MODELING OF NITRIC-OXIDE FORMATION IN SPARK-IGNITION ENGINES WITH A MULTIZONE BURNED GAS

A review of the literature on nitrogen oxide concludes that the thermal nitric oxide (NO) mechanism is the most relevant to the nitrogen oxide emissions from spark ignition engines. The review has also shown that the use of multiple burned gas zones is likely to be important for the accurate prediction of NO emissions. The multizone formulation developed here results in a particularly efficient formulation, as burning can occur within a zone. Previous multizone spark ignition engine simulations all appear to have created a new zone for each crank-angle increment during combustion. The present studies indicate that 5 or 10 burned zones are likely to be sufficient for most purposes. In reality there will be some mixing between the zones, and it could well be that with discrete burned zones, the most accurate simulation will be with a finite number of zones. The investigations with the multizone model showed lower predictions of NO than the single burned zone model; a result consistent with that of earlier workers. The outputs from the simulation have been compared with measurements from a gas engine, that encompass particularly wide variations in ignition timing (2-30 deg btdc) and nondimensional air fuel ratio (0.8 < lambda < 1.8). The earlier multizone NOx models had been compared with a limited range of mixtures close to stoichiometric, and in most cases the Minimum ignition advance for the Best Torque (MET ignition timing). The opportunity was also taken to compare the NOx predictions, resulting from using different recommendations for the reaction rates in the extended Zeldovich mechanism. Finally, mean cycle modeling and cycle-by-cycle modeling were compared. There was close agreement in NO predictions only in the region of the MBT ignition timing. The significance of cycle-by-cycle modeling should be investigated further, with carefully calculated burn rate data.