Browse > Article
http://dx.doi.org/10.5572/ajae.2016.10.2.067

Predicting PM2.5 Concentrations Using Artificial Neural Networks and Markov Chain, a Case Study Karaj City  

Asadollahfardi, Gholamreza (Civil Engineering Department, Kharazmi University)
Zangooei, Hossein (Civil Engineering Department, Kharazmi University)
Aria, Shiva Homayoun (Civil Engineering Department, Kharazmi University)
Publication Information
Asian Journal of Atmospheric Environment / v.10, no.2, 2016 , pp. 67-79 More about this Journal
Abstract
The forecasting of air pollution is an important and popular topic in environmental engineering. Due to health impacts caused by unacceptable particulate matter (PM) levels, it has become one of the greatest concerns in metropolitan cities like Karaj City in Iran. In this study, the concentration of $PM_{2.5}$ was predicted by applying a multilayer percepteron (MLP) neural network, a radial basis function (RBF) neural network and a Markov chain model. Two months of hourly data including temperature, NO, $NO_2$, $NO_x$, CO, $SO_2$ and $PM_{10}$ were used as inputs to the artificial neural networks. From 1,488 data, 1,300 of data was used to train the models and the rest of the data were applied to test the models. The results of using artificial neural networks indicated that the models performed well in predicting $PM_{2.5}$ concentrations. The application of a Markov chain described the probable occurrences of unhealthy hours. The MLP neural network with two hidden layers including 19 neurons in the first layer and 16 neurons in the second layer provided the best results. The coefficient of determination ($R^2$), Index of Agreement (IA) and Efficiency (E) between the observed and the predicted data using an MLP neural network were 0.92, 0.93 and 0.981, respectively. In the MLP neural network, the MBE was 0.0546 which indicates the adequacy of the model. In the RBF neural network, increasing the number of neurons to 1,488 caused the RMSE to decline from 7.88 to 0.00 and caused $R^2$ to reach 0.93. In the Markov chain model the absolute error was 0.014 which indicated an acceptable accuracy and precision. We concluded the probability of occurrence state duration and transition of $PM_{2.5}$ pollution is predictable using a Markov chain method.
Keywords
Air pollution; $PM_{2.5}$ concentration prediction; Artificial neural network; Markov chain;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Bahari, R.A., Ali Abssaspour, R., Pahlavi, P. (2014) Prediction of $PM_{2.5}$ concentrations using temperature inversion effects based on an artificial neural network, The ISPRS international conference of Geospatial information research, 15-17 November, Tehran, Iran.
2 Caputo, M., Gimenez, M., Schlamp, M. (2003) Intercomparison of atmospheric dispersion models. Atmospheric Environment 37, 2435-2449.   DOI
3 Chung, K.L., Farid AitSahlia (2003) Elementary Probability Theory: With Stochastic Processes and an Introduction to Mathematical Finance, Springer Undergraduate Texts in Mathematics and Technology, ISSN 0172-6056.
4 Cohen, S., Intrator, N. (2002) Automatic model selection in a hybrid perceptron/radial network; Information Fusion. Special Issue on Multiple Experts 3(4), 259-266.
5 Deng, X., Zhang, F., Rui, W., long, F., Wang, L., Feng, Z., Chen, D., Ding, W. (2013) $PM_{2.5}$-induced oxidative stress triggers autophagy in human lung epithelial A549 cells. Toxicology in Vitro 27(6), 1762-1770.   DOI
6 Dong, G.H., Zhang, P., Sun, B., Zhang, L., Chen, X., Ma, N. (2012) Long term exposure to ambient air pollution and respiratory disease mortality in Shenyang, China: a 12 year population - based retrospective cohort study. Respiration 84(5), 360-368.   DOI
7 Eleuteri, A., Tagliaferri, R., Milano, L. (2005) A novel information geometric approach to variable selection in MLP networks. Neural Network 18(10), 1309-1318.   DOI
8 Feng, X., Li, Q., Zhu, Y., Hou, J., Jin, L., Wang, J. (2015) Artificial neural network forecasting of $PM_{2.5}$ pollution using air mass trajectory based geographic model and wavelet transformation. Atmospheric Environment 107, 118-128.   DOI
9 Goss, C.H., Newsom, S.A., Schildcrout, J.S., Sheppard, L., Kaufman, J.D. (2004) Effect of ambient air pollution on pulmonary exacerbations and lung function in cystic fibrosis. American Journal of Respiratory and Critical Care Medicine 169(7), 816-821.   DOI
10 Hambli, R. (2011) Multiscale prediction of crack density and crack length accumulation in trabecular bone based on neural networks and finite element simulation. International Journal for Numerical Methods in Biomedical Engineering 27(4), 461-475.   DOI
11 Hanna, S.R., Paine, R., Heinold, D., Kintigh, E., Baker, D. (2007) Uncertainties in air toxics calculated by the dispersion models AERMOD and ISCST 3 in the Houston ship channel area. Journal of Applied Meteorology and Climatology 46, 1372-1382.   DOI
12 Kohavi, R., John, G.H. (1997) Wrappers for feature subset selection. Artificial Intelligence 97, 273-324.   DOI
13 Harsham, D.K., Bennett, M. (2008) A sensitivity study of validation of three regulatory dispersion models. American Journal of Environmental Sciences 4(1), 63-76.   DOI
14 Haykin, S. (1999) Neural networks: a comprehensive foundation. (2nd ed.) Upper Saddle River, New Jersey: Prentice Hal.
