Predlog modela za predviđanje koncentracije suspendovanih (PM2.5) čestica u vazduhu / Proposing the Predictive Model for Airborne PM2.5 Concentration

Energija, ekonomija, ekologija, 3, XXV (2023) (стр 39- 44)

АУТОР(И) / AUTHOR(S): Filip Nastić

Е-АДРЕСА / E-MAIL: filip.nastic@uni.kg.ac.rs

Download Full Pdf   

DOI: 10.46793/EEE23-3.39N

САЖЕТАК / ABSTRACT:

Rastući broj istraživanja ukazuje na negativan uticaj suspendovanih (PM) čestica na zdravlje ljudi. Jedan od načina da se izbegnu njihove negativne posledice, jeste blagovremena predikcija koncentracije PM2.5 u vazduhu. Znajući časovnu koncentraciju, građani bi mogli organizovati svoje dnevne aktivnosti sa ciljem smanjenja njhovog izlaganja intezivnom zagađenju. U cilju formiranja optimalnog modela za časovnu predikciju koncentracije PM2.5 u vazduhu, analizirane su prediktivne performanse tri različita algoritma mašinskog učenja: “Random forest”, “XGBoost” i “Light gradient boosting machine”. Koristeći pomenute algoritme mašinskog učenja stvoreni su modeli koji koristeći meteorološke i hronološke podatke mogu izvršiti predikciju koncentracije PM2.5 na časovnom nivou sa zadovoljavajućom tačnošću. Podaci o koncentraciji PM2.5 su prikupljeni laserskim senzorom u gradu Kragujevcu, čija su očitavanja preuzeta sa sensor.community otvorene baze podataka. Evaluacija modela je izvršena koristeći koeficijent determinacije (R2), osrednjenu apsolutnu grešku (MAE)  i koren srednje kvadratne greške (RMSE).

КЉУЧНЕ РЕЧИ / KEYWORDS:

Zagađenje vazduha, časovna predikcija, mašinsko učenje, PM2.5, zdravlje ljudi

ЛИТЕРАТУРА / REFERENCES:

  • Jurišević, N., Stojadinović, M., Končalović, D., Josijević, M., Gordić, D. Students’ Perceptions of Air Quality – an Opportunity for More Sustainable Urban Transport in the Medium-sized University city in the Balkans, Tehnika, Vol. 78, No. 4, pp. 455-463, 2023. https://doi.org/10.5937/tehnika2304455J
  • World Health Organization, Air quality and health, https://www.who.int/teams/environment-climate-change-and-health/air-quality-and-health/health-impacts/types-of-pollutants [pristupljeno 30.4.2023]
  • Nansai, K., Tohno, S., Chatani, S., Kanemoto, K., Kagawa, S., Kondo, Y., Takayanagi, W., Lenzen, M. Consumption in the G20 nations causes particulate air pollution resulting in two million premature deaths annually. Nature Communications, Vol. 12, No. 6286, 2021. https://doi.org/10.1038/s41467-021-26348-y
  • United States Environmental Protection Agency. Particulate Matter (PM) Basics, https://www.epa.gov/pm-pollution/particulate-matter-pm-basics [pristupljeno 30.4.2023]
  • World Health Organization, Regional Office for Europe. Health effects of particulate matter: policy implications for countries in eastern Europe, Caucasus and central Asia. 2013. https://iris.who.int/handle/10665/344854 [pristupljeno 30.4.2023]
  • Đurišić, Ž., Škrbić, B., Potencijal energije sunca i vetra za strateško planiranje dekarbonizacije proizvodnje električne energije u Srbiji, Energija, ekonomija, ekologija, Vol. 24, No. 4, pp. 1-11, 2022. https://doi.org/10.46793/EEE22-4.01D
  • Chen, Z., Chen, D., Zhao, C., Kwan, M., Cai, J., Zhuang, Y, Zhao, B., Wang, X., Chen, B., Yang, J., Li, R., He, B., Gao, B., Wang, K., Xu, B. Influence of meteorological conditions on PM5 concentrations across China: A review of methodology and mechanism, Environment International, Vol. 139, No. 105558, 2020. https://doi.org/10.1016/j.envint.2020.105558
  • Megaritis, A.G., Fountoukis, C., Charalampidis, P.E., Denier Van Der Gon, H.A.C., Pilinis, C., Pandis, S.N. Linking climate and air quality over Europe: Effects of meteorology on PM5 concentrations. Atmospheric Chemistry and Physics, Vol. 14, No. 18, pp. 10283-10298, 2014. https://doi.org/10.5194/acp-14-10283-2014
  • Zalakeviciute, R., López-Villada, J., Rybarczyk, Y. Contrasted effects of relative humidity and precipitation on urban PM5 pollution in high elevation urban areas, Sustainability, Vol. 10, No. 6, pp. 2064, 2018. https://doi.org/10.3390/su10062064
  • Wang, J., Ogawa, S. Effects of meteorological conditions on PM5 concentrations in Nagasaki, Japan, International Journal of Environmental Research and Public Health, Vol. 12, No. 8, pp. 9089-9101, 2015. https://doi.org/10.3390/ijerph120809089
  • Hernandez, G., Berry, T.-A., Wallis, S.L., Poyner, D. Temperature and humidity effects on particulate matter concentrations in a sub-tropical climate during winter, in Proc. International Proceedings of Chemical, Biological and Environmental Engineering, (ICECB 2017), https://doi.org/10.7763/IPCBEE.2017.V102.10
  • Chang, L.T.C., Scorgie, Y., Duc, H.N., Monk, K., Fuchs, D., Trieu, T. Major source contributions to ambient PM5 and exposures within the New South Wales Greater Metropolitan Region, Atmosphere, Vol. 10, No.3, pp. 138, 2019. https://doi.org/10.3390/atmos10030138
  • Doreswamy, H.K.S., Km, Y., Gad, I. Forecasting air pollution particulate matter (PM5) using machine learning regression models, Procedia Computer Science, Vol. 171, pp. 2057-2066, 2020. https://doi.org/10.1016/j.procs.2020.04.221
  • Kumar, S., Mishra, S., Singh, S.K. A machine learning-based model to estimate PM5 concentration levels in Delhi’s atmosphere, Heliyon, Vol. 6, No. 11, pp. E05618, 2020. https://doi.org/10.1016/j.heliyon.2020.e05618
  • Shahriar, S.A., Kayes, I., Hasan, K., Hasan, M., Islam, R., Awang, N.R., Hamzah, Z., Rah, A., Salam, M. Potential of ARIMA-ANN, ARIMA-SVM, DT and CatBoost for Atmospheric PM5 forecasting in Bangladesh, Atmosphere, Vol. 12, No. 1, pp. 100, 2021. https://doi.org/10.3390/atmos12010100
  • Zhong, J., Zhang, X., Gui, K., Wang, Y., Che, H., Shen, X., Zhang, L., Zhang, Y., Sun, J., Zhang, W. Robust prediction of hourly PM5 from meteorological data using LightGBM, National Science Review, Vol. 8, No. 10, nwaa307, 2021. https://doi.org/10.1093/nsr/nwaa307
  • Liu, H.Y., Schneider, P., Haugen, R., Vogt, M. Performance assessment of a low-cost PM 2.5 sensor for a near four-month period in Oslo, Norway, Atmosphere, Vol. 10, No. 2, pp. 41, 2019. https://doi.org/10.3390/atmos10020041
  • Archive – Sensor.Community.https://archive.sensor.community/ [pristupljeno 03.05.2023]
  • Data Access Viewer, https://power.larc.nasa.gov/data-access-viewer/ [pristupljeno 03.05.2023]
  • Shuvho, M.B.A., Chowdhury, M.A., Ahmed, S., Kashem, M.A. Prediction of solar irradiation and performance evaluation of grid connected solar 80KWp PV plant in Bangladesh, Energy Reports, Vol. 5, pp. 714-722, 2019. https://doi.org/10.1016/j.egyr.2019.06.011
  • Parameter tuning, https://catboost.ai/en/docs/concepts/parameter-tuning [pristupljeno 10.4.2023]
  • Pan, B. Application of XGBoost algorithm in hourly PM5 concentration prediction, in Proc. IOP Conference Series: Earth and Environmental Science, Vol. 113, 3rd International Conference on Advances in Energy Resources and Environment Engineering 8-10 December 2017, Harbin, China, 2018. https://doi.org/10.1088/1755-1315/113/1/012127
  • AlDaweesh, S.A. Predicting hourly particulate matter (PM5) concentrations using meteorological data, in Proc. International Conference of Computing, Electronics & Communications Engineering (iCCECE), pp. 136-140, London, 2019. https://doi.org/10.1109/iCCECE46942.2019.8941696
  • Sihag, P., Kumar, V., Afghan, F.R., Pandhiani, S.M., Keshavarzi, A. Predictive modeling of PM5 using soft computing techniques: case study-Faridabad, Haryana, India, Air Quality, Atmosphere & Health, Vol. 12, pp. 1511-1520, 2019. https://doi.org/10.1007/s11869-019-00755-z