Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ

ÖZKAL, Can Burak; Arslan, Özkan

doi:10.35208/ert.1000739

Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ

Can Burak ÖZKAL, (Tekirdağ Namık Kemal Üniversitesi, Çevre Mühendisliği Bölümü, Tekirdağ, Türkiye)

Özkan ARSLAN (Tekirdağ Namık Kemal Üniversitesi, Elektronik ve Haberleşme Mühendisliği Bölümü, Tekirdağ, Türkiye)

Environmental Research & Technology

1 0

Yıl: 2022 Cilt: 5 Sayı: 1 Sayfa Aralığı: 56 - 71 Metin Dili: İngilizce DOI: 10.35208/ert.1000739 İndeks Tarihi: 09-08-2022

Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ

Öz:

Air pollution-induced issues involve public health, environmental, agricultural and socio-economic aspects. Therefore, decision-makers need low-cost, efficient tools with high spatiotemporal representation for monitoring air pollutants around urban areas and sensitive regions. Air pollution forecasting models with different time steps and forecast lengths are used as an alternative and support to traditional air quality monitoring stations (AQMS). In recent decades, given their eligibility to reconcile the relationship between parameters of complex systems, artificial neural networks have acquired the utmost importance in the field of air pollution forecasting. In this study, different machine learning regression methods are used to establish a mathematical relationship between air pollutants and meteorological factors from four AQMS (A-D) located between Çerkezköy and Süleymanpaşa, Tekirdağ. The model input variables included air pollutants and meteorological parameters. All developed models were used with the intent to provide instantaneous prediction of the air pollutant parameter NOx within the AQMS and across different stations. In the GMDH (group method of data handling)-type neural network method (namely the self-organizing deep learning approach), a five hidden layer structure consisting of a maximum of five neurons was preferred and, choice of layers and neurons were made in a way to minimize the error. In all models developed, the data were divided into a training (%80) and a testing set (%20). Based on R2, RMSE, and MAE values of all developed models, GMDH provided superior results regarding the NOx prediction within AQMS (reaching 0.94, 10.95, and 6.65, respectively for station A) and between different AQMS. The GMDH model yielded NOx prediction of station B by using station A input variables (without using NOx data as model input) with R2, RMSE and MAE values 0.80, 10.88, 7.31 respectively. The GMDH model is found suitable for being employed to fill in the gaps of air pollution records within and across-AQMS.

Anahtar Kelime: NOx prediction NOx self-organizing deep learning artificial intelligence air pollutant GMDH

