Effects of the turbulence model and the spray model on predictions of the n-heptane jet fuel–air mixing and the ignition characteristics with a reduced chemistry mechanism
2018-01-19T11:41:29Z (GMT) by
Reynolds-averaged Navier–Stokes simulations with an improved spray model and a realistic chemistry mechanism are performed for turbulent spray flames under diesel-like conditions in a constant-volume chamber. Comprehensive numerical analyses including two turbulence models (the renormalisation group k–ε model and the standard two-equation k–ε model) with different model coefficients are made. The distribution of the fuel mixture fractions is a very important factor affecting the combustion process. In this study, we also use the entrainment gas-jet model, modifications of the the spray model coefficient and two turbulence models to investigate extensively the influence of the gas-jet theory model on the fuel–air mixture process. First, a non-reacting case is validated by comparing the liquid-phase penetration and the vapour-phase penetration and also the mixture fractions at different axis positions. Second, approriate methods are confirmed according to accurate mixture fraction distributions to validate the combustion process. Because of the large number of species and reactions, the calculation of chemically reacting flows is unaffordable, particularly for three-dimensional simulations. Hence, the dynamic adaptive chemistry method for efficient chemistry calculations is extended in this work to reduce the computational cost of the spray combustion process when a reduced chemistry mechanism is used. The results show that, in the evaporation case, the gas-jet theory model can be used to obtain a relatively accurate fuel vapour penetration length with different influential factors and that improved numerical methods can effectively reduce the mesh dependence for the spray evaporation process. It is demonstrated that the Schmidt number Sc and the turbulence models significantly influence the mixture fraction distribution. Very good agreement with available experimental data is found concerning the ignition delay time and the flame lift-off length for different oxygen concentrations owing to the accurate fuel mixture fraction.