DOI QR코드

DOI QR Code

Development of Surface Weather Forecast Model by using LSTM Machine Learning Method

기계학습의 LSTM을 적용한 지상 기상변수 예측모델 개발

  • Hong, Sungjae (Department of Atmospheric Sciences, Pusan National University) ;
  • Kim, Jae Hwan (Department of Atmospheric Sciences, Pusan National University) ;
  • Choi, Dae Sung (Department of Atmospheric Sciences, Pusan National University) ;
  • Baek, Kanghyun (Research Center for Climate Sciences, Pusan National University)
  • 홍성재 (부산대학교 대기과학과) ;
  • 김재환 (부산대학교 대기과학과) ;
  • 최대성 (부산대학교 대기과학과) ;
  • 백강현 (부산대학교 기후연구센터)
  • Received : 2020.11.23
  • Accepted : 2021.03.02
  • Published : 2021.03.31

Abstract

Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.

Keywords

References

  1. Chen, J., J. Yu, M. Song, and V. Valdmanis, 2019: Factor decomposition and prediction of solar energy consumption in the United States. J. Clean. Prod., 234, 1210-1220, doi:10.1016/j.jclepro.2019.06.173.
  2. Farzad, A., H. Mashayekhi, and H. Hassanpour, 2019: A comparative performance analysis of different activation functions in LSTM networks for classification. Neural Comput. Applic., 31, 2507-2521, doi:10.1007/s00521-017-3210-6.
  3. Feng, C., M. Cui, B.-M. Hodge, and J. Zhang, 2017: A data-driven multi-model methodology with deep feature selection for short-term wind forecasting. Appl. Energ., 190, 1245-1257, doi:10.1016/j.apenergy.2017.01.043.
  4. Gal, Y., and Z. Ghahramani, 2016: A theoretically grounded application of dropout in recurrent neural network. Preprints, 30th Conference on Neural Information Processing Systems, 14 pp [Available online at https://arxiv.org/abs/1512.05287].
  5. Hewage, P., M. Trovati, E. Pereira, and A. Behera, 2021: Deep learning-based effective fine-grained weather forecasting model. Pattern Anal. Applic., 24, 343-366, doi: 10.1007/s10044-020-00898-1.
  6. Holton, J., and G. Hakim, 2012: An Introduction to Dynamic Meteorology, 5th edition. Academic Press, 552 pp.
  7. Hochreiter, S., and J. Schmidhuber, 1997: Long short-term memory. Neural Comput., 9, 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  8. Hong, S.-Y., and M. Kanamitsu, 2014: Dynamical downscaling: Fundamental issues from an NWP point of view and recommendations. Asia-Pac. J. Atmos. Sci., 50, 83-104, doi:10.1007/s13143-014-0029-2.
  9. Jozefowicz, R., W. Zaremba, and I. Sutskever, 2015: An empirical exploration of recurrent network architectures. Proc., The 32nd International conference on machine learning, 37, ICML, 2342-2350, doi:10.5555/3045118.3045367.
  10. Kiperwasser, E., and Y. Goldberg, 2016: Simple and accurate dependency parsing using bidirectional LSTM feature representations. T. Assoc. Comput. Linguist., 4, 313-327, doi:10.1162/tacl_a_00101.
  11. KMA, 2020: Validation of Numerical Prediction System (2019). Tech. Rep., Numerical Modeling Center, Korea Meteorological Administration, 11-1360709-000001-10, 297 pp (in Korean).
  12. Lea, C., M. D. Flynn, R. Vidal, A. Reiter, and G. D. Hager, 2017: Temporal convolutional networks for action segmentation and detection. Proc., 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Institute of Electrical and Electronics Engineers Inc., 1003-1012, doi:10.1109/CVPR.2017.113.
  13. Majhi, B., D. Naidu, A. P. Mishra, and S. C. Satapathy, 2020: Improved prediction of daily pan evaporation using Deep-LSTM model. Neural Comput. Applic., 32, 7823-7838, doi:10.1007/s00521-019-04127-7.
  14. Poornima, S., and M. Pushpalatha, 2019: Prediction of rainfall using intensified LSTM based recurrent neural network with weighted linear units. Atmosphere, 10, 668, doi:10.3390/atmos10110668.
  15. Rumelhart, D. E., G. E. Hinton, and R. J. Williams, 1986: Learning representations by back-propagating errors. Nature, 323, 533-536. https://doi.org/10.1038/323533a0
  16. Srivastava, N., G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, 2014: Dropout: a simple way to prevent neural networks from overfitting. J. Machine Learn. Res., 15, 1929-1958.
  17. Tieleman, T., and G. Hinton, 2012: Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude. COURSERA: Neural Networks for Machine Learning, 4, 26-31.
  18. Wang, H., and B. Raj, 2017: On the origin of deep learning. Preprint, arXiv, 72 pp [Available online at https://arxiv.org/abs/1702.07800].
  19. Yang, B., S. Sun, J. Li, X. Lin, and Y. Tian, 2019: Traffic flow prediction using LSTM with feature enhancement. Neurocomputing, 332, 320-327, doi:10.1016/j.neucom.2018.12.016.
  20. Zhang, Q., H. Wang, J. Dong, G. Zhong, and X. Sun, 2017: Prediction of sea surface temperature using long short-term memory. IEEE Geosci. Remote S., 14, 1745-1749, doi:10.1109/LGRS.2017.2733548.
  21. Zhang, T., S. Song, S. Li, L. Ma, S. Pan, and L. Han, 2019: Research on gas concentration prediction models based on LSTM multidimensional time series. Energies, 12, 161, doi:10.3390/en12010161.