Evaluation of combustion models based on tabulated chemistry and presumed probability density function approach for diesel spray simulation

Abstract

Two combustion models of different complexity have been implemented in a RANS solver in the CFD platform OpenFOAM®. Both models rely on the flame prolongation of ILDM (FPI) method, which allows the use of detailed chemistry mechanisms at relatively low computational costs. The homogeneous auto-ignition (HAI) model, directly using the chemical data from the FPI tabulation, does not take into account subgrid turbulence-chemistry interaction. Therefore, the second, more advanced model combines the FPI method with a presumed conditional moment approach. This auto-ignition-presumed conditional moment (AI-PCM) model accounts for the fluctuations of the mixture fraction and the progress variable caused by the turbulent flow. Both models have been evaluated by means of a parametric study of a single diesel spray at varying initial temperatures and oxygen concentration levels. The results obtained with the CFD models have been compared with experimental data from the engine combustion network (ECN). The comparison of the two models demonstrates the important role of the subgrid turbulence-chemistry interaction on the accuracy of the auto-ignition process and the diesel flame structure, as indicated by the agreement of the AI-PCM predictions with the measured data.