Detection of anomalous Nitrogen Dioxide concentration of Ankara: a Reconstruction-based approach

Yıl 2024, ERKEN GÖRÜNÜM, 1 - 1

Mustafa Murat Arat

https://doi.org/10.2339/politeknik.1419512

Öz

Air quality significantly impacts human health, particularly in urban areas, leading to global morbidity and mortality. Elevated air pollutant levels pose health risks, emphasizing the need for timely monitoring and detection. This study adopts an innovative approach to identify anomalies of daily NO2 concentration levels in a district of Ankara, Turkey. Leveraging both traditional statistical approaches and state-of-the-art techniques, the research aims to provide real-time alerts. Employing a multivariate strategy, the study generates new features based on historical and current data, and incorporates periodic variables, as well. Among the methods explored, Variational Autoencoder emerges as noteworthy, exhibiting superior performance with %98 recall, %82 precision and %0.12 false alarm rate. This approach not only demonstrates a high true positive rate, enhancing its efficacy in anomaly detection but also effectively mitigates false alarms, preventing alert fatigue. By using advanced methodologies with a focus on NO2 levels, the study contributes to proactive measures for public health, enabling prompt responses to potential air quality issues.

Anahtar Kelimeler

Air quality, anomaly detection, variational autoencoders, NO2 concentration, machine learning

Ankara'daki anormal Azot Dioksit konsantrasyonunun tespiti: Rekonstrüksiyona-dayalı bir yaklaşım

Öz

Hava kalitesi, özellikle kentsel bölgelerde önemli ölçüde insan sağlığını etkilemekte ve tüm dünyada morbiditeye ve mortaliteye yol açmaktadır. Hava kirleticilerinin yüksek seviyeleri sağlık riski oluşturarak zamanında izleme ve tespitin gerekliliğini vurgulamaktadır. Bu çalışma, Ankara’nın bir ilçesindeki günlük NO2 konsantrasyon seviyesindeki anomalileri belirlemek için yenilikçi bir yaklaşım benimsemektedir. Geleneksel istatistiksel yöntemler ile birlikte son teknoloji teknikleri kullanan bu araştırma gerçek zamanlı uyarılar sağlamayı amaçlamaktadır. Çok değişkenli bir stratejiyi benimseyen çalışma, geçmiş ve güncel verilere dayalı yeni öznitelikleri ve ayrıca periyodik değişkenleri de analizlere dahil etmektedir. İncelenen yöntemler arasında, %98 duyarlılık, %82 kesinlik and %0.12 yanlış alarm oranı ile Varyasyonel Otokodlayıcı yöntemi üstün performans sergileyerek dikkat çekmektedir. Elde edilen model, sadece yüksek bir gerçek pozitif oranı elde etmekle kalmayıp, aynı zamanda yanlış alarmları etkili bir şekilde azaltarak alarm yorgunluğunu önlemektedir. NO2 seviyelerine odaklanan gelişmiş metodolojiler kullanan bu çalışma, halk sağlığına yönelik proaktif önlemlere katkıda bulunarak potansiyel hava kalitesi sorunlarına hızlı yanıt verilmesini sağlamaktadır.

Anahtar Kelimeler

Hava kalitesi, anomali tespiti, varyasyonel otokodlayıcılar, NO2 konsantrasyonu, makine öğrenmesi