15 Jones, R.M., Nicas, M. (2014) Benchmarking of a Markov multizone model of contaminant transport. Annals of Occupational Hygiene 58(8), 1018-1031.   DOI
16 Kohohen, T. (1984) Self-organization and associative memory. New York: Springer-Verlag.
17 Krause, P., Boyle, D.P., Base, F. (2005) Comparison of different efficiency criteria for hydrological model assessment. Advances in Geosciences 5, 89-97.   DOI
18 Kukkonen, J., Partanen, L., Karppinen, A., Ruuskanen, J., Junninen, H., Kolehmainen, M., Li, P., Xin, J.Y., Wang, Y.S., Wang, S.G., Li, G.X., Pan, X.C., Liu, Z.R., Wang, L.L. (2015) Reinstate regional transport of $PM_{2.5}$ as a major cause of severe haze in Beijing. Proceeding of the National Academy of Sciences of the United States of America 112, E2739-E2740.   DOI
19 Kuncheva, L. (2004) Combining Pattern Classifiers: Methods and Algorithms. Wiley, New York, USA.
20 Kurt, A., Gulbagci, B., Karaca, F., Alagha, O. (2008) An online air pollution forecasting system using neural networks. Environment International 34, 592-598.   DOI
21 Logofet, D.O., Lensnaya, E.V. (2000) The mathematics of Markov models: what Markov chains can really predict in forest successions. Ecological Modelling 2(3), 285-298.
22 Niska, H., Rantamaki, M., Hiltunen, T., Karppinen, A., Kukkonen, J., Ruuskanen, J. (2005) Evaluation of an integrated modelling system containing a multi-layer perceptron model and the numerical weather prediction model HIRLAM for the forecasting of urban airborne pollutant concentrations. Atmospheric Environment 39(35), 6524-6536.   DOI
23 Nicas, M. (2014) Markov modeling of contaminant concentrations in indoor air. American Journal of Environmental Sciences, 61(4), 484-491.
24 Niska, H., Dorling, S., Chatterton, T., Foxall, R., Cawley, G. (2003) Extensive evaluation of neural network models for the prediction of $NO_2$ and $PM_{10}$ concentrations, compared with a deterministic modeling system and measurements in central Helsinki. Atmospheric Environment 37, 4539-4550.   DOI
25 Niska, H., Heikkinen, M., Kolehmainen, M. (2006) Genetic algorithms and sensitivity analysis applied to select inputs of a multi-layer perceptron for the prediction of air pollutant time-series. Chapter Intelligent data engineering and automated learning-IDEAL2006 volume 4224 of the series lecture notes in computer science pp. 224-231 springer publisher.
26 Orr, M.J.L. (1996) Introduction to radial basis function networks, University of Edinbergh, EH89LW.
27 Owega, S., Khan, B.U.Z., Evans, G.J., Jervis, R.E., Fila, M. (2006) Identification of long-range aerosol transport patterns to Toronto via classification of back trajectories by cluster analysis and neural network techniques. Chemo Metrics and Intelligent Laboratory Systems 83(1), 26-33.   DOI
28 Romanof, N. (1982), A Markov chain model for the mean daily $SO_2$ concentrations. Atmospheric Environment 16(8), 1895-1897.   DOI
29 Shamshad, A., Bawadi, M.A., Wan Hussin, W.M.A., Majid, T.A., Sanusi, S.A.M. (2005) First and second order Markov chain models for synthetic generation of wind speed time series. Energy 30, 693-708.   DOI
30 Rumelhart, D.E., McClelland, J.L. (1986) Parallel distribution processing: Exploration in the microstructure of cognition, Cambridge, MA: MIT Press.
31 Slaughter, J.C., Lumley, T., Sheppard, L., Koenig, J.Q., Shapiro, G.G. (2003) Effects of ambient air pollution on symptom severity and medication use in children with asthma. Annals of Allergy, Asthma and Immunology 91(4), 346-353.   DOI
32 Slini, T., Kaprara, A., Karatzas, K., Moussiopoulos, N. (2006) $PM_{10}$ forecasting for Thessaloniki, Greece. Environ. Modell. Softw. 21, 559-565.   DOI
33 Song, X.M. (1996) Radial basis function networks for empirical modeling of chemical process. MSc thesis, University of Helsinki.
34 Sun, W., Zhang, H., Palazoglu, A., Singh, A., Zhang, W., Liu, S. (2013) Prediction of 24-hour-average $PM_{2.5}$ concentrations using a hidden Markov model with different emission distributions in Northern California. Science of the Total Environment 443, 93-103.   DOI
35 Taylor, H., Karlin, S. (1998) An Introduction to Stochastic Modeling. Academic Press, San Diego, California.
36 Voukantsis, D., Karatzas, K., Kukkonen, J., Rasanen, T., Karppinen, A., Kolehmainen, M. (2011) Intercomparison of air quality data using principal component analysis, and forecasting of $PM_{10}$ and $PM_{2.5}$ concentrations using artificial neural networks, in Thessaloniki and Helsinki. Science of the Total Environment 409, 1266-1276.   DOI
37 Zurada, J.M. (1992) Introduction to Artificial Neural Systems, PWS; Singapore, 195-196.
38 Wang, X., Liu, W. (2012) Research on Air Traffic Control Automatic System Software Reliability Based on Markov Chain. Physics Procedia 24, 1601-1606.   DOI
39 Wilks, D.S. (2006) Statistical methods in the atmospheric sciences. 2nd ed. Academic Press, xvii, 627 p.
40 Zickus, M., Greig, A.J., Niranjan, M. (2002) Comparison of four machine learning methods for predicting $PM_{10}$ concentration in Helsinki, Finland. Water, Air and Soil Pollution 2(5), 717-729.   DOI