Vehicular Emissions Models Using Mobile6.2 And Field Data

Regression models to predict vehicular emissions for different categories of vehicles for different pollutants are presented in this thesis. Vehicular emissions are affected by numerous variables which, among others, include speed, temperature, acceleration, deceleration, driving behavior and meteorological data. Regression models are developed based on data obtained from Mobile 6.2 and on-board emissions measurements. The U.S. Department of Transportation (US DOT) conducted sensitivity analysis of Mobile6 where they evaluated different parameters used to find the emission factors, such as vehicle miles traveled, speed, humidity, etc. The sensitivity analysis investigated the overall Mobile6.2 model behavior for various conditions. In the analysis, speed was observed to be the most significant variable for all emission types. In this thesis, the regression model for estimating the emission factor for different classes of vehicles for different pollutants considers speed as the predictor variable. CO2 emission rate is estimated in Mobile 6.2 in a very simplistic way. The CO2 calculations are based on the average fuel economy performance estimates built into the model or supplied by the user. For other pollutants, Mobile6.2 considers various factors, such as the ambient temperature, speeds, humidity, etc., but the CO2 emission rates are not adjusted for the speed, temperature, fuel content, etc. Therefore, in this thesis, a model is proposed for estimating the CO2 emission rate considering speed as the predictor variable based on the data obtained from on-board emission measurements. Finally, an analysis is performed to study the affect of acceleration and deceleration on the emission rates.