Experimental and numerical investigation of the structure of flames under engine conditions

Abstract

The turbulent combustion of premixed gases is investigated under engine conditions. The chemiluminescence of the CH radical generated by a two-dimensional flame is registered quantitatively with high spatial resolution at different times. A telecentric optical integration evaluates volume elements of identical state and directly provides adequately averaged quantities. The spatial distribution of the local mass burning rate and of the local flame surface density is obtained from the emission measurements. This analysis is based on the flamelet assumption and uses the global mass burning rate calculated from the pressure record as a calibration basis. It was found that the maximum of the flame surface density hardly depends on the turbulence intensity and that the increase of turbulent flame speed with increasing turbulence is essentially achieved by a growth of the thickness of the flame brush. The turbulent flame speed does not reach a quasi-steady state during the phase of free propagation of the flame under our experimental conditions. The measured flame surface density is compared to the results of a numerical simulation using the transport equation for the flame surface density of the coherent flame model (CFM). This model was not able to reproduce the qualitative trends of the measurements. In particular, the calculated maximum of the flame surface density significantly increased with higher turbulence intensities. whereas the flame thickness remained constant. Modifications of the CFM are suggested, leading to a qualitative agreement with the experimental observation.