Soot Prediction in an Aircraft Combustor at Realistic Operation Conditions

This article presents numerical simulations of the Rolls-Royce BR700 combustor operating at various realistic conditions. Emphasis is put on the prediction of soot emissions. Three-dimensional steady Reynolds-averaged Navier-Stokes (RANS) simulations were performed employing the kϵ-model for turbulence closure. Combustion is modelled by finterate chemistry and the turbulence-chemistry interactions are captured by an APDF-approach (assumed probability density function) for temperature fluctuations. The injection of the liquid fuel Jet A-1 is described by a spray model applying Lagrangian methods for spray transport and atomization. The multi-component fuel is modelled as surrogate of n-decane, iso-octane and toluene. Reaction kinetics are described by a detailed mechanism, which is optimised for Jet A-1 oxidation and accurately resolves the reaction paths up to the smallest aromatic soot precursors benzene and toluene. Heavier PAHs (Polycyclic Aromatic Hydrocarbons) are lumped to sections and hence, modelled by a sectional approach, which leads to soot nucleation. The soot particle dynamics are described by a two equation model. Due to the model’s efficiency feasible computational costs are realised. Overall, four operation points are investigated, which are take-off, climb, approach and taxi. The simulation results at the combustor exit are compared to the experimentally determined Smoke Number of the engine exhaust gas by means of empirical correlations. The comparison shows a very good agreement. In particular, the Smoke Number’s trend is predicted well with respect to the thrust of the operation points.