Selection of appropriate constraints for dimension reduction in MILD combustion simulations via RCCE

Modeling MILD (Moderate or Intense Low-oxygen Dilution) combustion is well known to be computation- ally demanding due to the need of an accurate description of the chemistry, because the low Damköhler number. The present work investigates the potentials of Rate-Controlled Constrained Equilibrium for the dimension reduction of kinetic schemes for a burner fed with a mixture of ethylene and air emulating MILD combustion conditions. A method based on the Computational Singular Perturbation theory is proposed for the identification of the species to be retained in the reduced representation of the chemistry. A significant reduction was achieved by using only 10 species (out of 34 species in the full kinetic scheme) and 3 elements; the results from the reduced model were in very good agreement with those of the full kinetic scheme. The sensitivity of predictions to the turbulence and combustion models was also preliminary assessed, indicating the need of modifying the residence time constant of the Eddy Dissipation Concept for MILD combustion conditions.