OCR Text |
Show the recent years, N O x modelling in turbulent diffusion flames remains a subject for intense fundamental research. Traditionally, four NO kinetic formation mechanisms have been considered^5]; the thermal or Zeldovich mechanism, the Fenimore-prompt mechanism, the N2O intermediate mechanism and, for fuels with bound nitrogen, the fuel-NO mechanism. These mechanisms are mainly controlled by radicals concentration in the flame, thus modelling the departure from equilibrium of radical concentrations in flames is considered of major importance. Modelling N O formation in turbulent flames is a challenging task, which needs to take into account sufficiently detailed kinetics for non-equilibrium, prompt-NO or fuel-NO chemistry, as well as flame radiation and turbulence effects. Primary models have been developed by Peters^6-7], Kent and Bilger^, etc., assuming equilibrium chemistry. These models consider a conserved scalar-the mixture fraction, to describe the mixing of the reactants. The chemical composition and the flame temperature are thus directly related to this scalar. The effect of the fluctuations in the flow field is calculated using a beta function for the Probability Density Function (PDF) of the mixture fraction. Alternative turbulent closure models^] have been also used, and non-equilibrium effects have been modelled with different levels of chemical complexity, intermediate between the partial equilibrium two step assumption!10] and the detailed kinetics approach.!11] This effort presents the fundamental aspects of a model developed at Air Liquide, and compares the results obtained with this model to experimental measurements, as well as with alternate modelling results. 3 |