ePrints

Predicting real-time roadside CO and NO₂ concentrations using neural networks

Lookup NU author(s): Professor Margaret Carol Bell CBEORCiD

Downloads

• Published version [.pdf]

Abstract

The main aim of this paper is to develop a model based on neural network (NN) theory to estimate real-time roadside CO and NO2 concentrations using traffic and meteorological condition data. The location of the study site is at a road intersection in Melton Mowbray, which is a town in Leicestershire, U.K. Several NNs, which can be classified into three types, namely, the multilayer perceptron, the radial basis function, and the modular network, were developed to model the nonlinear relationships that exist in the pollutant concentrations. Their performances are analyzed and compared. The transferability of the developed models is studied using data collected from a road intersection in another city. It was concluded that all NNs provide reliable estimates of pollutant concentrations using limited information and noisy data. © 2008 IEEE.

Publication metadata

Author(s): Zito P, Chen H, Bell MC

Publication type: Article

Publication status: Published

Journal: IEEE Transactions on Intelligent Transportation Systems

Year: 2008

Volume: 9

Issue: 3

Pages: 514-522

Print publication date: 01/09/2008

ISSN (print): 1524-9050

ISSN (electronic): 1558-0016

Publisher: IEEE

URL: http://dx.doi.org/10.1109/TITS.2008.928259

DOI: 10.1109/TITS.2008.928259

Altmetrics

Share