Yapay Zeka ve IoT Kullanarak İzmir'deki Hava Kalitesinin Tahmini

Abstract

Hava kirliliği, Türkiye'nin üçüncü büyük şehri olan İzmir'de önemli bir endişe kaynağıdır ve insan sağlığı ve çevre üzerinde ciddi olumsuz etkileri vardır. Şehir, endüstriyel faaliyetler ve yoğun trafik nedeniyle hava kalitesi sorunlarıyla karşı karşıyadır. Nesnelerin İnterneti (IoT) teknolojisi, hava kalitesini düşüren faktörlerin sürekli izlenmesini ve ölçülmesini sağlar. Hava kirliliğini tahmin etmek için IoT'den yararlanmak, olası olumsuz etkileri azaltmada çok önemlidir. Bu çalışmada, hava kirliliği tahminleri makine öğrenimi, derin öğrenme ve zaman serisi analiz yöntemleri kullanılarak gerçekleştirilmiştir. PM10 ve SO2 seviyelerine ilişkin veriler, 1996'dan 2024'e kadar İzmir'deki yedi lokasyondan toplandı. PM10 ve SO2 ölçümlerini değerlendirmek için kullanılan modeller Destek Vektör Regresyonu (SVR), Mevsimsel Otoregresif Entegre Hareketli Ortalama (SARIMA), Uzun Kısa Süreli Bellek (LSTM) ve Genişletilmiş Uzun Süreli Bellek'i (xLSTM) içeriyordu. Bu modeller arasında xLSTM, LSTM modeline göre biraz daha düşük R² puanlarına rağmen en düşük hata ölçütlerini elde ederek hem PM10 hem de SO2 seviyelerini tahmin etmede genel olarak en iyi performansı göstermiştir.

Keywords

• Hava Kalitesi
• Yapay Zeka
• Nesnelerin İnterneti
• Hava Tahmini
• xLSTM

Predicting Air Quality in Izmir Using Artificial Intelligence and IoT

Abstract

ir pollution is a significant concern in Izmir, the third-largest city in Turkey, and it has serious adverse effects on human health and the environment. The city faces air quality issues due to industrial activities and heavy traffic. The Internet of Things (IoT) technology enables continuous monitoring and measurement of factors that diminish air quality. Utilizing IoT to predict air pollution is crucial in mitigating potential adverse effects. In this study, air pollution predictions were conducted using machine learning, deep learning, and time series analysis methods. Data on PM10 and SO2 levels were collected from seven locations in Izmir from 1996 to 2024. The models used for evaluating PM10 and SO2 measurements included Support Vector Regression (SVR), Seasonal Autoregressive Integrated Moving Average (SARIMA), Long Short-Term Memory (LSTM), and Extended Long-Term Memory (xLSTM). Among these models, xLSTM demonstrated the best overall performance for predicting both PM10 and SO2 levels, achieving the lowest error metrics despite slightly lower R² scores than the LSTM model.