Belge Türü: Makale Makale Türü: Araştırma Makalesi Erişim Türü: Erişime Açık

[1] Y. Dokuz, A. Bozdağ, and B. Gökçek, “Use of ma- chine learning methods for estimation and spatial distribution of air quality parameters,” Nigde Omer Halisdemir University Journal of Engineering Sci- ences, Vol. 9, pp. 37–47, 2020. (in Turkish)
[2] S.M. Cabaneros, J.K. Calautit, and B.R. Hughes, A re- view of artificial neural network models for ambient air pollution prediction, Environmental Modelling & Software, Vol. 119, pp. 285–304, 2019. [CrossRef]
[3] A. Yakın, and R. Behçet, “Effect of different types of fuels tested in a gasoline engine on engine per- formance and emissions,” International Journal of Hydrogen Energy, Vol. 46, pp. 33325–33338, 2021. [CrossRef]
[4] H. Zhang, Y. Wang, J. Hu, Q. Ying, and X.M. Hu, “Relationships between meteorological parameters and criteria air pollutants in three megacities in Chi- na,” Environment Reseach Vol. 140, pp. 242–254, 2015. [CrossRef]
[5] H.K. Elminir, “Dependence of urban air pollutants on meteorology,” Science of the Total Environment, Vol. 350, pp. 225–237, 2005. [CrossRef]
[6] E. Demirci, and B. Cuhadaroglu, “Statistical anal- ysis of wind circulation and air pollution in urban Trabzon,” Energy Building, Vol. 31, pp. 49–53, 2000. [CrossRef]
[7] S. Munir, M. Mayfield, D. Coca, S.A. Jubb, and O. Osammor, “Analysing the performance of low-cost air quality sensors, their drivers, relative benefits and calibration in cities—a case study in Sheffield,” Environmental Monitoring and Assessment, Vol. 191, Article 94, 2019. [CrossRef]
[8] N. Castell, F.R. Dauge, P. Schneider, M. Vogt, U. Le- rner, B. Fishbain, D. Broday, and A. Bartonova, “Can commercial low-cost sensor platforms contribute to air quality monitoring and exposure estimates?,” En- vironment International, Vol. 99 pp. 293–302, 2017. [CrossRef]
[9] M. Castelli, F.M. Clemente, A. Popovič, S. Silva, and L. Vanneschi, “A machine learning approach to pre- dict air quality in California,” Complexity, Vol. 2020, Article 8049504, 2020. [CrossRef]
[10] K. Prasad, and A.K. Gorai, P. “Goyal, Development of ANFIS models for air quality forecasting and in- put optimization for reducing the computational cost and time,” Atmospheric Environment, Vol. 128, pp. 246–262, 2016. [CrossRef]
[11] N. L iu, X. Liu, R. Jayaratne, and L. Morawska, “A study on extending the use of air quality monitor data via deep learning techniques,” Journal of Cleaner Production, Vol. 274, Article 122956, 2020. [CrossRef]
[12] R. Dongol, “Evaluation of the usability of low-cost sensors for public air quality information,” [Unpub- lished Master Thesis], Universitetet I Oslo Depart- ment of Informatics Programming and Networks, 2015.
[13] H. M aleki, A. Sorooshian, G. Goudarzi, Z. Baboli, Y. Tahmasebi Birgani, and M. Rahmati, “Air pollu- tion prediction by using an artificial neural network model,” Clean Technologies and Environmental Pol- icy, Vol. 21, pp. 1341–1352, 2019. [CrossRef]
[14] S.R. Shams, A. Jahani, S. Kalantary, M. Moeinaddi- ni, and N. Khorasani, “Artificial intelligence accura- cy assessment in NO2 concentration forecasting of metropolises air,” Scientific Reports, Vol. 11, pp. 1–9, 2021. [CrossRef]
[15] J. Xu, Y. Xu, H. Wang, C. Guo, H. Qiu, Y. He, Y. Zhang, X. Li, and W. Meng, “Occurrence of antibiot- ics and antibiotic resistance genes in a sewage treat- ment plant and its effluent-receiving river,” Chemo- sphere, Vol. 119, pp. 1379–1385, 2015. [CrossRef]
[16] U. Pak, J. Ma, U. Ryu, K. Ryom, U. Juhyok, K. Pak, and C. Pak, “Deep learning-based PM2.5 prediction considering the spatiotemporal correlations: A case study of Beijing, China,” Science of the Total Envi- ronment, Vol. 699, Article 133561, 2020. [CrossRef]
[17] H.K. Cigizoglu, K. Alp, and M. Kömürcü, “Estima- tion of air pollution parameters using artificial neu- ral networks,” Advances in Air Pollution Modeling for Environmental Security, pp. 63–75, 2005. [Cross- Ref]
[18] A. Alimissis, K. Philippopoulos, C.G. Tzanis, and D. Deligiorgi, “Spatial estimation of urban air pollution with the use of artificial neural network models”, Atmospheric Environment, Vol. 191, pp. 205–213, 2018. [CrossRef]
[19] F. Kunt, Z.C. Ayturan, and S. Dursun, “Used some modelling applications in air pollution estimates,” Journal of International Environment Applied Sci- ence Vol. 11, pp. 418–425, 2016.
[20] Y.A. Ayturan, Z.C. Ayturan, H.O. Altun, C. Kongoli, F.D. Tuncez, S. Dursun, and A. Ozturk, “Short-term prediction of pm2.5 pollution with deep learning methods,” Global Nest Journal, Vol. 22, pp. 126–131, 2020.
[21] M. Krishan, S. Jha, J. Das, A. Singh, M.K. Goyal, and C. Sekar, “Air quality modelling using long short- term memory (LSTM) over NCT-Delhi, India,” Air Quality, Atmosphere & Health Vol. 12, 899–908, 2019. [CrossRef]
[22] G. Varol, B. Tokuç, S. Ozkaya, and Ç. Çağlayan, “Air quality and preventable deaths in Tekirdağ, Turkey,” Air Quality, Atmosphere & Health, Vol. 14, pp. 843–853, 2021. [CrossRef]
[23] Thrace Development Agency, Thrace Region Plan (Rep.), 2019. Available at: https://www.trakya- ka.org.tr/upload/Node/33264/xfiles/trakya_bol- ge_%0Aplani_2014-2023.pdf. Accessed on Feb 2022, 06.
[24] A. Vardar, R. Okursoy, and Y. Tekin, “Local wind characteristics for east Thrace, Turkey ,” Energy Sources, Part B: Economics, Planning, and Policy, Vol. 7, pp. 1–9, 2012. [CrossRef]
[25] A. Liaw, and M. Wiener, “Classification and regres-sion by random forest,” R News, Vol. 2, pp. 18–22, 2002.
[26] F. Murtagh, “Multilayer perceptrons for classifica- tion and regression,” Neurocomputing, Vol. 2, 183– 197, 1991. [CrossRef]
[27] T. Kondo, “GMDH neural network algorithm using the heuristic self-organization method and its ap- plication to the pattern identification problem,” In Proceedings of the 37th SICE Annual Conference. International Session Papers, IEEE, pp. 1143–1148, 1998.
[28] S.-K. Oh, and W. Pedrycz, “The design of self-orga- nizing polynomial neural networks,” Information Sciences, Vol. 141, pp. 237–258, 2002. [CrossRef]
[29] S.J. Farlow, “Self-organizing methods in modeling: GMDH type algorithms,” CRC Press, Florida, 2020. [CrossRef]
[30] J. Ma, Y. Ding, J.C.P. Cheng, F. Jiang, and Z. Wan, “A t emporal-spatial interpola tion and extrapola- tion method based on g eographic Long Short-Term Memory neural network for PM2,” Journal of Cleaner Production, Vol. 237, Article 117729, 2019. [CrossRef]
[31] Ö. Akyürek, O. Arslan, ve A. Karademir, “SO2 Ve PM10 hava kirliliği parametrelerinin CBS ile ko- numsal analizi: Kocaeli örneği,” TMMOB Coğrafi Bilgi Sistemleri Kongresi, Cilt 12, 2013.
[32] M. Savic, I. Mihajlovic, and Z. Zivkovic, “An anfis– based air quality model for prediction of SO2 con- centration in urban area,” Serbian Journal of Man- agement, Vol. 8, pp. 25–38, 2013. [CrossRef]
[33] A.B. Chelani, C.V.C. Rao, K.M. Phadke, and M.Z. Hasan, “Prediction of sulphur dioxide concentration using artificial neural networks,” Environmental Mod- elling & Software, Vol. 17, pp. 159–166, 2002. [CrossRef]
[34] G. Asadollahfardi, H. Zangooei, and S.H. Aria, “Pre- dicting PM 2.5 concentrations using artificial neural networks and markov chain, a case study Karaj City,” Asian Journal of Atmospheric Environment, Vol. 10 pp. 67–79, 2016. [CrossRef]
[35] L. Rafati, M. Ehrampoush, A. Talebi, and M. Mokhtari, Z. Kheradpisheh, H. Dehghan, “Mod- elling the formation of Ozone in the air by using Adaptive Neuro-Fuzzy Inference System (ANFIS) (Case study: city of Yazd, Iran),” Desert, Vol. 19, pp. 131–135, 2014.
[36] S. Teixeira, P. Pereira, and F. Ferreira, “Co-creating data based on human nose perceptions to study odour nuisance from an oilseed industry,” 16th In- ternational Conference on Environmental Science and Technology, 2019.

