LES/FDF studies of probability distribution of mixture fraction in inhomogeneous inlet turbulent jet flames*

Mixture fraction is a key parameter that can be used for quantifying and modeling mixing in turbulent non-premixed, partially premixed, and stratified flames. It is a central quantity in many existing combustion models such as the flamelet models and the conditional moment closure. Developing capabilities to accurately predict mixture fraction is thus crucial for turbulent combustion modeling. Focus in the past has mainly been put on the prediction of the low-order moments such as the mean and variance of mixture fraction. The prediction of the full statistical information of mixture fraction has not been thoroughly studied. In this work, we examine the predictive capability of large-eddy simulation (LES)/filtered density function (FDF) for the probabilistic distribution of the mixture fraction. A series of piloted methane jet flames with inhomogeneous inlet flows is chosen as the test case which has detailed measurement data for the distribution of the mixture fraction. Different statistical distributions are examined such as the single-point PDF and the FDF. The capability of capturing these distributions by using LES/FDF is examined thoroughly.