Hourly ozone prediction for a 24-h horizon using neural networks

This study is an attempt to verify the presence of non-linear dynamics in the ozone time series by testing a ''dynamic'' model, evaluated versus a ''static'' one, in the context of predicting hourly ozone concentrations, one-day ahead. The ''dynamic'' model uses a recursive structure involving a cascade of 24 multilayer perceptrons (MLP) arranged so that each MLP feeds the next one. The ''static'' model is a classical single MLP with 24 outputs. For both models, the inputs consist of ozone and of exogenous variables: past 24-h values of meteorological parameters and of NO"2; concerning the ozone inputs, the ''static'' model uses only the 24 past measurements, while the ''dynamic'' one uses, also, the previously forecast ozone concentrations, as soon as they are predicted by the model. The outputs are, for both configurations, ozone concentrations for a 24-h horizon. The performance of the two models was evaluated for an urban and a rural site, in the greater Paris. Globally, the results indicate a rather good applicability of these models for a short-term prediction of ozone. We notice that the results of the recursive model were comparable with those obtained via the ''static'' one; thus, we can conclude that there is no evidence of non-linear dynamics in the ozone time series under study.

[1]  Gavin C. Cawley,et al.  Modelling SO2 concentration at a point with statistical approaches , 2004, Environ. Model. Softw..

[2]  C. R. Reeves,et al.  Artifical Neural Nets and Genetic Algorithms: Proceedings of the International Conference in Innsbruck, Austria, 1993 , 1993 .

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

[4]  Yoshua Bengio,et al.  Pattern Recognition and Neural Networks , 1995 .

[5]  Gavin C. Cawley,et al.  Error Functions for Prediction of Episodes of Poor Air Quality , 2002, ICANN.

[6]  Stephen Dorling,et al.  Statistical surface ozone models: an improved methodology to account for non-linear behaviour , 2000 .

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

[8]  S. Vitabile,et al.  Two-days ahead prediction of daily maximum concentrations of SO2, O3, PM10, NO2, CO in the urban area of Palermo, Italy , 2007 .

[9]  Milan Paluš,et al.  Nonlinearity and Prediction of Air Pollution , 2001 .

[10]  S. I. V. Sousa,et al.  Multiple linear regression and artificial neural networks based on principal components to predict ozone concentrations , 2007, Environ. Model. Softw..

[11]  J. Lodge.  Air quality guidelines for Europe: WHO regional publications, European series, No. 23, World Health Organization, 1211 Geneva 27, Switzerland; WHO publications center USA, 49 Sheridan Avenue, Albany, NY 12210, 1987, xiii + 426 pp. price: Sw. fr. 60 , 1988 .

[12]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[13]  Martin Fodslette Møller,et al.  A scaled conjugate gradient algorithm for fast supervised learning , 1993, Neural Networks.

[14]  Oms. Regional Office for Europe.  Air quality guidelines for Europe , 1996, Environmental science and pollution research international.

[15]  T. S. Dye,et al.  Guideline for developing an ozone forecasting program , 1999 .

[16]  Holger R. Maier,et al.  Neural networks for the prediction and forecasting of water resource variables: a review of modelling issues and applications , 2000, Environ. Model. Softw..

[17]  Gavin C. Cawley,et al.  A rigorous inter-comparison of ground-level ozone predictions , 2003 .

[18]  J. L. Carrasco-Rodriguez,et al.  Effective 1-day ahead prediction of hourly surface ozone concentrations in eastern Spain using linear models and neural networks , 2002 .

[19]  Gabriel Ibarra-Berastegi,et al.  Regression and multilayer perceptron-based models to forecast hourly O3 and NO2 levels in the Bilbao area , 2006, Environ. Model. Softw..

[20]  Robert Vautard,et al.  Validation of a hybrid forecasting system for the ozone concentrations over the Paris area , 2001 .

[21]  P. Viotti,et al.  Atmospheric urban pollution: applications of an artificial neural network (ANN) to the city of Perugia , 2002 .

[22]  Gavin C. Cawley,et al.  Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations, compared with a deterministic modelling system and measurements in central Helsinki , 2003 .

[23]  Gavin C. Cawley,et al.  Statistical models to assess the health effects and to forecast ground-level ozone , 2006, Environ. Model. Softw..

[24]  David F. Shanno,et al.  Conjugate Gradient Methods with Inexact Searches , 1978, Math. Oper. Res..

[25]  Saleh M. Al-Alawi,et al.  Assessment and prediction of tropospheric ozone concentration levels using artificial neural networks , 2002, Environ. Model. Softw..

[26]  W. Dab,et al.  Air pollution and doctors' house calls: Results from the ERPURS system for monitoring the effects of air pollution on public health in Greater Paris, France, 1991-1995 , 1997 .

[27]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .

[28]  Uwe Schlink,et al.  Non-parametric Short-Term Prediction of Ozone Concentration in Berlin: Preconditions and Justification , 2002 .

[29]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[30]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[31]  Yves Candau,et al.  Air pollutant emissions prediction by process modelling - Application in the iron and steel industry in the case of a re-heating furnace , 2007, Environ. Model. Softw..