Journal of Convergence for Information Technology (융합정보논문지)

Convergence Society for SMB (중소기업융합학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison and analysis of prediction performance of fine particulate matter(PM2.5) based on deep learning algorithm

딥러닝 알고리즘 기반의 초미세먼지(PM2.5) 예측 성능 비교 분석

  • Kim, Younghee (Department of Convergence Engineering, Graduate School of Venture, Hoseo University) ;
  • Chang, Kwanjong (Department of Convergence Engineering, Graduate School of Venture, Hoseo University)
  • 김영희 (호서대학교 벤처대학원 융합공학과) ;
  • 장관종 (호서대학교 벤처대학원 융합공학과)
  • Received : 2021.02.06
  • Accepted : 2021.03.20
  • Published : 2021.03.28
Download PDF
⟨ Previous Next ⟩

Abstract

This study develops an artificial intelligence prediction system for Fine particulate Matter(PM2.5) based on the deep learning algorithm GAN model. The experimental data are closely related to the changes in temperature, humidity, wind speed, and atmospheric pressure generated by the time series axis and the concentration of air pollutants such as SO2, CO, O3, NO2, and PM10. Due to the characteristics of the data, since the concentration at the current time is affected by the concentration at the previous time, a predictive model for recursive supervised learning was applied. For comparative analysis of the accuracy of the existing models, CNN and LSTM, the difference between observation value and prediction value was analyzed and visualized. As a result of performance analysis, it was confirmed that the proposed GAN improved to 15.8%, 10.9%, and 5.5% in the evaluation items RMSE, MAPE, and IOA compared to LSTM, respectively.

본 연구는 딥러닝(Deep Learning) 알고리즘 GAN 모델을 기반으로 초미세먼지(PM2.5) 인공지능 예측시스템을 개발한다. 실험 데이터는 시계열 축으로 생성된 온도, 습도, 풍속, 기압의 기상변화와 SO2, CO, O3, NO2, PM10와 같은 대기오염물질 농도와 밀접한 관련이 있다. 데이터 특성상, 현재시간 농도가 이전시간 농도에 영향을 받기 때문에 반복지도학습(Recursive Supervised Learning) 예측 모델을 적용하였다. 기존 모델인 CNN, LSTM의 정확도(Accuracy)를 비교분석을 위해 관측값(Observation Value)과 예측값(Prediction Value)간의 차이를 분석하고 시각화했다. 성능분석 결과 제안하는 GAN이 LSTM 대비 평가항목 RMSE, MAPE, IOA에서 각각 15.8%, 10.9%, 5.5%로 향상된 것을 확인하였다.

Keywords

References

  1. H. G. Yoo et al. (2020). Impact of Meteorological Conditions on the PM2.5 and PM10 concentrations in Seoul. Journal of Climate Change Research, 11(5-2), 521-528. DOI : 10.15531/KSCCR.2020.11.5.521
  2. S. H. Lee et al. (2019, 4). Characteristics of PM2.5 in Gwangju Evaluated by Factor Analysis. Journal of Environmental Science International, 28(4), 413-422 DOI : 10.5322/JESI.2019.28.4.413
  3. S. J. Oh, J. W. Koo & U. M. Kim. (2017). Concentration Prediction Technique Based on Locality of Fine Dust Generation. The Institute of Electronics and Information Engineers, 1357-1360
  4. A. Chaloulakou, G. Grivas & N. Spyrellis. (2003). Neural Network and multiple regression model for PM10 prediction in Athens: A comparative assessment. Journal of the Air & Waste Management Association, 53(10), 1183-1190. https://doi.org/10.1080/10473289.2003.10466276
  5. M. M. Dedovic, S. Avadakovic, I. Turkovic, N. Dautbasic & T. Konjic. (2016). Forecasting PM10 concentrations using neural networks and system for improving air quality. Proceeding of 2016 XI International Symposium on Telecommunications, 1-6.
  6. D. J. Lim, T. H. Kim, R. Lee & H. M. Jung. (2017). LSTM-based Particulate Matter prediction for efficient road scattering dust removal path proposal. Korea Information Processing Society, 24(2), 1258-1261.
  7. K. P. Ra, M. C. Kim, M. J. Kim, S. T. Lim & Y. G. Sim. (2019). A Study on The Prediction of The Fine-Dust Concentration Using RNN/LSTM. The Institute of Electronics and Information Engineers, 1400-1405.
  8. K. W. Cho, Y. J. Jung, C. G. Kang & C. H. Oh. (2019). Conformity Assessment of Machine Learning Algorithm for Particulate Matter Prediction. The Korea Institute of Information and Communication Engineering, 23(1), 20-26.
  9. I. H. Shin, Y. H. Moon & Y. J. Lee. (2019). Deep Learning Models for Fine Dust Prediction in Smart Cities. Journal of Computing Science and Engineering, 397-399
  10. Y. Bengio, P. Simard & P. Frasconi. (1994). Learning long-term dependencies with gradient descent is difficult. Neural Networks, IEEE Transactions on, 5(2), 157-166. https://doi.org/10.1109/72.279181
  11. J. Y. Choi, D. H. Lee, J. Y. Kim & K. M. Jung. (2019), Air pollution prediction using deep learning based model. Journal of Computing Science and Engineering. 859-861.
  12. C. J. Huang, P. H. Kuo. (2018). A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities. Sensors, 18(7):2220 https://doi.org/10.3390/s18072220
  13. K. Zhang, G. Zhong, J. Dong, S. Wang & Y. Wang. (2019). Stock Market Prediction Based on Generative Adversarial Network. Procedia Computer Science, 147. 400-406. DOI:10.1016/j.procs.2019.01.256.
  14. K. W. Cho, Y. J. Jung, J. S. Lee & C. H. Oh. (2019). PM10 Particulate Matters Concentration Prediction using LSTM. The Korea Institute of Information and Communication Engineering, 23(2), 632-634
  15. I. Goodfellow et al. (2014). Generative adversarial nets. Advances in Neural Information Processing Systems 27, Montreal, Quebec, Canada, 2672-2680.
  16. Y. J. Lee, K. H. Seok. (2018). A study on the performance of generative adversarial networks. The Korean Data and Information Science Society, 29(5), 1155-1167. https://doi.org/10.7465/jkdi.2018.29.5.1155
  17. I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin & A. Courville, (2017). Imporved Training of Wasserstein GANs. arXiv preprint arXiv:1704.00028.