Near wall combustion modeling in spark ignition engines. Part A: Flame–wall interaction

Abstract

Research and design in the field of spark ignition engines seek to achieve high performance while conserving fuel economy and low pollutant emissions. For the evaluation of various engine configurations, numerical simulations are favored, since they are quick and less expensive than experiments. Various zero-dimensional combustion models are currently used. Both flame front reactions and post-flame processes contribute to the heat release rate. The first part of this study focuses on the role of the flame front on the heat release rate, by modeling the interaction of the flame front with the chamber wall. Post-flame reactions are dealt with in Part B of the study.The basic configurations of flame quenching in laminar flames are also applicable in turbulent flames, which is the case in spark ignition engines. A simplified geometric model of the combustion chamber was used to calculate the mean flame surface, the flame volume and the distribution of flame surface as a function of the distance from the wall. The flame–wall interaction took into account the geometry of the combustion chamber and of the flame, aerodynamic turbulence and the in-cylinder pressure and temperature conditions, through a phenomenological attenuation function of the wrinkling factor. A modified global wrinkling factor as a function of the mean surface distance distribution from the wall was calculated. The impact of flame–wall interaction was simulated for four configurations of the sparkplug position and length: centered and lateral position, and standard and projected tip sparkplug.Results show that the position of the sparkplug has a greater influence than its length. The heat release rate was significantly altered in the lateral sparkplug position, when running the model with flame wall interaction. With the projected tip sparkplug, the impact of the wall on the initial flame kernel was delayed, since the flame is not close to the cylinder head. The maximum pressure was reduced when using the flame–wall interaction model for all four configurations. The attenuation of the wrinkling factor and of the mean flame surface at the end of combustion was captured by taking into account the impact of the chamber geometry, while this is not the case in global models, which impose an exponential decay of the heat release rate. The measured values for the four sparkplug configurations gave consistent results in terms of cylinder pressure and heat release.