An evaluation of flame surface density models for turbulent premixed jet flames

We report a computational study of a turbulent premixed jet flame stabilized in a laminar coflow of a reactant mixture at the same stoichiometric ratio as the turbulent flame. The performance of four different flame surface density (FSD) models, earlier used by Duclos et al. [1] for studying flame propagation in constant density frozen turbulence, and that of the reaction progress variable approach of Bray and Moss (BM) is evaluated by comparison with existing experimental data. We also describe a sensitivity analysis of predictions to inlet turbulence quantities and inlet flame surface density. The predictions of the Mantel and Borghi (MB) model are observed to be highly sensitive to inlet turbulence quantities, while those of the coherent flame (CF) model and BM models are less sensitive. The mean velocity and temperature predictions are insensitive to the inlet FSD distribution because of the strong production and dissipation terms in the transport equation for FSD. The predictions of mean velocity and temperature profiles are compared with the experimental data of Chen et al. [2]. Among the FSD models, the CF and MB models gave good estimates of mean velocity and temperature. The Cant, Pope, and Bray (CPB) model overpredicted the mean temperatures slightly. The flame surface density model of Cheng and Diringer (CD) predicted too high temperatures. This is shown to be resulting from an overprediction of the generation of flame surface density. All the models predict the increase in flame brush thickness with distance from the injector exit, but the MB model provides the best quantitative estimates of the flame brush thickness.