Mathematical modelling of spark ignition engine combustion

TLDR: In this paper, a classification of model types is constructed on this basis and the mathematical procedures involved in their development are outlined, and the most detailed, multi-dimensional (MDM) models have the potential to resolve the turbulent, temperature and species fluctuations both spatially and temporally.

Abstract: Spark ignition engine combustion can be modelled with a varying degree of detail depending on how the governing equations are formulated and how the turbulent flow is considered to affect the flame propagation process. A classification of model types is constructed on this basis and the mathematical procedures involved in their development are outlined. The simpler, ‘zero-dimensional’ models invariably incorporate “phenomenological” sub-models to describe particular aspects of the combustion process and are especially suitable for parametric studies. They evaluate engine performance satisfactorily and have been instrumental in determining the mechanisms of formation/destruction of certain exhaust pollutants. The most detailed, multi-dimensional (MDM) models have the potential to resolve the turbulent, temperature and species fluctuations both spatially and temporally. Currently, they are restricted in this by computer storage and cost constraints and by inadequate sub-grid scale models for turbulence and chemical reactions. “Hybrid” models retain a little of the detail and spatial resolution of MDM models but without the associated computational cost difficulties. These couple the momentum conservation equation with certain of the “phenomenological” sub-models in the “zero-dimensional” classification and could evolve as the most suitable and cost effective for spark ignition engines.