Comparison of well-mixed and multiple representative interactive flamelet approaches for diesel spray combustion modelling

The application of detailed chemistry to the computational fluid dynamics simulation of combustion process in diesel engines has many potentials, including the possibility to predict auto-ignition, diffusion flame structure, stabilisation and soot formation in a wide range of operating conditions, also taking into account the effects of different fuel types. Among the approaches that were proposed over the years, the ones that are mostly used in practical calculations can be divided into two main categories: the first assumes each cell to be a well-stirred reactor, while the second employs the flamelet assumption to describe both auto-ignition and turbulent diffusion flame propagation. Despite the fact that both types of model have been widely validated over the years, a detailed comparison between them appears to be very useful in order to understand better the relevant parameters governing auto-ignition, flame stabilisation and the formation of pollutant emissions. This work is focused on a comparison of two different combustion models that were recently implemented by the authors in an open-source code. The first assumes each cell to be a homogeneous reactor and neglects interaction between turbulence and chemistry, while in the second, multiple laminar flamelets are used to represent the structure of a turbulent diffusion flame. Suitable techniques for online reaction rate tabulation and chemical mechanism reduction are also incorporated, to make the use of bigger mechanisms possible (up to 150 species). The two models are compared and validated by simulating constant-volume diesel combustion in a wide range of operating conditions, including variations of ambient temperature and oxygen concentration. Comparison between the computed and experimental data on flame structure, auto-ignition and flame lift-off enables an understanding of the main relevant differences between the models in the way both auto-ignition and flame stabilisation processes are predicted. © 2013 Taylor & Francis