Prediction of Air Quality Index of Sihhiye Region by Extreme Learning Machines and Artificial Neural Networks

Year 2022, Volume: 02 Issue: 01, 1 - 18, 31.07.2022

Burhan Baran

Abstract

With this study, it was aimed to estimate the air quality index (AQI) in the Sihhiye region with both extreme learning machines (ELM) and artificial neural networks (ANN) algorithms. For this purpose, seven parameters that could affect the AQI had been chosen. These parameters were PM10, SO2, CO, temperature, humidity, pressure and wind speed. Firstly, correlation analysis was performed between the AQI and these seven parameters. According to the results of the analysis, it was concluded that the strongest relation with the AQI were with PM10 from the atmospheric parameters and the pressure from the meteorological parameters. The parameter values for August, October, November and December of 2018 year were determined as training data. The parameter values for the first 14 days of January and February of 2019 year were determined as test data. AQI values were classified mathematically between 1 and 6. Classification studies were applied to both raw data and normalized data. In the classification process, different training functions and hidden neuron numbers were used in algorithms. 3-fold cross-validation was performed for the reliability of the results. The activation function and neuron numbers with highest performance were applied to actual test data. Finally, mathematical classification results were compared with the predicted classification values of AQI. According to the results obtained, in the classification studies conducted with both raw and normalized data, it was observed that ELM algorithm achieved more successful results than ANN algorithm. The success rates were 85.71% in raw data and 71.43% in normalized data.

Keywords

Air quality index, Extreme learning machine, Artificial neural networks, Classification, Correlation.

Sıhhiye Bölgesi Hava Kalitesi İndeksinin Aşırı Öğrenme Makineleri ve Yapay Sinir Ağları ile Tahmini

Abstract

Bu çalışma ile Sıhhiye bölgesindeki hava kalitesi indeksinin (HKİ) hem aşırı öğrenme makineleri (AÖM) hem de yapay sinir ağları (YSA) algoritmaları ile tahmin edilmesi amaçlanmıştır. Bu amaçla, HKİ’yi etkileyebilecek yedi adet parametre seçilmiştir. Bu parametreler PM10, SO2, CO, sıcaklık, nem, basınç ve rüzgâr hızıdır. İlk olarak, HKİ ile bu yedi parametre arasında korelasyon analizi yapılmıştır. Analiz sonucuna göre HKİ ile en güçlü ilişkinin atmosferik parametrelerden PM10 ile, meteorolojik parametrelerden ise basınç ile olduğu sonucuna ulaşılmıştır. 2018 yılının Ağustos, Ekim, Kasım ve Aralık aylarına ait parametre değerleri eğitim verisi olarak belirlenmiştir. 2019 yılının Ocak ve Şubat aylarına ait ilk 14 günlük parametre verileri ise test verisi olarak belirlenmiştir. HKİ değerleri 1 ile 6 arasında matematiksel olarak sınıflandırılmıştır. Sınıflandırma çalışmaları hem ham veriler hem de normalize edilmiş veriler ile gerçekleştirilmiştir. Sınıflandırma sürecinde algoritmalarda farklı eğitim fonksiyonları ve gizli nöron sayıları kullanılmıştır. Sonuçların güvenilirliği için 3-kat çapraz doğrulama yapılmıştır. En yüksek performansa sahip aktivasyon fonksiyonları ve nöron sayıları gerçek test verilerine uygulanmıştır. Son olarak, HKİ’nin matematiksel sınıflandırma sonuçları ile tahmini sınıflandırma sonuçları karşılaştırılmıştır. Elde edilen sonuçlara göre hem ham hem de normalize veriler ile yapılan sınıflandırma çalışmalarında AÖM algoritmasının YSA algoritmasından daha başarılı sonuçlar elde ettiği görülmüştür. Başarım oranları ham verilerde %85.71, normalize verilerde %71.43 olarak gerçekleşmiştir.

Keywords

Hava kalitesi indeksi, Aşırı öğrenme makineleri, Yapay sinir ağları, Sınıflandırma, Korelasyon.

