A Comparative Study of Premixed Turbulent Combustion with PDF and RANS Methods

A methane-air turbulent premixed flame is simulated via probability density function (PDF) and Reynolds averaged Navier-Stokes (RANS) methods. In the PDF approach, molecular mixing is modelled through the modified Curl’s model. A Monte Carlo method is used to solve the PDF transport equation. Also, the run time averaging and local time stepping procedures are incorporated to increase the accuracy and reduce the computational time of the PDF simulation. In the RANS approach, the averaged chemical reaction rate term is modeled by the eddy breakup-finite rate model. A finite difference discretization on a staggered grid is utilized to obtain the numerical solution for the RANS equations. The characteristics and differences of the two above mentioned methods, including computational time and predicted mean fields, are investigated in detail for premixed flames. It is observed that the discrepancy of the predicted mean fields between the two methods is large especially in regions near the flame. In addition, the predicted flame length by the PDF method is approximately half the flame length predicted by the RANS method.