MODELLING EMISSIONS FROM AN INSTRUMENTED CAR

There are a number of ways of using a full chassis dynamometer to determine emissions from a vehicle. Some methods have been discussed in Watson et al (1983). One method of determining emissions is to use the dynamometer to simulate a predefined drive cycle. The emissions are collected in a bag and the amount of emissions and the components of the gas are determined. The advantage of doing this is that simulation of traffic conditions can be strictly controlled, however the correlation between dynamometer test results and actual on road results are not easily quantifiable (Taylor and Young 1988). Another method is to use a chassis dynamometer to determine the various emission levels of the vehicle at various levels of engine power and speed. The vehicle will then be driven in real traffic conditions, determining the vehicle emissions. The benefit of this method is that the results are more likely to reflect vehicle emissions from a vehicle under the influence of real traffic conditions. Such a method has been used widely used by researchers (Watson et al 1985, Taylor and Young 1996, Taylor et al 1995, Dentonkelaar 1994). Watson et al (1985) use the dynamometer to determine fuel consumption equations, leading to the determination of coefficients to derive equations for estimating vehicle emissions which can be applied to a vehicle driving on the road within a traffic stream. Enhancements have been made to the models derived to take into account the variability encountered in determining emissions (Sheikh 1991, Watson et al 1992). The method described in this paper is using the dynamometer to derive an engine map of a vehicle, which is similar to the approach taken by Watson et al (1985). The engine map is a three-dimensional graph with manifold pressure (x axis) and engine speed (y axis) in the two- dimensional plane, and the emissions levels in the z-coordinate. These three dimensional maps were also derived for emissions of CO, CO2, HC, and Nox. They all generally have the same characteristic in that the higher the rpm and lower the manifold pressure the higher the emissions produced. This method has been proven to be affective in determining vehicle emissions under real traffic situations (Taylor and Young 1996, Taylor et al 1995, De Maria 1993a, De Maria 1993b, De Maria 1994). The engine mapping model of the TSC research vehicle will constitute the main thrust of paper.