Modelling and Simulation of the Combustion of Ethanol blended Fuels in a SI Engine using a 0D Coherent Flame Model 2009-24-0016

Motor fuels are today increasingly blended with oxygenate components to reduce global CO₂ emissions. Among these components, biomass-derived ethanol is very popular for spark ignition engine operation as it is not only a renewable source of energy, but it also allows to increase the engine power and thermal efficiency. Indeed, ethanol has the advantage of a high latent heat of vaporization leading to the so-called “cooling effect” which allows to increase the air-mass flow rate in the engine while reducing the charge temperature. This last property of ethanol combined with its high octane index make the engine less sensitive to knock. Then, although ethanol is characterised by high combustion speeds, optimal values of spark advance can be maintained on a larger range of engine operating conditions and high compression ratios as well as increased levels of downsizing can be used, all these aspects contributing to improve fuel consumptions.

However, the real potential of ethanol blended fuels still has to be explored and their impact on engine control strategies has to be investigated, especially considering the possible fuel composition variability during the engine life. Both issues can nowadays be addressed at low cost using system simulations of the whole engine, provided that the models used correctly account for the effect of the fuel composition on combustion processes.

This paper deals with the extension of a 0-dimensional coherent flame model to the combustion of ethanol blended gasoline for the simulation of heat release, knock and pollutants in SI engines. This extension mainly relies on the combination of a new laminar flame speed correlation, a modified set of chemical reactions in the flame front and an adapted correlation for the knock delay. The proposed developments are validated on a wide experimental database including many engine operating conditions as well as ethanol volume fractions ranging from 0% to 30%. Parametric variations in terms of spark advance and fuel/air ratio are also performed to compare optimal engine settings obtained from the simulations and at the engine bench. A good agreement is observed, showing the interest of using system simulations to predict the influence of the fuel composition on SI engines operation.