Vehicle emission models using Australian fleet data

Dia, Hussein and Boongrapue, Noppakun (2015) Vehicle emission models using Australian fleet data. Road and Transport Research, 24 1: 14-26.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Author Dia, Hussein
Boongrapue, Noppakun
Title Vehicle emission models using Australian fleet data
Journal name Road and Transport Research   Check publisher's open access policy
ISSN 1037-5783
Publication date 2015-03
Year available 2015
Sub-type Article (original research)
Open Access Status Not Open Access
Volume 24
Issue 1
Start page 14
End page 26
Total pages 13
Place of publication Vermont South, VIC, Australia
Publisher ARRB Group
Collection year 2016
Language eng
Formatted abstract
This paper presents the development of second-bysecond vehicle emission models under hot stabilised settings and variable speed, acceleration, air-tofuel ratio and torque conditions using Australian fleet data. The models were developed using 27 passenger and light commercial vehicles with more than 64 500 instantaneous observations obtained from dynamometer tests. Both neural network and regression analysis models were developed to predict fuel consumption and pollutant emissions at different levels of speed, acceleration, air-to-fuel ratio and torque. The results showed that modelling individual vehicle emissions provides superior results to an aggregate modelling approach in which a 'characteristic' vehicle is used to represent dissimilar vehicle populations. The best performing models from the two modelling approaches showed high degrees of correlation between predicted and actual data (e.g. 96%-98% accuracy for prediction of fuel consumption, 85%-87% accuracy for prediction of hydrocarbons, 70%-90% accuracy for prediction of carbon monoxide, and 88%-97% accuracy for prediction of nitrogen oxides (NOx)), depending on the type of passenger and light commercial vehicles being modelled. Finally, the paper highlights some of the advantages and shortcomings of the two modelling approaches when considering a universal model to predict multiple emissions outputs. The models developed in this study will provide researchers and practitioners with a better understanding of the environmental impacts resulting from a wide range of transport schemes in the Australian context.
Keyword Artificial Neural-Network
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status Non-UQ

Document type: Journal Article
Sub-type: Article (original research)
Collections: School of Civil Engineering Publications
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in Thomson Reuters Web of Science Article
Google Scholar Search Google Scholar
Created: Sun, 13 Sep 2015, 00:17:48 EST by System User on behalf of School of Civil Engineering