Modeling hot-running carbon monoxide emissions: A comparison of speed-based and engine-based approaches.

摘要

The Clean Air Act of 1990 place a great deal of importance an the use of transportation controls to meet air quality standards, in non-attainment areas. The approach to estimating the positive and negative impacts of transportation measures requires estimation of current emissions in an emissions inventory, followed by a determination of what changes in emissions would occur given changes in the operation and use of the transportation system.; Research suggests that current models significantly underestimate mobile source emissions, thus there are many improvements to the modeling approach and structure that should be considered. One area where improvements{09}to the current modeling approach could have significant impact is the manner in which hot running exhaust emissions are defined as the pollutants that exit the tailpipe of a fully warmed-up vehicle, and exclude evaporative and engine-start emissions. The current MOBILE-5a approach to hot running emissions is based on the average emissions from a portion of the Federal Test Procedure (FTP) for various model year and vehicle classes. A derived relationship between the average trip speed and emissions ratios is employed to correct the baseline FTP exhaust emission rate for different average operating speeds. This work shows that average speed has deficiencies in modeling hot running trip emissions, and agrees with previous research in this area.; This research develops and evaluates five alternative models to the existing average speed modeling approach. Emissions are predicted as a function of: (1) {A0}speed on a second-by second basis, (2) {A0}speed and acceleration on a second-by-second basis, (3) {A0}proportion of a trip with activity above a throttle-based enrichment trigger, (4) {A0}throttle position on a second-by-second basis, and (5) {A0}a full engine-map model utilizing throttle, engine speed, and manifold absolute pressure. The average-speed model is also evaluated for purposes of comparison. Speed-based models explain, at best, one third to one half of the variation in trip emissions, which could be due to a poor ability to predict exhaust carbon monoxide (CO) concentration, as well as a limited ability to predict mass air flow. Engine-based models showed a substantial increase in modeling capability over the speed-based models. The new models are based upon limited data collected from a variety of vehicles during the Federal Test Procedure Improvement Project. Although the data are limited, the conclusions lead to new and improved comprehensive modeling approaches. Recommendations for further research efforts are provided.