The Neural Network Modeling of Suspended Particulate Matter with Autoregressive Structure

Mehmet AKTAN, Hanefi BAYRAKTAR

Ekoloji, 2010, Issue 74, Pages: 32-37

OPEN ACCESS

Download Full Text (PDF)

Abstract

Pollution sources and emissions, and their interactions with terrain and the atmosphere are necessary in developing appropriate air pollution management plans and action strategies. In this study, we will investigate the relationship between the total suspended particulate matter (TSP) concentration and meteorological parameters such as wind speed and direction, temperature, air pressure, precipitation and relative humidity. The TSP measurements o f the past two days, sunshine duration, and sunshine amount in the winter season (November through March) in the city o f Erzurum between the years 1990 and 2007 were investigated. The artificial neural network (ANN) models were constructed using a mixed autoregressive relationship to realize the stochastic nature o f the TSP levels for each month o f the winter season. The impact o f wind direction on TSP concentration was introduced to the model by defining two variables. Linear regression models were also constructed to check the performance o f the neural networks. The most significant factors affecting the TSP concentration are found to be the TSP level o f the previous day (TSP(_i), expected temperature (temp), wind speed (w), air pressure (p), and precipitation (pc).

Keywords

Autoregressive, Erzurum, meteorological parameters, neural network, particulate matter

References

  • Aktan M (2008) Artificial neural network modeling o f winter season sulphur dioxide concentrations in Erzurum city. Fresenius Environmental Bulletin 17, 2, 218-223.
  • Bayraktar H, Turalioglu FS (2005) A Kriging-based approach for locating a sampling site-in the assessment of air quality. Stochastic Environmental Research Risk Assessment 19, 4, 301-305.
  • Berkowicz R (2000) A simple model for urban background pollution. Environmental Monitoring and Assessment 65, 259-267.
  • Berkowicz R (2000) OSPM - a parameterised street pollution model. Environmental Monitoring and Assessment 65, 323-331.
  • Bertsekas DP (1995) Nonlinear Programming. Athena Scientific, Belmont.
  • Bridgman HA, Davies TD, Jickells T, Hunova I, Tovey K, Bridges K, Surapipith V (2002) Air Pollution in the Krusne Hory region, Czech Republic during the 1990s. Atmospheric Environment 36, 3375-3389.
  • Chao Z (1990) Urban climate and air pollution in Shangai. Energy Build 15, 647-656.
  • Quhadaroglu B, Demirci E (1997) Influence o f some meteorological factors on air pollution in Trabzon city. Energy Build 25, 179-184.
  • Elbir T, Muezzinoglu A, Bayram A (2000) Evaluation o f some air pollution indicators in Turkey. Environment International 26, 5-10.
  • Haykin S (1994) Neural networks: a comprehensive foundation. Macmillan, New York.
  • Henry RC, Chang YS, Spiegelman CH (2002) Locating nearby sources o f air pollution by nonparametric regression o f atmospheric concentrations on wind direction. Atmospheric Environment 36, 2237-2244.
  • Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Networks 2, 359-366.
  • Hyslop N P (2009) Impaired visibility: the air pollution people see. Atmospheric Environment 43, 182-195.
  • Khare M, Nagendra SSM (2007) Artificial Neural Networks in Vehicular Pollution Modelling. Springer-Verlag, Berlin, Heidelberg.
  • Kurt A, Gulbagci B, Karaca F, Alagha O (2008) An online air pollution forecasting system using neural networks. Environment International 34, 592-598.
  • Liu J, Mauzerall DL, Horowitz LW (2009) Evaluating Inter-Continental Transport of Fine Aerosols: (2) Global Health Impact. Atmospheric Environment DOI: 10.1016/j.atmosenv.2009.05.032.
  • Montgomery DC, Runger GC (2003) Applied Statistics and Probability for Engineers. John Wiley and Sons Inc., New York.
  • Ripley BD (1996) Pattern recognition and neural networks. Cambridge University Press, Cambridge.
  • Topçu N, Keskinler B, Bayramoğlu M, Akçay M (1993) Air pollution modeling in Erzurum city. Envimental Pollution 79, 9-13.
  • Taşdemir Y, Çağlar Ö (2002) Uludag Üniversitesi Kampüsü'nde Ölçülen Partikül Madde Kuru Çökelme Akılan. Ekoloji 11, 42, 8-11.
  • Yıldırım Y, Bayramoğlu M (2006) Adaptive neuro-fuzzy based modelling for prediction of air pollution daily levels in city o f Zonguldak. Chemosphere 63, 1575-1582.