A Stochastic Model of Turbulent Mixing for the Prediction of Burn Rate in a Spark-Ignition Engine

Abstract

A model for predicting the burn rate of a premixed and stratified internal combustion engine is developed. The model involves an ignition period dominated by mslecular and turbulent diffusion and a combustion period dominated by turbulent diffusion.

A correlation based on a deterministic model for the structure of turbulence is used. This correlation employs turbulent flow characteristics and the laminar flame speed. A statistical model based on turbulence characteristics which employs a Monte Carlo technique to follow the mixing process is developed to describe the combustion period. This model is used to predict the effects of spark advance, exhaust gas recirculation, and stroke variation on the burn rate of a premixed charge engine with no intake swirl. The effect of mixture stratification is also explored.

The model agrees well with the experimental data and predicts the correct trend for the effect of spark advance, exhaust gas recirculation, and stroke variation. With increased mixture stratification, the model predicts a reduction in available work.