DOI QR코드

DOI QR Code

Machine learning-based Fine Dust Prediction Model using Meteorological data and Fine Dust data

기상 데이터와 미세먼지 데이터를 활용한 머신러닝 기반 미세먼지 예측 모형

  • KIM, Hye-Lim (Dept. of Urban Engineering, Gyeongsang National University) ;
  • MOON, Tae-Heon (Dept. of Urban Engineering, ERI. Gyeongsang National University)
  • 김혜림 (경상대학교 도시공학과 대학원) ;
  • 문태헌 (경상대학교 도시공학과)
  • Received : 2021.02.22
  • Accepted : 2021.03.10
  • Published : 2021.03.31

Abstract

As fine dust negatively affects disease, industry and economy, the people are sensitive to fine dust. Therefore, if the occurrence of fine dust can be predicted, countermeasures can be prepared in advance, which can be helpful for life and economy. Fine dust is affected by the weather and the degree of concentration of fine dust emission sources. The industrial sector has the largest amount of fine dust emissions, and in industrial complexes, factories emit a lot of fine dust as fine dust emission sources. This study targets regions with old industrial complexes in local cities. The purpose of this study is to explore the factors that cause fine dust and develop a predictive model that can predict the occurrence of fine dust. weather data and fine dust data were used, and variables that influence the generation of fine dust were extracted through multiple regression analysis. Based on the results of multiple regression analysis, a model with high predictive power was extracted by learning with a machine learning regression learner model. The performance of the model was confirmed using test data. As a result, the models with high predictive power were linear regression model, Gaussian process regression model, and support vector machine. The proportion of training data and predictive power were not proportional. In addition, the average value of the difference between the predicted value and the measured value was not large, but when the measured value was high, the predictive power was decreased. The results of this study can be developed as a more systematic and precise fine dust prediction service by combining meteorological data and urban big data through local government data hubs. Lastly, it will be an opportunity to promote the development of smart industrial complexes.

미세먼지는 질병, 산업·경제에 부정적인 영향을 미치고 있어 국민들은 미세먼지에 대해 예민하게 반응하고 있다. 따라서 미세먼지의 발생을 예측할 수 있다면, 미리 대응책을 마련할 수 있어 생활과 경제에 도움이 될 수 있다. 미세먼지의 발생은 기상과 미세먼지 배출원의 밀집 정도에 영향을 받는다. 산업부문은 미세먼지 배출량이 가장 많으며, 그 중에 산단은 공장들이 미세먼지 배출원이 되어 더 많은 미세먼지를 배출하는 문제가 있다. 본 연구는 지방도시에서 노후산업단지가 있는 지역을 선정하여, 미세먼지를 일으키는 요인을 탐색하고, 미세먼지 발생을 예측할 수 있는 예측모형을 개발하고자 한다. 기상 데이터와 미세먼지 관련 데이터를 활용하였고, 다중회귀분석을 통해 미세먼지 발생에 영향을 미치는 변수를 추출하였다. 이를 토대로 머신러닝 회귀학습기 모형으로 학습하여 예측력이 높은 모형을 추출하였고, 검증용 데이터를 이용하여 예측 모형의 성능을 검증하였다. 그 결과, 예측력이 높은 모형은 선형회귀모형, 가우스 과정 회귀모형, 서포트 벡터 머신으로 나타났으며, 훈련용 데이터의 비율과 예측력은 비례하지 않은 것으로 나타났다. 또한 예측치와 실측치 차이의 평균치는 크지 않지만, 미세먼지 실측치가 높을 때, 예측력이 다소 떨어지는 것으로 나타났다. 본 연구의 결과는 지자체 데이터 허브를 통해 기상데이터와 관련 도시 빅데이터를 결합함으로써 보다 체계적이고 정밀한 미세먼지 예측 서비스로 개발이 가능할 것이며, 스마트산단의 발전을 촉진하는 계기가 될 것이다.

