DOI QR코드

DOI QR Code

심층신경망을 이용한 레이더 영상 학습 기반 초단시간 강우예측

Very short-term rainfall prediction based on radar image learning using deep neural network

  • 윤성심 (한국건설기술연구원 국토보전연구본부) ;
  • 박희성 (한국건설기술연구원 국토보전연구본부) ;
  • 신홍준 (한국수력원자력(주) 중앙연구원)
  • Yoon, Seongsim (Korea Institute of Civil Engineering and Building Technology) ;
  • Park, Heeseong (Korea Institute of Civil Engineering and Building Technology) ;
  • Shin, Hongjoon (Central Research Institute, Korea-Hydro & Nuclear Power)
  • 투고 : 2020.10.21
  • 심사 : 2020.11.06
  • 발행 : 2020.12.31

초록

본 연구에서는 강우예측을 위해 U-Net과 SegNet에 기반한 합성곱 신경망 네트워크 구조에 장기간의 국내 기상레이더 자료를 활용하여 심층학습기반의 강우예측을 수행하였다. 또한, 기존 외삽기반의 강우예측 기법인 이류모델의 결과와 비교 평가하였다. 심층신경망의 학습 및 검정을 위해 2010부터 2016년 동안의 기상청 관악산과 광덕산 레이더의 원자료를 수집, 1 km 공간해상도를 갖는 480 × 480의 픽셀의 회색조 영상으로 변환하여 HDF5 형태의 데이터를 구축하였다. 구축된 데이터로 30분 전부터 현재까지 10분 간격의 연속된 레이더 영상 4개를 이용하여 10분 후의 강수량을 예측하도록 심층신경망 모델을 학습하였으며, 학습된 심층신경망 모델로 60분의 선행예측을 수행하기 위해 예측값을 반복 사용하는 재귀적 방식을 적용하였다. 심층신경망 예측모델의 성능 평가를 위해 2017년에 발생한 24개의 호우사례에 대해 선행 60분까지 강우예측을 수행하였다. 임계강우강도 0.1, 1, 5 mm/hr에서 평균절대오차와 임계성공지수를 산정하여 예측성능을 평가한 결과, 강우강도 임계 값 0.1, 1 mm/hr의 경우 MAE는 60분 선행예측까지, CSI는 선행예측 50분까지 참조 예측모델인 이류모델이 보다 우수한 성능을 보였다. 특히, 5 mm/hr 이하의 약한 강우에 대해서는 심층신경망 예측모델이 이류모델보다 대체적으로 좋은 성능을 보였지만, 5 mm/hr의 임계 값에 대한 평가결과 심층신경망 예측모델은 고강도의 뚜렷한 강수 특징을 예측하는 데 한계가 있었다. 심층신경망 예측모델은 예측시간이 길어질수록 공간 평활화되는 경향이 뚜렷해지며, 이로 인해 강우 예측의 정확도가 저하되었다. 이류모델은 뚜렷한 강수 특성을 보존하기 때문에 강한 강도 (>5 mm/hr)에 대해 심층신경망 예측모델을 능가하지만, 강우 위치가 잘못 이동하는 경향이 있다. 본 연구결과는 이후 심층신경망을 이용한 레이더 강우 예측기술의 개발과 개선에 도움이 될 수 있을 것으로 판단된다. 또한, 본 연구에서 구축한 대용량 기상레이더 자료는 향후 후속연구에 활용될 수 있도록 개방형 저장소를 통해 제공될 예정이다.

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.

키워드

과제정보

본 논문은 한국수력원자력(주)에서 재원을 부담하여 한국건설기술연구원에서 수행한 연구결과입니다(No. 2018-기술-20).