References

• [1] Shaban, K.B., Kadri, A. and Rezk, E., 2016. Urban Air Pollution Monitoring System With Forecasting Models. IEEE Sensors Journal, 16 (8). DOI: 10.1109/JSEN.2016.2514378.
• [2] Yang, Z. and Wang, J., 2017. A new air quality monitoring and early warning system: Air quality assessment and air pollutant concentration prediction. Environmental Research, https://www.sciencedirect.com/science/journal/00139351/158/supp/C158, 105-117. https://doi.org/10.1016/j.envres.2017.06.002.
• [3] Karamchandani, S. and Gupta, D., 2016. Pervasive monitoring of carbon monoxide and methane using air quality prediction. 3rd International Conference on Computing for Sustainable Global Development (INDIACom).
• [4] Karacı, A., 2018. Akıllı Şehir Hava Takip Sistemi ve Astım Hastaları için PM2.5 Konsantrasyonu Ölçüm Aracının Geliştirilmesi, Mühendislik Bilimleri ve Tasarım Dergisi, 6:3, 418 – 425. https://doi.org/10.21923/jesd.412665.
• [5] Turalioglu, F.S., 2005. An Assessment on Variation of Sulphur Dioxide and Particulate Matter in Erzurum (Turkey). Environmental Monitoring Assessment, 104, 119–130.
• [6] EPA, 2019. Birleşik Devletler Çevre Koruma Ajansı. Karbon monoksit Hava Kirliliği. https://www.epa.gov/.
• [7] Rao, K.S. Devi, G.L. and Ramesh, N., 2019. Air Quality Prediction in Visakhapatnam with LSTM based Recurrent Neural Networks. International Journal of Intelligent Systems and Applications, 2, 18-24. DOI: 10.5815/ijisa.2019.02.03.
• [8] Temiz, İ. and Turgut, D., 2015. Time Series Analysis and Forecasting For Air Pollution In Ankara: A Box-Jenkıns Approach. Alphanumeric Journal, 3 (2), 131-138. https://doi.org/10.17093/aj.2015.3.2.5000148347.
• [9] Sevinç, E., 2022. An empowered AdaBoost algorithm implementation: A COVID-19 dataset study. Computers & Industrial Engineering, 165:107912. doi: 10.1016/j.cie.2021.107912.
• [10] Cihan, P., Ozel, H., & Ozcan, H. K. (2021). Modeling of atmospheric particulate matters via artificial intelligence methods. Environmental Monitoring and Assessment, 193 (5), 1-15. https://doi.org/10.1007/s10661-021-09091-1.
• [11] Baran, B., 2021. Air Quality Index Prediction in Besiktas District by Artificial Neural Networks and K Nearest Neighbours. Journal of Engineering Sciences and Design, 9(1): 52 – 63. DOI: 10.21923/jesd.671836.
• [12] Shishegaran, A., Saeedi, M., Kumar, A. and Ghiasinejad, H., 2020. Prediction of air quality in Tehran by developing the nonlinear ensemble model. Journal of Cleaner Production, 259, 120825. https://doi.org/10.1016/j.jclepro.2020.120825.
• [13] Wang, J., Du, P., Hao, Y., Ma, X., Niu, T. and Yang, W., 2020. An innovative hybrid model based on outlier detection and correction algorithm and heuristic intelligent optimization algorithm for daily air quality index forecasting. Journal of Environmental Management, 255, 109855. https://doi.org/10.1016/j.jenvman.2019.109855.
• [14] Baran, B., 2019. Aşırı Öğrenme Makineleri ile Rüzgar Hızına Bağlı Enerji Tahmini: Malatya Örneği. 1. Ulusal Mühendislik ve Teknoloji Kongresi (UMTK). 56-62.
• [15] Liu, H., Li, Q., Yu, D. and Gu, Y., 2019. Air Quality Index and Air Pollutant Concentration Prediction Based on Machine Learning Algorithms. Applied Sciences, 9 (4069), 1-9. https://doi.org/10.3390/app9194069.
• [16] Sevinç, E., 2019. A Novel Evolutionary Algorithm for Data Classification Problem With Extreme Learning Machines, IEEE Access, 7:122419-122427, doi: 10.1109/ACCESS.2019.2938271.
