimization of a Hydrocarbon Fuel Ignition Model using Genetic Algorithms

The autoignition predictions of the Shell hydrocarbon fuel ignition model have been improved for initial conditions similar to those in diesel engine environments. Modifications made to the model include a more accurate calculation of the heat release of the fuel, a new mass balance for the products of the termination reactions and revised enthalpies of the Shell model radical species. In addition, the twenty-six kinetic parameters of the model were optimized using a genetic algorithm methodology guided by auto-ignition results obtained from a detailed kinetic mechanism. The optimization was performed for a broad range of conditions that is representative to the operating conditions in diesel engines at the start-of-injection (SOI) with equivalence ratios from 0.5 to 4.0, initial pressures from 40 to 120 bar, initial temperatures from 650 K to 1175 K and EGR percentages from 0 to 75 %, for two single component hydrocarbons species, nheptane and tetradecane. Finally, the model was implemented into the KIVA -3V CFD code in order to access its agreement with experimental data for engine applications. The results were found to be promising in both cases of short and long ignition delays.