A system-level, data driven model was developed to predict gas temperature in the exhaust manifolds of naturally aspirated spark ignited engines during vehicle operation. The model is based on data gathered from 67 vehicle tests.
The data were collected over the last few years, from a dozen cars and trucks, spanning a range of rated power from 127 to 350 hp, engine displacements from 2 to 8 liters, Line-4, V-6 and V-8 engine configurations, vehicle mass from 1500 to nearly 9000 kg, trailer mass from zero to nearly 4000 kg, different vehicle drive schedules, different vehicle speeds, varying road grades up to a maximum in excess of 9% and ambient temperatures of 40°C.
The large number of engine and vehicle design and operational variables that can influence exhaust gas temperature was limited to high-level variables known early in a vehicle development program. The variables were then combined into five dimensionless parameter groups and an expression relating these parameters was obtained.
Though the model is empirical by nature, the diversity of data used in its' creation and the use of dimensionless groups should make the model widely applicable to cars and trucks.