Kaynakça

• [1] Brauer M., Hoek G., Smit H. A., d. Jongste J. C., Gerritsen J., Postma D. S., Kerkhof M. and Brunekreef B., "Air pollution and development of asthma, allergy and infections in a birth cohort", Eur. Respir. J., 29: 879-888, (2007).
• [2] Liu Y., Pan J., Zhang H., Shi C., Li G., Peng Z., Ma J., Zhou Y. and Zhang L., "Short-Term Exposure to Ambient Air Pollution and Asthma Mortality", Am. J. Respir. Crit. Care Med., 200(1): 24-32, (2019).
• [3] Rajagopalan S., Al-Kindi S. G. and Brook R. D., "Air Pollution and Cardiovascular Disease: JACC State-of-the-Art Review", J. Am. Coll. Cardiol., 72(17): 2054-2070, (2018).
• [4] Tuśnio N., Fichna J., Nowakowski P. and Tofiło P., "Air Pollution Associates with Cancer Incidences in Poland", Appl. Sci., 10(21), (2020).
• [5] Balogun H., Rantala A., Antikainen H., Siddika N., Amegah A., Ryti N., Kukkonen J., Sofiev M., Jaakkola M. and Jaakkola J., "Effects of Air Pollution on the Risk of Low Birth Weight in a Cold Climate", Appl. Sci., 10(18), (2020).
• [6] http://tinyurl.com/EUTZPforAWS, "EU Action Plan: 'Towards Zero Pollution for Air, Water and Soil'”,(2021).
• [7] http://tinyurl.com/WHOAQG, "WHO global air quality guidelines", (2021).
• [8] http://tinyurl.com/APDataDownload
• [9] https://www.havaizleme.gov.tr
• [10] Anenberg S. C., Mohegh A., Goldberg D. L., Kerr G. H., Brauer M., Burkart K., Hystad P., Larkin A., Wozniak S. and Lamsal L., "Long-term trends in urban NO2 concentrations and associated paediatric asthma incidence: estimates from global datasets", Lancet Planet Health, 6(1): 49–58, (2022).
• [11] Braei M. and Wagner S., "Anomaly Detection in Univariate Time-series: A Survey on the State-of-the-Art", arXiv preprint, (2020).
• [12] Shams S. R., Jahani A., Kalantary S., Moeinaddini M. and Khorasani N., "Artificial intelligence accuracy assessment in NO2 concentration forecasting of metropolises air", Sci. Rep., 11(1): 1805, (2021).
• [13] Yammahia A. A. and Aunga Z., "Forecasting the concentration of NO2 using statistical and machine learning methods: A case study in the UAE", Heliyon, 9(2), (2023).
• [14] Gastaldo P., Liu B., Zhang L., Wang Q. and Chen J., "A Novel Method for Regional NO2 Concentration Prediction Using Discrete Wavelet Transform and an LSTM Network", Comput. Intell. Neurosci., 2021, (2021).
• [15] Jesemann A.-S., Matthias V., Böhner J. and Bechtel B., "Using Neural Network NO2-Predictions to Understand Air Quality Changes in Urban Areas—A Case Study in Hamburg", Atmosphere, 13(11): 1929, (2022).
• [16] Cabaneros S. M. S., Calautit J. K. S. and Hughes B. R., "Hybrid Artificial Neural Network Models for Effective Prediction and Mitigation of Urban Roadside NO2 Pollution", Energy Procedia, 142: 3524-3530, (2017).
• [17] Aggarwal A. and Toshniwal D., "Detection of anomalous nitrogen dioxide (NO2) concentration in urban air of India using proximity and clustering methods", J&AWMA, 69(7): 805-822, (2019).
• [18] van Zoest V., Stein A., and Hoek G., "Outlier Detection in Urban Air Quality Sensor Networks", Water Air Soil Pollut., 229(111), (2018).
• [19] Torres J. M., Nieto P. G., Alejano L. and Reyes A., "Detection of outliers in gas emissions from urban areas using functional data analysis", J. Hazard. Mater., 186(1): 144-149, (2011).
• [20] Can A. and Özsoy H., "A Different Perspective on Air Pollution Measurements", Journal of Polytechnic, 26(1): 329 - 344, (2023).
• [21] Radojević D., Antanasijević D., Perić-Grujić A., Ristić M., Pocajt V., "The significance of periodic parameters for ANN modeling of daily SO2 and NOx concentrations: A case study of Belgrade, Serbia", Atmos. Pollut. Res., 10(2): 621-628, (2019).
