A phenomenologically based computer model to predict soot and NO_x emission in a direct injection diesel engine

A phenomenological model based on the multizone method is described. The model requires a relatively low amount of central processing unit (CPU) time and memory for each run and, consequently, is useful for parameter studies. The flame morphology predicted by the model is compared to that recently uncovered in experimental studies. As a demonstration of its utility, the model is used to study the effect of the oxygen content in intake air on the emission of particulate matter (PM) and NO_x from a direct injection diesel engine. A parametric study reveals that while oxygen-enriched intake air is useful for reducing PM, the reduction comes with a large increase in NO_x emission. The PM-NO_x trade-off curve for the oxygenated intake air is worse than that for the baseline air case.