• [17] Zou, Z., Cai, T. and Cao, K., 2019. An urban big data-based air quality index prediction: A case study of routes planning for outdoor activities in Beijing. Environment and Planning B: Urban Analytics and City Science. https://doi.org/10.1177/2399808319862292.
• [18] Bai, Y., Li, Y., Wang, X., Xie, J. and Li, C., 2016. Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions. Atmospheric Pollution Research, 7 (3), 557-566. https://doi.org/10.1016/j.apr.2016.01.004.
• [19] Baran, B., 2019. Prediction of Air Quality Index by Extreme Learning Machines. 2019 International Artificial Intelligence and Data Processing Symposium (IDAP), 1-8. doi: 10.1109/IDAP.2019.8875910.
• [20] Zhang, J. and Ding, W., 2017. Prediction of Air Pollutants Concentration Based on an Extreme Learning Machine: The Case of Hong Kong. International Journal of Environmental Research and Public Health, 14 (2), 114. doi: 10.3390/ijerph14020114.
• [21] Sevinç, E., 2018. Activation functions in single hidden layer feed-forward neural networks. Selcuk University Journal of Engineering Sciences, 1-13.
• [22] Zhu, S., Lian, X., Liu, H., Hu, J., Wang, Y.Y. and Che, Z., 2017. Daily air quality index forecasting with hybrid models: A case in China. Environmental Pollution, 231 (2), 1232-1244. DOI: 10.1016/j.envpol.2017.08.069
• [23] Peng, H., Lima, A.R., Teakles, A., Jin, J., Cannon, A.J. and Hsieh, W.W., 2017. Evaluating hourly air quality forecasting in Canada with nonlinear updatable machine learning methods. Air Quality, Atmosphere, & Health, 10 (2), 195-211.
• [24] Jose, L.A., 2017. Probabilistic forecasting for extreme NO2 pollution episodes. Environmental Pollution, 229, 321-328. https://doi.org/10.1016/j.envpol.2017.05.079.
• [25] Patricio, P. and Ernesto, G., 2016. Forecasting hourly PM2.5 in Santiago de Chile with emphasis on night episodes. Atmospheric Environment, 124, 22-27. https://doi.org/10.1016/j.atmosenv.2015.11.016.
• [26] Avşar, E., 2015. Balıkesir İli Burhaniye İlçesi (İskele Mahallesi) hava kalitesinin Değerlendirilmesi. Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, 4(1): 68-82. https://doi.org/10.17798/beufen.40291
• [27] Vong, CM., Fai Ip, W., Wong, PK. and Chiu, CC., 2014. Predicting minority class for suspended particulate matters level by extreme learning machine. Neurocomputing, 128, 136-144. https://doi.org/10.1016/j.neucom.2012.11.056.
• [28] Moustris, K.P., Ziomas, I.C. and Paliatsos, A.G., 2010. 3-Day-Ahead Forecasting of Regional Pollution Index for the Pollutants NO2, CO, SO2, and O3 Using Artificial Neural Networks in Athens, Greece. Water, Air, & Soil Pollution, 209 (1–4), 29–43.
• [29] Biancofiore, F., Busilacchio, M., Verdecchia, M., Tomassetti, B., Aruffo, E., Bianco, S., Di Tommaso, S., Colangeli, C., Rosatelli, G. and Di Carlo, P., 2017. Recursive neural network model for analysis and forecast of PM10 and PM2.5. Atmospheric Pollution Research, 8, 652–659. https://doi.org/10.1016/j.apr.2016.12.014.
• [30] Mekparyup, J. and Saithanu, K., 2020. Air Quality Index Prediction in the Eastern Regions of Thailand with Accuracy of Neural Networks. International Journal of Applied Engineering Research, 15 (5), 436-444.
• [31] Liu, B.C., Binaykia, A., Chang, P.C., Tiwari, M.K. and Tsao, C.C., 2017. Urban air quality forecasting based on multi-dimensional collaborative Support Vector Regression (SVR): A case study of Beijing-Tianjin-Shijiazhuang. PLoSONE, 12 (7), e0179763. https://doi.org/10.1371/journal.pone.0179763.