• [22] http://tinyurl.com/TSIVehicle2023, “Turkish Statistical Institute - Turkish Statistics for Road Motor Vehicles” (2023).
• [23] http://tinyurl.com/MoEUCCAPCAAP, “Republic of Türkiye - Ministry of Environment, Urbanisation and Climate Change - Ankara Province Clean Air Action Plan 2020-2024", (2020).
• [24] Rousseeuw P. J. and Croux C., "Alternatives to the Median Absolute Deviation," JASA, 88(424): 1273-1283, (1993).
• [25] Géron A., "Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow", 2nd Edition, O'Reilly Media, Inc., (2019).
• [26] http://tinyurl.com/ISRTELL, “Information System of Regulations - Turkish Environmental Law”, (1983).
• [27] http://tinyurl.com/ISRByLaw, “Information System of Regulations By-law on Air Quality Assessment and Management", (2008).
• [28] http://tinyurl.com/ISRCircular, “Information System of Regulations - Circular on Air Quality Assessment and Management”, (2013).
• [29] Iglewicz B. and Hoaglin D. C., "How to Detect and Handle Outliers", ASQC Quality Press, Milwaukee, Wisconsin, (2004).
• [30] Tukey J.W. and McLaughlin D.H., "Less Vulnerable Confidence and Significance Procedures for Location Based on a Single Sample: Trimming/Winsorization 1", Sankhya: Indian J. Stat., 25(3): 331-352, (1963).
• [31] Pearson R., "Outliers in process modeling and identification", IEEE Trans. Control Syst. Technol., 10(1): 55-63, (2002).
• [32] Schölkopf B., Platt J. C., Shawe-Taylor J., Smola A.J. and Williamson R.C., "Estimating the support of a high-dimensional distribution", Neural Computation, 13(7): 1443–1471, (2001)
• [33] Liu F.T., Ting K.M. and Zhou Z.-H., "Isolation-Based Anomaly Detection", ACM Transactions on Knowledge Discovery from Data, 6(1): 1–39, (2012).
• [34] Bengio Y., Lamblin P., Popovici D. and Larochelle H., "Greedy Layer-Wise Training of Deep Networks", Advances in Neural Information Processing Systems 19 (NIPS 2006), Vancouver, B.C., Canada, (2006).
• [35] Zong B., Song Q., Min M.R., Cheng W., Lumezanu C., Cho D.-K. and Chen H., "Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection", International Conference on Learning Representations, Vancouver, B.C., Canada, (2018).
• [36] Gong D., Liu L., Le V., Saha B., Mansour M.R., Venkatesh S. and Hengel A., "Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection", arXiv preprint, (2019).
• [37] Kingma D.P. and Welling M., "Auto-Encoding Variational Bayes", International Conference on Learning Representations, Scottsdale, Arizona, USA, (2013)
• [38] Awadh K. and Akbaş A., "Intrusion Detection Model Based on TF.IDF and C4.5 Algorithms", Journal of Polytechnic, 24(4): 1691-1698, (2021).
• [39] Duncan B.N., Yoshida Y., de Foy B., Lamsal L.N., Streets D.G., Lu Z., Pickering K.E. and Krotkov N.A., "The observed response of Ozone Monitoring Instrument (OMI) NO2 columns to NOx emission controls on power plants in the United States: 2005-2011", Atmos. Environ., 81: 102-111, (2013).
• [40] Matandirotya N. and Burger R., "An assessment of NO2 atmospheric air pollution over three cities in South Africa during 2020 COVID-19 pandemic", Air Qual Atmos Health, 16: 263-276, (2023).
• [41] U.S. Environmental Protection Agency, "Overview of Nitrogen Dioxide (NO2) Air Quality in the United States," 29 June 2023. [Online]. Available: http://tinyurl.com/epaNO2. [Accessed 28 December 2023].
• [42] Dėdelė A. and Miškinytė A., "A. Seasonal variation of indoor and outdoor air quality of nitrogen dioxide in homes with gas and electric stoves", Environ. Sci. Pollut. Res., 23: 17784-17792, (2016).
• [43] Kavalcı Yılmaz E. and Bakır H., "Hyperparameter Tunning and Feature Selection Methods for Malware Detection”, Journal of Polytechnic preprint, (2024).