Large-eddy simulation of piloted diffusion flames using multi-environment probability density function models

Two piloted partially premixed jet flames (Sandia Flames D and F) are calculated using large-eddy simulation. A multi-environment turbulent combustion model, which depicts the sub-grid probability density function (pdf) as a weighted summation of a small number of Dirac delta functions in composition space, is used for modeling of turbulence chemistry interaction. It was coupled with a 19-species reduced mechanism and an In Situ adaptive tabulation algorithm for chemical source term integration. The impact of number of environments is investigated in this study. While predictions of Flame D by both two- and three-environment models are in good agreement with experimental data, only the three-environment model correctly captures the strong extinction and re-ignition in Flame F. The current study highlights the capability of the multi-environment pdf model for modeling of turbulent flames that feature strong local extinction and re-ignition.