Keywords

• Air Quality
• Artificial Intelligence
• IoT
• Weather Forecasting
• xLSTM

References

1. Abhijith, K. V., Kumar, P., Gallagher, J., McNabola, A., Baldauf, R., Pilla, F., Broderick, B., Di Sabatino, S., & Pulvirenti, B. (2017). Air pollution abatement performances of green infrastructure in open road and built-up street canyon environments: A review. Atmospheric Environment, 162, 71–86.
2. Albuali, A., Srinivasagan, R., Aljughaiman, A., & Alderazi, F. (2023). Scalable lightweight IoT-based smart weather measurement system. Sensors, 23(12), 5569. https://doi.org/10.3390/s23125569
3. Ambildhuke, G., & Banik, B. G. (2022). IoT-based portable weather station for irrigation management using real-time parameters. International Journal of Advanced Computer Science and Applications, 13(5), 267–278.
4. Atali, A., Eren, B., Erden, C., & Atali, G. (2022). LSTM derin öğrenme yaklaşımı ile hava kalitesi verilerinin tahmini: Sakarya örneği [Forecasting air quality data with an LSTM deep learning, approach: The case of Sakarya]. Academic Perspective Procedia, 5(3), 477–484.
5. Aydin, S., Tasyürek, M., & Öztürk, C. (2021). Derin öğrenme yöntemi ile İç Anadolu Bölgesi ve çevresi hava kirliliği tahmini [Air pollution forecasting for Central Anatolia and surroundings using a deep learning method]. Avrupa Bilim ve Teknoloji Dergisi / European Journal of Science and Technology, 29, 168–173.
6. Bansal, P., & Quan, S. J. (2024). Examining temporally varying nonlinear effects of urban form on urban heat island using explainable machine learning: A case of Seoul. Building and Environment, 247, 110957.
7. Bernardes, G. F. L. R., Ishibashi, R., Ivo, A. A. S., Rosset, V., & Kimura, B. Y. L. (2023). Prototyping low-cost automatic weather stations for natural disaster monitoring. Digital Communications and Networks, 9(4), 941–956. https://doi.org/10.1016/j.dcan.2022.05.002
8. Bolla, S., Anandan, R., & Thanappan, S. (2022). Weather forecasting method from sensor-transmitted data for smart cities using IoT. Scientific Programming, 2022, 1426575. https://doi.org/10.1155/2022/1426575

