Prediction of Air Pollutant Levels by Using Artificial Neural Networks and Statistical Methods

One of the environmental problems adversely affecting human health and welfare in many part of world is air pollution. Pollution monitoring data can be utilized to predict concentrations of air pollutants for short-term using artificial intelligence approaches and multivariate regression analysis. In this paper, artificial neural networks (ANNs) and multivariate regression modeling (MRM) techniques have been comparatively employed to forecast one-hour ahead concentration of particulate air pollution (PM10). An hourly based data was composed by including meteorological factors and particulate concentrations for the years 2015-2016. ANN(12-7-1) model with R2 of 0.887 and RMSE of 19.89 yielded fairly rational predictions over hourly dataset while the best MRM models produced lower scores with R2 of 0.848 and RMSE of 21.33 in general. ANN models simulating the time series data better among the models identified have been chosen in further tuning in the prediction of next hour’s concentration of PM10.

[1]  Salim Lahmiri,et al.  A Comparative Study Of Backpropagation Algorithms In Financial Prediction , 2011 .

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

[3]  Wei Huang,et al.  Systematic review of Chinese studies of short-term exposure to air pollution and daily mortality. , 2013, Environment international.

[4]  Runhe Shi,et al.  Ensemble and enhanced PM10 concentration forecast model based on stepwise regression and wavelet analysis , 2013 .

[5]  Reza Modarres,et al.  Daily air pollution time series analysis of Isfahan City , 2005 .

[6]  Jorge Reyes,et al.  Prediction of PM2.5 concentrations several hours in advance using neural networks in Santiago, Chile , 2000 .

[7]  Le Jian,et al.  An application of ARIMA model to predict submicron particle concentrations from meteorological factors at a busy roadside in Hangzhou, China. , 2012, The Science of the total environment.

[8]  Q Tong,et al.  Measurements and analysis of criteria pollutants in New Delhi, India. , 2001, Environment international.

[9]  Mikko Kolehmainen,et al.  Evolving the neural network model for forecasting air pollution time series , 2004, Eng. Appl. Artif. Intell..

[10]  Who Europe.  Air Quality Guidelines Global Update 2005: Particulate Matter, ozone, nitrogen dioxide and sulfur dioxide , 2006 .

[11]  Ding-Xuan Zhou,et al.  Learning gradients by a gradient descent algorithm , 2008 .

[12]  P. Goyal,et al.  Statistical models for the prediction of respirable suspended particulate matter in urban cities , 2006 .

[13]  S S Huang,et al.  Forecasts Using Neural Network versus Box-Jenkins Methodology for Ambient Air Quality Monitoring Data , 2000, Journal of the Air & Waste Management Association.

[14]  Ferhat Karaca,et al.  An online air pollution forecasting system using neural networks. , 2008, Environment international.

[15]  D. Dockery,et al.  Health Effects of Fine Particulate Air Pollution: Lines that Connect , 2006, Journal of the Air & Waste Management Association.

[16]  Joel L. Horowitz,et al.  Statistical analysis of the maximum concentration of an air pollutant: Effects of autocorrelation and non-stationarity , 1979 .

[17]  Kirk R. Smith,et al.  Global review of national ambient air quality standards for PM10 and SO2 (24 h) , 2011, Air Quality, Atmosphere & Health.

[18]  Athanasios Sfetsos,et al.  A new methodology development for the regulatory forecasting of PM10. Application in the Greater Athens Area, Greece , 2010 .

[19]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[20]  Fatih Taşpınar,et al.  Improving artificial neural network model predictions of daily average PM10 concentrations by applying principle component analysis and implementing seasonal models , 2015, Journal of the Air & Waste Management Association.

[21]  Ahmad Zia Ul-Saufie,et al.  Future daily PM10 concentrations prediction by combining regression models and feedforward backpropagation models with principle component analysis (PCA) , 2013 .

[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]  Anikender Kumar,et al.  Forecasting of daily air quality index in Delhi. , 2011, The Science of the total environment.

[24]  L. Folinsbee.  Human health effects of air pollution. , 1993, Environmental health perspectives.