Keywords

References

  1. AirKorea. 2021. Data Retrieve. https://www.airkorea.or.kr. (Accessed January 25, 2021).
  2. An, M.H. and Ryoo, M.H. 2016. Modeling Stochastic Volatility Using Gaussian Processes. The Korean Journal Of Financial Engineering 2(0):101-113.
  3. Bae, S.W. and Yu, J.S. 2018. Predicting the Real Estate Price Index Using Machine Learning Methods and Time Series Analysis Model. Korean Association For Housing Policy Studies 26(1):107-133. https://doi.org/10.24957/hsr.2018.26.1.107
  4. Cha, J.W. and Kim, J.Y. 2018. Development of Data Mining Algorithm for Implementation of Fine Dust Numerical Prediction Model. Korea Institute of information and Communication Engineering 22(4):595-601.
  5. Cho, K.W., Jeong, Y.J., Lee, J.S. and Oh, C.H. 2019. PM10 Particulate Matters Concentration Prediction using Korea Institute of information and Communication Engineering 23(2):632-634.
  6. Choi, C.H., Kim, J.S., Kim, J.H., Kim, H.Y., Lee, W.J. and Kim, H.S. 2017. Development of Heavy Rain Damage Prediction Function Using Statistical Methodology. Korean Society of Hazard Mitigation 17(3):331-338. https://doi.org/10.9798/kosham.2017.17.3.331
  7. Heo, S.Y., Kim, J.Y. and Moon, T.H. Predicting Crime Risky Area Using Machine Learning. Journal of the Korean Association of Geographic Information Studies 2018. 21(4):64-80. https://doi.org/10.11108/KAGIS.2018.21.4.064
  8. Jeon, S.H. and Son, Y.S. 2018. Prediction of fine dust PM10 using a deep neural network model. The Korean Journal of Applied Statistics 31(2):265-285. https://doi.org/10.5351/KJAS.2018.31.2.265
  9. Jeong, Y.H. and Park, J.K. 2017. Energy Storage system Strategy under Gaussian Process Regression. The Korean Institute of Industrial Engineers 2017(04):2690-2695.
  10. Jung, Y.J., Cho, K.W., Lee, J.S. and Oh, C.H. 2020. PM10 Binary Classification Model based on SVM Algorithm. Korea Institute of information and Communication Engineering 24(1):308-310.
  11. Kim, B.H., Lee, H.Y. and Lee, S.M. 2019. The Inutitute of Electronic and information Engineers 6(11):53-60.
  12. Kim, H. 2020. The Prediction of PM2.5 in Seoul through XGBoost ensemble. Journal of the Korean Data Analysis Society 22(4):1661-1671. https://doi.org/10.37727/jkdas.2020.22.4.1661
  13. Kim, J.S., Choi, C.H., Kim, D.H., Lee, M.J. and Kim, H.S. 2017. Development of Heavy Rain Damage Prediction Function Using Artificial Neural Network and Multiple Regression Model. Korean Society of Hazard Mitigation 17(6):73-80. https://doi.org/10.9798/kosham.2017.17.6.73
  14. Kyung, S.Y., Kim, Y.S., Kim, W.J., Park, M.S., Song. J.W., Yum, H.K., Yoon, H.G., Rhee, C.K. and Jeong, S.H. 2015. Guideline for the prevention and managementof particulate matter/Asian dust particleinduced adverse health effect on the patients withpulmonary diseases. Journal of the Korean Medical Association 58(11):1060-1069. https://doi.org/10.5124/jkma.2015.58.11.1060
  15. Lee, A.R. and Jeong, S.J. 2019. Korean Meteorological Society 2019(10):357-357.
  16. Lee, D.W. and Lee, S.W. 2020. Hourly Prediction of Particulate Matter (PM2.5) Concentration Using Time Series Data and Random Forest. Korea Information Processing Society 9(4):129-136.
  17. Lee, Y.S. and Moon, P.J. 2017. A Comparison and Analysis of Deep Learning Framework. Korea Institute of Electronic Communication Sciences 12(1):115-122.
  18. Lim, J.M. 2019. An Estimation Model of Fine Dust Concentration Using Meteorological Environment Data and Machine Learning. Journal of Information Technology Services 18(1):173-186. https://doi.org/10.9716/KITS.2019.18.1.173
  19. Lim, J.M., Ko, S.H. and Kim, J.W. 2018. An estimation model of fine dust concentration using weather data and machine learning. Korea Society of IT Services 2018:691-694.
  20. Ministry of Environment. 2017. Comprehensive measures for fine dust management. pp.1-37.
  21. Ministry of Trade, Industry and Energy. 2020. Implementation strategies of Smart Green industrial complex. pp.1-27.
  22. Oh, J.M., Shin, H.S., Shin, Y.S. and Jeong. H.C. 2017. Forecasting the Particulate Matter in Seoul using a Univariate Time Series Approach. Korea Information Processing Society 19(5):2457-2468.
  23. Open MET Data Portal. 2021. Data. https://data.kma.go.kr. (Accessed January 25, 2021).
  24. Sohn, K.T. and Kim, D.H. 2015. Development of statistical forecast model for PM10 concentration over Seoul. Journal of the Korean data & information science society 26(2):289-299. https://doi.org/10.7465/jkdi.2015.26.2.289
  25. Sung, J.H. and Cho, Y.S. 2019. Machine Learning Approach for Pattern Analysis of Energy Consumption in Factory. Korea Information Processing Society 8(4):87-92.
  26. Sung, S.H., Kim, S.J. and Ryu, M.H. 2020. A Comparative Study on the Performance of Machine Learning Models for the Prediction of Fine Dust: Focusing on Domestic and Overseas Factors. Korea Society Of Innovation 15(4):339-357.
  27. Yeo, M.S. and Kim, J.H. 2019. The Society Of Air-Conditioning And Refrigerating Engineers Of Korea 48(12):44-50.
  28. Yoon, C.J., Ko, H.G., Bang, M.S. and Kwon, H.B. 2018. An Analysis of the Rail Wear Measurements for the Prediction of Particulate Matter Emission in Urban Railway. Journal of Korean Society for Urban Railway 6(4):339-350. https://doi.org/10.24284/jkosur.2018.12.6.4.339