ตัวแบบผสมอารีมา-ซัพพอร์ตเวกเตอร์แมชชีนเน้นองค์ประกอบเชิงบวกสำหรับการพยากรณ์ปริมาณฝุ่นละอองขนาดไม่เกิน 10 ไมครอน
Article Sidebar
Main Article Content
Abstract
Abstract
In this research, a hybrid model of ARIMA and SVM is developed and proposed for PM10 forecasting in Na Phra Lan sub-district, Chaloem Phra Kiat district, Saraburi province based on an assumption of additive components. In this regard, the ARIMA model, which has dominated in linear forecasting problems, is used to predict linear components of the PM10 time series. Whilst, the SVM model is a well-known approach with superior performance of nonlinear forecasting model, which is used to formulate nonlinear and complex function in order to predict nonlinear components of the PM10 time series. In addition, the developed model is compared to both ARIMA model and SVM model based on six accuracy measures. The empirical results revealed that the SVM model based on radial kernel function slightly outperforms ARIMA model. In contrast, the developed hybrid model can provide lower errors than both ARIMA and SVM models. The comparison between ARIMA model and SVM model indicated that the PM10 time series may not be explicit linear or nonlinear patterns. Subsequently, the developed hybrid model can reduce a risk of using improper forecasting model. Consequently, the hybrid ARIMA-SVM model can be a promising tool for PM10 forecasting in other area in order to provide useful information to monitor and make a critical decision on air quality that is not only a very challenging task but also a public concerned problem for human health.
Keywords: combined model; air pollution forecasting; ARIMA; support vector machine; particulate matter
Article Details
References
[2] สถานการณ์มลพิษประเทศไทย ปี 2558 รอบ 6 เดือน, กรมควบคุมมลพิษ กระทรวงทรัพยากรธรรมชาติและสิ่งแวดล้อม, แหล่งที่มา : http://www.oic.go.th/FILEWEB/CABINFOCENTER3/DRAWER056/GENERAL/DATA0000/00000425.PDF, 1 ตุลาคม 2560.
[3] Güler, N. and İşçi, Ö.G., 2016, The regional prediction model of PM 10 concentrations for Turkey, Atmospheric Res. 180: 64-77.
[4] Song, Y., Qin, S., Qu, J. and Liu, F., 2015, The forecasting research of early warning systems for atmospheric pollutants: A case in Yangtze River Delta region, Atmospheric Environ. 118: 58-69.
[5] Yang, Z. and Wang, J., 2017, A new air quality monitoring and early warning system: Air quality assessment and air pollutant concentration prediction, Environ. Res. 158: 105-117.
[6] Xu, Y., Yang, W. and Wang, J., 2017, Air quality early-warning system for cities in China, Atmospheric Environ. 148: 239-257.
[7] Zafra, C., Ángel, Y. and Torres, E., 2017, ARIMA analysis of the effect of land surface coverage on PM10 concentrations in a high-altitude megacity, Atmospheric Pollut. Res. 8: 660-668.
[8] Jian, L., Zhao, Y., Zhu, Y.P., Zhang, M.B. and Bertolatti, D., 2012, An application of ARIMA model to predict submicron particle concentrations from meteorological factors at a busy roadside in Hangzhou, China, Sci. Total Environ. 426: 336-345.
[9] Dumitrache, R.C., Iriza, A., Maco, B.A., Barbu, C.D., Hirtl, M., Mantovani, S., Nicola, O., Irimescu, A., Craciunescu, V., Ristea, A. and Diamandi, A., 2016, Study on the influence of ground and satellite observations on the numerical air-quality for PM10 over Romanian territory, Atmospheric Environ. 143: 278-289.
[10] Nieto, P.G., Combarro, E.F., del Coz Díaz, J.J. and Montañés, E., 2013, A SVM-based regression model to study the air quality at local scale in Oviedo urban area (Northern Spain): A case study, Appl. Math. Comput. 219: 8923-8937.
[11] Sánchez, A.S., Nieto, P.G., Fernández, P.R., del Coz Díaz, J.J. and Iglesias-Rodríguez, F.J., 2011, Application of an SVM-based regression model to the air quality study at local scale in the Avilés urban area (Spain), Math. Comput. Model. 54: 1453-
1466.
[12] Hirtl, M., Mantovani, S., Krüger, B.C., Triebnig, G., Flandorfer, C., Bottoni, M. and Cavicchi, M., 2014, Improvement of air quality forecasts with satellite and ground based particulate matter observations, Atmospheric Environ. 84: 20-27.
[13] Wang, P., Zhang, H., Qin, Z. and Zhang, G., 2017, A novel hybrid-Garch model based on ARIMA and SVM for PM2.5 concentrations forecasting, Atmospheric Pollut. Res. 8: 850-860.
[14] Khashei, M. and Bijari, M., 2011, A new hybrid methodology for nonlinear time series forecasting, Model. Simulat. Eng. 2011: 15.
[15] Khashei, M. and Bijari, M., 2011, A novel hybridization of artificial neural networks and ARIMA models for time series forecasting, Appl. Soft Comput. 11: 2664-2675.
[16] Zhu, B. and Wei, Y., 2013, Carbon price forecasting with a novel hybrid ARIMA and least squares support vector machines methodology, Omega 41: 517-524.
[17] Guo, Y., Wang, G., Zhang, X. and Deng, W., 2014, An improved hybrid ARIMA and support vector machine model for water quality prediction, pp. 411-422, International Conference on Rough Sets and Knowledge Technology, Springer, Cham.
[18] de Oliveira, J.F. and Ludermir, T.B., 2016, A hybrid evolutionary decomposition system for time series forecasting, Neurocomputing 180: 27-34.
[19] สรุปข้อมูลคุณภาพอากาศ พ.ศ. 2557-2559, สำนักจัดการคุณภาพอากาศและเสียง กรมควบคุมมลพิษ, แหล่งที่มา : http://aqnis.pcd. go.th, 1 ตุลาคม 2560.
[20] Khandakar, Y. and Hyndman, R.J., 2008, Automatic time series forecasting: the forecast Package for R, J. Stat. Software 27: 1-22.
[21] Sujjaviriyasup, T., 2017, Forecasting Thailand’s monthly export quantity of rubber compound using support vector machine model, Srinakharinwirot Sci. J. 33(1): 205-220. (In Thai)
[22] Muller, J. and Bogenberger, K., 2015, Time series analysis of booking data of a free-floating carsharing system in Berlin, Transp. Res. Proc. 10: 345-354.
[23] Kokic, P., Crimp, S. and Howden, M., 2014, A probabilistic analysis of human influence on recent record global mean temperature changes, Climate Risk Manage.
3: 1-12.
[24] Ghodhi, Z., Silva, S.E. and Hassani, H., 2015, Bicoid Signal Extraction with a Selection of Parametric and Nonpara metric Signal Processing Techniques, Genomics Proteomics Bioinformatics 13: 183-191.
[25] Haworth, J., Shawe-Taylor, J., Cheng, T. and Wang, J., 2014, Local online kernel ridge regression for forecasting of urban travel times, Transp. Res. C 46: 151-178.
[26] Götz, M., Richerzhagen, M., Bodenstein, C., Cavallaro, G., Glock, P., Riedel, M. and Benediktsson, J.A., 2015, On scalable data mining techniques for earth science, Proc. Comput. Sci. 51: 2188-2197.
[27] Appelhans, T., Mwangomo, E., Hardy, D.R., Hemp, A. and Nauss, T., 2015, Evaluating machine learning approaches for the interpolation of monthly air temperature at Mt.Kilimanjaro, Tanzania, Spatial Stat. 14: 91-113.