In English

Time Alignment of Engine Emission Measurements

Guo Dong
Göteborg : Chalmers tekniska högskola, 2015. Diploma work - Department of Applied Mechanics, Chalmers University of Technology, Göteborg, Sweden, ISSN 1652-8557; 2015:15, 2015.
[Examensarbete på avancerad nivå]

Generally, steady state engine emission measurements are used to evaluate exhaust, which is far from the real condition. When driven in urban environments, the vehicle is affected by traffic lights, bus stations, pedestrians, etc. Therefore, transient driving states account for 50% - 90% of all working conditions, in which 30% - 50% are acceleration. Vehicles are, most of the time, working at starting, accelerating, decelerating and such transient states, and the air flow as well as the fuel flow rates are changing simultaneously. This will cause obvious change of the exhaust emissions such like hydrocarbon (HC) and nitrogen oxide (NOX). Thus the emission parameters measured under steady state cannot represent the real exhaust level, which makes it necessary to study the transient emissions and measuring technique. One cornerstone of dynamic emissions testing is the time alignment between measured emissions and exhaust flow from the engine. In the future, emissions development on engines will increase and thus increases the importance of time alignment. This means that new more accurate methods needs to be developed to increase accuracy. In this study, 15.77% v/v CO2 is dosed at the empty reactor setup, thus to observe the convolution effect of the exhaust system. In the absence of any catalytic or reactive material that interact with the gas phase species, the same step function is expected to be observed as the measured signal in the outlet. Nevertheless, the result is often a monotonically increasing function of time which slowly reaches the inlet concentration after certain elapsed time. Hydrodynamic dispersion is the reason of this phenomenon, because it will increase the residence time of fluid elements. Consequently, signal distortion will also affect the results of reactive experiments, and therefore it is necessary to characterize the reactor setup with respect to these perturbing effects in order to properly interpret the results of transient experiments. Having this fulfilled, the next step would be to propose a method to retrieve the true response of the reactor system for quantitative data analysis. The residence time distribution of the detector and exhaust system was analyzed, thus to get the transfer function of the instruments. Through deconvolution of the step response and sinus response under certain conditions, we got the transfer function under various exhaust flow. Thus we can get the actual output emissions from the measured data. In this project we did various experiments under certain conditions, and in the next step we need to verify the usage of deconvolution under random conditions. So the methodology we developed in this thesis could not be used in reality for now, as more works need to be done in the future.

Nyckelord: Time alignment, concentration measurements, transient experiments, deconvolution, engine, emissions, exhaust



Publikationen registrerades 2015-11-24. Den ändrades senast 2015-11-24

CPL ID: 226229

Detta är en tjänst från Chalmers bibliotek