Title: ANN-based PM prediction model for assessing the temporal variability of PM10, PM2.5 and PM1 concentrations at an urban roadway
Authors: B. Srimuruganandam; S.M. Shiva Nagendra
Addresses: Environmental Engineering Division, School of Mechanical and Building Sciences, VIT-University, Vellore-632014, India ' Environmental and Water Resources Engineering Division, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai-600036, India
Abstract: In this paper, hourly particulate matter (PM10, PM2.5 and PM1) concentrations, traffic and meteorological parameters monitored at urban roadway in India during November 2007-May 2009 are used to develop 1-hr and 24-hr average artificial neural network (ANN)-based PM prediction models. Maximum of eight meteorological and five traffic characteristic variables have been used in the models' formulation. Three scenarios were considered - considering both meteorological and traffic characteristics as input parameters; only meteorological inputs; and only traffic characteristics input data. The performance of all the developed models are evaluated on the basis of index of agreement (IA) and other statistical parameters, viz. the mean and the deviations of the observed and predicted concentrations, mean bias error, mean square error, systematic and unsystematic root mean square error, coefficient of determination and linear best fit constant and gradient. The performance of 24-hr average ANN-based PM model (ANNDMT) showed good performance (IA > 0.85 i.e. more than 85% of the predictions are error free) on the validation dataset. While, the 1-hr average ANN-based PM model (ANNHMT) showed reasonable performance (IA < 0.75) on the validation dataset. This study demonstrates that ANN is capable of modelling complex PM dispersion phenomenon at urban roads having heterogeneous traffic.
Keywords: particulate matter; heterogeneous traffic; artificial neural networks; ANNs; back-propagation; PM prediction modelling; meteorology; urban roads; PM dispersion; pollutant dispersion modelling; air pollution; vehicle emissions.
International Journal of Environmental Engineering, 2015 Vol.7 No.1, pp.60 - 89
Available online: 06 May 2015 *Full-text access for editors Access for subscribers Purchase this article Comment on this article