Correlation of air pollution and meteorological data using neural networks

In order to develop an environmental forecasting tool, the Neural Network method of computational intelligence is investigated. For this purpose, hourly and daily time series of CO, NO2 and O3, as well as a variety of meteorological variables are employed in various multi-layer percepton (MLP) models, in order to provide reliable air quality forecasts, using as a test case the city of Athens, Greece. The performance of the two most satisfactory models are presented thoroughly and compared using certain statistical indices. Results verify both the potential and the complicated nature of the method.

[1]  Sandy D. Balkin.  A statistical implementation of feedforward neural networks for univariate time series forecasting , 2000 .

[2]  Christos Zerefos,et al.  Forecasting peak pollutant levels from meteorological variables , 1995 .

[3]  D. Melas,et al.  Neural Network Model for Predicting Peak Photochemical Pollutant Levels , 2000, Journal of the Air & Waste Management Association.

[4]  C. Willmott.  Some Comments on the Evaluation of Model Performance , 1982 .

[5]  Patricio Perez,et al.  Prediction of NO and NO2 concentrations near a street with heavy traffic in Santiago, Chile , 2001 .

[6]  M. C. Hubbard,et al.  A Comparison of Nonlinear Regression and Neural Network Models for Ground-Level Ozone Forecasting , 2000, Journal of the Air & Waste Management Association.

[7]  I. Zurbenko,et al.  Moderating the Influence of Meteorological Conditions on Ambient Ozone Concentrations. , 1996, Journal of the Air & Waste Management Association.

[8]  Robert E. Davis,et al.  Statistics for the evaluation and comparison of models , 1985 .

[9]  Roy M. Harrison,et al.  Regression modelling of hourly NOx and NO2 concentrations in urban air in London , 1997 .

[10]  A. Comrie.  Comparing Neural Networks and Regression Models for Ozone Forecasting , 1997 .

[11]  N Moussiopoulos,et al.  Statistical analysis of environmental data as the basis of forecasting: an air quality application. , 2002, The Science of the total environment.

[12]  Scott M. Robeson,et al.  Evaluation and comparison of statistical forecast models for daily maximum ozone concentrations , 1990 .

[13]  William F. Ryan,et al.  Forecasting severe ozone episodes in the Baltimore metropolitan area , 1995 .

[14]  M. W Gardner,et al.  Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences , 1998 .

[15]  R. Lippmann,et al.  An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[16]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[17]  D G Gajghate,et al.  Prediction of Ambient PM10 and Toxic Metals Using Artificial Neural Networks , 2002, Journal of the Air & Waste Management Association.

[18]  M. Kolehmainen,et al.  Neural networks and periodic components used in air quality forecasting , 2001 .

[19]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[20]  Trevor D. Davies,et al.  Regression and stochastic models for air pollution—I. Review, comments and suggestions , 1994 .