9. Elbasi, E., Mostafa, N., AlArnaout, Z., Zreikat, A. I., Cina, E., Varghese, G., Shdefat, A., Topcu, A. E., Abdelbaki, W., Mathew, S., & Zaki, C. (2023). Artificial intelligence technology in the agricultural sector: A systematic literature review. IEEE Access, 11, 171–202. https://doi.org/10.1109/ACCESS.2022.3232485
10. Elsaraiti, M., & Merabet, A. (2021). A comparative analysis of the ARIMA and LSTM predictive models and their effectiveness for predicting wind speed. Energies, 14(20), 6782.
11. Geng, D., Zhang, H., & Wu, H. (2020). Short-term wind speed prediction based on principal component analysis and LSTM. Applied Sciences, 10(13), 4416.
12. Ioannou, K., Karampatzakis, D., Amanatidis, P., Aggelopoulos, V., & Karmiris, I. (2021). Low-cost automatic weather stations in the Internet of Things. Information, 12(4), 146. https://doi.org/10.3390/info12040146
13. Izmir Metropolitan Municipality. (2024). Hava kalitesi ölçüm değerleri [Air quality measurement values][Dataset]. Retrieved September 16, 2025, from https://ulasav.csb.gov.tr/dataset/35-hava-kalitesiolcum-degerleri
14. Jamil, H., Umer, T., Ceken, C., & Al-Turjman, F. (2021). Decision-based model for real-time IoT analysis using big data and machine learning. Wireless Personal Communications, 121(4), 2947–2959. https://doi.org/10.1007/s11277-021-08857-7
15. Karvelis, P., Mazzei, D., Biviano, M., & Stylios, C. (2020). PortWeather: A lightweight onboard solution for real-time weather prediction. Sensors, 20(11), 3181. https://doi.org/10.3390/s20113181
16. Kaya, S. M., Isler, B., Abu-Mahfouz, A. M., Rasheed, J., & AlShammari, A. (2023). An intelligent anomaly detection approach for accurate and reliable weather forecasting at IoT edges: A case study. Sensors, 23(5), 2426. https://doi.org/10.3390/s23052426
17. Leelavinodhan, P. B., Vecchio, M., Antonelli, F., Maestrini, A., & Brunelli, D. (2021). Design and implementation of an energy-efficient weather station for wind data collection. Sensors, 21(11), 3831. https://doi.org/10.3390/s21113831
18. Lu, B., Wang, R., Qin, Z., & Wang, L. (2023). A practice-distributed thunder-localization system with crowd sourced smart IoT devices. Sensors, 23(9), 4186. https://doi.org/10.3390/s23094186
19. Mabrouki, J., Azrour, M., Dhiba, D., Farhaoui, Y., & El Hajjaji, S. (2021). IoT-based data logger for weather monitoring using Arduino-based wireless sensor networks with remote graphical application and alerts. Big Data Mining and Analytics, 4(1), 25–32. https://doi.org/10.26599/BDMA.2020.9020018
20. Mehmood, A., Lee, K.-T., & Kim, D.-H. (2023). Energy prediction and optimization for smart homes with weather metric-weight coefficients. Sensors, 23(7), 3640. https://doi.org/10.3390/s23073640
21. Mohapatra, D., & Subudhi, B. (2022). Development of a cost-effective IoT-based weather monitoring system. IEEE Consumer Electronics Magazine, 11(5), 81–86. https://doi.org/10.1109/MCE.2021.3136833
22. Nie, H., Liu, G., Liu, X., & Wang, Y. (2012). Hybrid of ARIMA and SVMs for short-term load forecasting. Energy Procedia, 16, 1455–1460. Patel, P., Kalyanam, R., He, L., Aliaga, D., & Niyogi, D. (2023). Deep learning-based urban morphology for city-scale environmental modeling. PNAS Nexus, 2(3), pgad027. https://doi.org/10.1093/pnasnexus/pgad027
23. Roy, D. S. (2020). Forecasting the air temperature at a weather station using deep neural networks. Procedia Computer Science, 178, 38– 46.
24. Singh, D. K., Sobti, R., Jain, A., Malik, P. K., & Le, D.-N. (2022). LoRa-based intelligent soil and weather condition monitoring with Internet of Things for precision agriculture in smart cities. IET Communications, 16(5), 604–618. https://doi.org/10.1049/cmu2.12352
25. Suresh, P., Aswathy, R. H., Arumugam, S., Albraikan, A. A., Al-Wesabi, F. N., Hilal, A. M., & Alamgeer, M. (2022). IoT with evolutionary algorithm-based deep learning for smart irrigation system. Computers, Materials & Continua, 71(1), 1713–1728.
26. Tsalikidis, N., Mystakidis, A., Koukaras, P., Ivaškevičius, M., Morkūnaitė, L., Ioannidis, D., Fokaides, P. A., Tjortjis, C., & Tzovaras, D. (2024). Urban traffic congestion prediction: A multi-step approach utilizing sensor data and weather information. Smart Cities, 7(1), 233–253.
27. Tukey, J. W. (1977). Exploratory data analysis. Addison-Wesley.
28. Venter, Z. S., Hassani, A., Stange, E., Schneider, P., & Castell, N. (2024). Reassessing the role of urban green space in air pollution control. Proceedings of the National Academy of Sciences, 121(6), e2306200121.
29. Vos, P. E. J., Maiheu, B., Vankerkom, J., & Janssen, S. (2013). Improving local air quality in cities: To tree or not to tree? Environmental Pollution, 183, 113–122.
30. Wang, Z., Hu, K., Wang, Z., Yang, B., & Chen, Z. (2024). Impact of urban neighborhood morphology on PM2.5 concentration distribution at different scale buffers. Land, 14(1), 7.
31. Woo, W., Richards, W., Selker, J., & Udell, C. (2023). WeatherChimes: An open IoT weather station and data sonification system. HardwareX, 13, e00402. https://doi.org/10.1016/j.ohx.2023.e00402
32. Wu, Q., Wang, Y., Sun, H., Lin, H., & Zhao, Z. (2023). A system coupling GIS and CFD for atmospheric pollution dispersion simulation in urban blocks. Atmosphere, 14(5), 832.
33. Yang, J., Yu, M., Liu, Q., Li, Y., Duffy, D. Q., & Yang, C. (2022). A high spatiotemporal resolution framework for urban temperature prediction using IoT data. Computers & Geosciences, 159, 104991. https://doi.org/10.1016/j.cageo.2021.104991
34. Yu, M., Xu, F., Hu, W., Sun, J., & Cervone, G. (2021). Using long shortterm memory (LSTM) and Internet of Things (IoT) for localized surface temperature forecasting in an urban environment. IEEE Access, 9, 137406–137418.