참고문헌

  1. Agrawal, S., Barrington, L., Bromberg, C., Burge, J., Gazen, C., and Hickey, J. (2019). Machine learning for precipitation nowcasting from radar images. accessed 08 December 2020, .
  2. Ayzel, G., Scheffer, T., and Heistermann, M. (2020). "RainNet v1.0: A convolutional neural network for radar-based precipitation nowcasting." Geoscientific Model Development, Vol. 13, pp. 2631-2644. https://doi.org/10.5194/gmd-13-2631-2020
  3. Badrinarayanan, V., Kendall, A., and Cipolla, R. (2017). "SegNet: A deep convolutional encoder-decoder architecture for image segmentation." IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, pp. 2481-2495. https://doi.org/10.1109/TPAMI.2016.2644615
  4. Bellon, A., and Austin, G.L. (1978). "The evaluation of two years of real-time operation of a short-term precipitation forecasting procedure (SHARP)." Journal of Applied Meteorology and Climatology, Vol. 17, pp. 1778-1787. https://doi.org/10.1175/1520-0450(1978)017<1778:TEOTYO>2.0.CO;2
  5. Dahl, G.E., Sainath, T.N., and Hinton, G.E. (2013). "Improving deep neutral networks for LVCSR using rectified linear units and dropout." Proceedings 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, Vancouver, Canada, pp. 8609-8613.
  6. Dixon, M., and Wiener, G. (1993). "TITAN: Thunderstorm identification, tracking, analysis, and nowcasting-a radar-based methodology." Journal of Atmospheric and Oceanic Technology, Vol. 10, pp. 785-797. https://doi.org/10.1175/1520-0426(1993)010<0785:TTITAA>2.0.CO;2
  7. Handwerker, J. (2002). "Cell tracking with TRACE3D - a new algorithm." Atmospheric Research, Vol. 61, pp. 15-34. https://doi.org/10.1016/S0169-8095(01)00100-4
  8. Hilst, G.R., and Russo, J.A. (1960). An objective extrapolation technique for semiconservative fields with an application to radar patterns. Tech. Memo. No. 3, Travelers Weather Research Center, Hartford, CT, U.S.
  9. Iglovikov, V., and Shvets, A. (2018). TernausNet: U-Net with VGG11 Encoder pre-trained on imagenet for image segmentation. accessed 08 December 2020, .
  10. Johnson, J.T., MacKeen, P.L., Witt, A., DeWayne Mitchell, E., Stumpf, G.J., Eilts, M.D., and Thomas, K.W. (1998). "The storm cell identification and tracking algorithm: An enhanced WSR88D algorithm." Weather and Forecasting, Vol. 13, pp. 263-276. https://doi.org/10.1175/1520-0434(1998)013<0263:TSCIAT>2.0.CO;2
  11. Kessler, E., and Russo, J.H. (1963). "A program for the assembly and display of radar-echo distributions." Journal of Applied Meteorology and Climatology, Vol. 2, pp. 582-593. https://doi.org/10.1175/1520-0450(1963)002<0582:APFTAA>2.0.CO;2
  12. Kim, G.S., and Kim, J.P. (2008). "Development of a short-term rainfall forecasting model using weather radar data." Journal of Korea Water Resources Association, Vol. 41, No. 10, pp. 1023-1034. https://doi.org/10.3741/JKWRA.2008.41.10.1023
  13. Kingma, D.P., and Ba, J. (2015). "Adam: A method for stochastic optimization." Proceedings 3rd International Conference on Learning Representations. ICLR 2015, San Diego, CA, U.S.
  14. Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012). "ImageNet classification with deep convolutional neural networks." Advances in neural information processing systems, Vol. 25, No. 2, doi: 10.1145/3065386.
  15. Kuligowski, R.J., and Barros, A.P. (1998). "Localized precipitation forecasts from a numerical weather prediction model using artificial neural networks." Weather and Forecasting, Vol. 13, No. 4, pp. 1194-1204. https://doi.org/10.1175/1520-0434(1998)013<1194:LPFFAN>2.0.CO;2
  16. LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998) "Gradient-based learning applied to document recognition." Proceedings of the IEEE, Vol. 86, No. 11, pp. 2278-2324. https://doi.org/10.1109/5.726791
  17. Lee, S., Cho, S., and Wong, P.M. (1998), "Rainfall prediction using artificial neural networks." Journal of Geographic Information and Decision Analysis, Vol. 2, No. 2, pp. 233-242.
  18. Lin, C., Vasic, S., Kilambi, A., Turner, B., and Zawadzki, I. (2005). "Precipitation forecast skill of numerical weather prediction models and radar nowcasts." Geophysical Research Letters, Vol. 32, p. L14801, doi: 10.1029/2005GL023451.
  19. Marshall, J.S., and Palmer, W.M. (1948). "The distribution of raindrops with size." Journal of Meteorology, Vol. 5, pp. 165-166. https://doi.org/10.1175/1520-0469(1948)005<0165:TDORWS>2.0.CO;2
  20. Nair, V., and Hinton, G.E. (2010). "Interpersonal informatics: Making social influence visible." Proceedings of the 27th International Conference on Machine Learning, Omnipress, Haifa, Israel, pp. 807-814.
  21. Nakakita, E., Ikebuchi, S., Nakamura, T., Kanmuri, M., Okuda, M., Yamaji, A., and Takasao T. (1996). "Short-term rainfall prediction method using a volume scanning radar and GPV data from numerical weather prediction." Journal of Geophysical Research, Vol. 101, No. D21, pp. 26181-26197. https://doi.org/10.1029/96JD01615