APA	ÖZKAL C, Arslan Ö (2022). Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. , 56 - 71. 10.35208/ert.1000739
Chicago	ÖZKAL Can Burak,Arslan Özkan Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. (2022): 56 - 71. 10.35208/ert.1000739
MLA	ÖZKAL Can Burak,Arslan Özkan Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. , 2022, ss.56 - 71. 10.35208/ert.1000739
AMA	ÖZKAL C,Arslan Ö Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. . 2022; 56 - 71. 10.35208/ert.1000739
Vancouver	ÖZKAL C,Arslan Ö Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. . 2022; 56 - 71. 10.35208/ert.1000739
IEEE	ÖZKAL C,Arslan Ö "Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ." , ss.56 - 71, 2022. 10.35208/ert.1000739
ISNAD	ÖZKAL, Can Burak - Arslan, Özkan. "Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ". (2022), 56-71. https://doi.org/10.35208/ert.1000739

APA	ÖZKAL C, Arslan Ö (2022). Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. Environmental Research & Technology, 5(1), 56 - 71. 10.35208/ert.1000739
Chicago	ÖZKAL Can Burak,Arslan Özkan Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. Environmental Research & Technology 5, no.1 (2022): 56 - 71. 10.35208/ert.1000739
MLA	ÖZKAL Can Burak,Arslan Özkan Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. Environmental Research & Technology, vol.5, no.1, 2022, ss.56 - 71. 10.35208/ert.1000739
AMA	ÖZKAL C,Arslan Ö Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. Environmental Research & Technology. 2022; 5(1): 56 - 71. 10.35208/ert.1000739
Vancouver	ÖZKAL C,Arslan Ö Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ. Environmental Research & Technology. 2022; 5(1): 56 - 71. 10.35208/ert.1000739
IEEE	ÖZKAL C,Arslan Ö "Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ." Environmental Research & Technology, 5, ss.56 - 71, 2022. 10.35208/ert.1000739
ISNAD	ÖZKAL, Can Burak - Arslan, Özkan. "Developing a GMDH-type neural network model for spatial prediction of NOx : A case study of Çerkezköy, Tekirdağ". Environmental Research & Technology 5/1 (2022), 56-71. https://doi.org/10.35208/ert.1000739