• [32] Ganesh, S.S., Arulmozhivarman, P. and Tatavarti, R., 2017. Forecasting Air Quality Index Using an Ensemble of Artificial Neural Networks and Regression Models. Journal of Intelligent Systems. https://doi.org/10.1515/jisys-2017-0277.
• [33] Saatcioglu, T., Alp, K., Hanedar, A. and Avşar, E., (2011). Effect of the marmaray project on air pollution in Istanbul: An ive model application, Fresenius Environmental Bulletin. 20 (9): 2340-2349,
• [34] Dragomir, E.G., 2010. Air Quality Index Prediction using K-Nearest Neighbor Technique. Bulletin of PG University of Ploiesti, Series Mathematics, Informatics, Physics, 62 (1), 103-108.
• [35] Jiao, Y., Wang, Z. and Zhang, Y., 2019. Prediction of Air Quality Index Based on LSTM. IEEE 8th Joint International Information Technology and Artificial Intelligence Conference (ITAIC), Chongqing-China, 17-20. doi: 10.1109/ITAIC.2019.8785602.
• [36] Kadilar, G.O and Kadilar, C., 2017. Assessing air quality in Aksaray with time series analysis. AIP Conference Proceedings, 1833, 020112.
• [37] Avşar, E., Hanedar, A., Toroz, I., Alp, K. and Kaynak, B., (2010). Investigation of PM10 Concentrations And Noise Levels of The Road Sweepers Operating In Istanbul-Turkey: A Case Study, Fresenius Environmental Bulletin 19 (9b).
• [38] AQI Calculator, 2018. Hava Kalitesi İndeksi Hesaplama Aracı. https://app.cpcbccr.com/ccr_docs/AQI%20-Calculator.xls.
• [39] YSAb, 2019. Yapay Sinir Ağları. https://blogs.mathworks.com/loren/2015/08/04/artificial-neural-networks-for-beginners/.
• [40] Time and Data, 2019. Ankara İli 2018-2019 Hava Durumu Bilgisi.https://www.timeanddate.com/weather/turkey/ankara/historic.
• [41] Taşdelen, B., 2019. Korelasyon ve Regresyon Analizi. Erişim adresi: https://docplayer.biz.tr/47627312-Korelasyon-ve-regresyon-analizi-doc-dr-bahar-tasdelen.html.
• [42] Huang, G.B., Zhu, Q.Y. and Siew, C.K., 2006. Extreme Learning Machine: Theory and Applications. Neurocomputing, 70, 489-501. https://doi.org/10.1016/j.neucom.2005.12.126.
• [43] Kaya, Y., 2014. A Fast Intelligent Diagnosis System For Thyroid Diseases Based On Extreme Learning Machine. Bilim ve Teknoloji Dergisi A-Uygulamalı Bilimler ve Mühendislik, 15 (1). https://doi.org/10.18038/btd-a.89202.
• [44] Baran, B., 2019. Sınır Değerler Arasında Kalan Evsel Atıksu Numune Analizi Sonucunun Aşırı Öğrenme Makineleri ile Sınıflandırılması. Mühendislik Bilimleri ve Tasarım Dergisi, 7 (1), 18 – 25. https://doi.org/10.21923/jesd.457085.
• [45] YSAa, 2019. Yapay Sinir Ağları. http://kod5.org/yapay-sinir-aglari-ysa-nedir/.
• [46] İnalpulat, M., Kızıl, Ü., Bilgücü, E. and Genç, L., 2016. E-Nose Identification of Milk Somatic Cell Count, Çanakkale Onsekiz Mart Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 2:1, 22-35.
• [47] Saplıoğlu, K., Küçükerdem Öztürk, T.S. and Şenel, FA., 2020. Eksik Hidrolojik Verilerin Simbiyotik Organizmalar Arama Algoritması ile Tahmini, Çanakkale Onsekiz Mart Üniversitesi Fen Bilimleri Enstitüsü Dergisi Açık Erişim, 6:1, 93-104. https://doi.org/10.28979/comufbed.628846.
• [48] FFBPNNa, 2019. Feed-forward Back Propagation Neural Network.https://stackoverflow.com/questions/28403782/what-is-the-difference-between-back-propagation-and-feed-forward-neural-network.
• [49] Yavuz, S. and Deveci M., 2012. İstatiksel Normalizasyon Tekniklerinin Yapay Sinir Ağın Performansına Etkisi. Erciyes Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 40 (2), 167-187.