  22. Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., and Prabhat. (2019). "Deep learning and process understanding for data-driven Earth system science." Nature, 566, pp. 195-204, doi: 10.1038/s41586-019-0912-1.
  23. Reyniers, M. (2008). "Quantitative precipitation forecasts based on radar observations: Principles, algorithms and operational systems." Royal Meteorological Institute, Belgium.
  24. Rinehart, R.E., and Garvey, E.T. (1978). "Three-dimensional storm motion detection by conventional weather radar." Nature, Vol. 273, pp. 287-289. https://doi.org/10.1038/273287a0
  25. Ronneberger, O., Fischer, P., and Brox, T. (2015). "U-Net: Convolutional networks for biomedical image segmentation." Medical Image Computing and Computer-Assisted Intervention MICCAI 2015, Springer International Publishing, Lecture Notes in Computer Science, doi: 10.1007/978-3-319-24574-4_28.
  26. Seo, J.H., Lee, Y.H., and Kim, Y.H. (2012), "Feature selection to predict very short-term heavy rainfall based on differential evolution." Journal of Korean Institute of Intelligent Systems, Vol. 22, No. 6, pp. 706-714. https://doi.org/10.5391/JKIIS.2012.22.6.706
  27. Shi, E., Li, Q., Gu, D., and Zhao, Z. (2018). "A method of weather radar echo extrapolation based on convolutional neural networks, in: Multimedia modeling." Springer International Publishing, Lecture Notes in Computer Science, pp.16-28, https://doi.org/10.1007/978-3-319-73603-7_2.
  28. Shi, X., Chen, Z., Wang, H., Yeung, D., Wong, W., and Woo, W. (2015). Convolutional LSTM network: A machine learning approach for precipitation nowcasting, accessed 8 December 2020, .
  29. Shi, X., Gao, Z., Lausen, L., Wang, H., Yeung D., Wong, W., and Woo, W. (2017) "Deep learning for precipitation nowcasting: A benchmark and a new model." 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, U.S.
  30. Shiiba, M., Takasao, T., and Nakakita, E. (1984). "Investigation of short-term rainfall prediction method by a translation model." Proceeding 28th Japanese Conference on Hydraulics, JSCE, pp. 423-428.
  31. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R. (2014). "Dropout: A simple way to prevent neural networks from overfitting." The Journal of Machine Learning Research, Vol. 15, No. 1, pp. 1929-1958.
  32. Srivastava, R.K., Greff, K., and Schmidhuber, J. (2015). "Training very deep networks." Advances in Neural Information Processing Systems, Curran Associates, Inc., Red Hook, NY, U.S., pp. 2377-2385.
  33. Sugimoto, S., Nakakita E., and Ikebuchi, S. (2001). "A stochastic approach to short-term rainfall prediction using a physically based conceptual rainfall model." Journal of Hydrology, Vol. 242, pp. 137-155. https://doi.org/10.1016/S0022-1694(00)00390-5
  34. Sun, J., Xue, M., Wilson, J.W., Zawadzki, I., Ballard, S.P., Onvlee Hooimeyer, J., Joe, P., Barker, D.M., Li, P.-W., Golding, B., Xu, M., and Pinto, J. (2014). "Use of NWP for nowcasting convective precipitation: Recent progress and challenges." Bulletin of the American Meteorological Society, Vol. 95, pp. 409-426, doi: 10.1175/BAMS-D-11-00263.1.
  35. Sutskever, I., Vinyals, O., and Le, Q.V. (2014). "Sequence to sequence learning with neural networks." Advances in Neural Information Processing Systems 27, Curran Associates, Inc., Red Hook, NY, U.S., pp. 3104-3112.
  36. Tran, Q.K., and Song, S.K. (2019). "Computer vision in precipitation nowcasting: Applying image quality assessment metrics for training deep neural networks." Atmosphere, Vol. 10, p. 244. https://doi.org/10.3390/atmos10050244
  37. Tuttle, J., and Gall, R. (1999). "A single-radar technique for estimating the winds in tropical cyclones." Bulletin of the American Meteorology Society, Vol. 80, pp. 653-668. https://doi.org/10.1175/1520-0477(1999)080<0653:ASRTFE>2.0.CO;2
  38. Tuttle, J.D., and Foote, G.B. (1990). "Determination of boundary layer airflow from a single doppler radar." Journal of Atmospheric and Oceanic Technology, Vol. 7, pp. 218-232. https://doi.org/10.1175/1520-0426(1990)007<0218:DOTBLA>2.0.CO;2
  39. Xu, H., and Ge, D. (2020). "A novel image edge smoothing method based on convolutional neural network." International Journal of Advanced Robotic Systems, SAGE journals, Vol.17, No. 3, pp.1-11, doi: 10.1177/1729881420921676.
  40. Yoon, S.S. (2017). "Development of radar-based quantitative precipitation forecasting using spatial-scale decomposition method for urban flood management." Journal of Korea Water Resources Association, Vol. 50, pp. 335-346. https://doi.org/10.3741/JKWRA.2017.50.5.335
  41. Yoon, S.S. (2019). "Adaptive blending method of radar-based and numerical weather prediction QPFs for urban flood forecasting." Remote Sensing, Vol. 11, p. 642. https://doi.org/10.3390/rs11060642
  42. Yoon, S.S., and Bae, D.H. (2010). "The applicability assesment of the short-term rainfall forecasting using translation model." Journal of Korea Water Resources Association, Vol. 43, No. 8, pp. 695-707. https://doi.org/10.3741/JKWRA.2010.43.8.695