Numerical investigation of MILD combustion using multi-environment Eulerian probability density function modeling

Abstract

In the present paper, the flames imitating Moderate and Intense Low Oxygen Dilution (MILD) combustion are studied using the Probability Density Function (PDF) modeling approach. Two burners which imitate MILD combustion are considered for the current study: one is Adelaide Jet-in-Hot-Coflow (JHC) burner and the other one is Delft-Jet-In-Hot-Coflow (DJHC) burner. 2D RANS simulations have been carried out using Multi-environment Eulerian Probability Density Function (MEPDF) approach along with the Interaction-by-Exchange-with-Mean (IEM) micro-mixing model. A quantitative comparison is made to assess the accuracy and predictive capability of the MEPDF model in the MILD combustion regime. The computations are performed for two different jet speeds corresponding to Reynolds numbers of Re = 4100 and Re = 8800 for DJHC burner, while Re = 10000 is considered for the Adelaide burner. In the case of DJHC burner, for Re = 4100, it has been observed that the mean axial velocity profiles and the turbulent kinetic energy profiles are in good agreement with the experimental database while the temperature profiles are slightly over-predicted in the downstream region. For the higher Reynolds number case (Re = 8800), the accuracy of the predictions is found to be reduced. Whereas in the case of Adelaide burner, the computed profiles of temperature and the mass fraction of major species (CH4, H2, N2, O2) are found to be in excellent agreement with the measurements while the discrepancies are observed in the mass fraction profiles of CO2 and H2O. In addition, the effects of differential diffusion are observed due to the presence of H2 in the fuel mixture.