Home // ADVCOMP 2010, The Fourth International Conference on Advanced Engineering Computing and Applications in Sciences // View article
Authors:
Mohamed Tarek Khadir
Keywords: Pollutants concentration prediction, Artificial Neural Network, Elman Network, Radial Basis Function, neurocomputing.
Abstract:
This paper describes the development of air pollutants concentration prediction models of five different pollutants (O3, PM10, SO2, NOx, COx), using Radial Basis Function, and Elman Networks, two neurocomputing paradigms. Each Artificial Neural network (ANN) predicts, therefore the concentration of the five different pollutants. These models are developed in order to give 12 hours ahead prediction for the region of Annaba, northeast of Algeria (north of Africa). Receiving the measurement of air pollutant concentration and the metrological parameters (wind speed, temperature and humidity) at time t; the models are designed to predict air pollutant concentration at t+12 hours. Once predicted pollutant concentrations are obtained, and the validity of each ANN model is proven, the performances of both ANN models are comprehensively compared and assessed. Conclusions are finally drawn and the use of a particular ANN network over another is justified on the light of the obtained results.
Pages: 28 to 33
Copyright: Copyright (c) IARIA, 2010
Publication date: October 25, 2010
Published in: conference
ISSN: 2308-4499
ISBN: 978-1-61208-101-4
Location: Florence, Italy
Dates: from October 25, 2010 to October 30, 2010