DOI QR코드

DOI QR Code

Forecasting the Precipitation of the Next Day Using Deep Learning

딥러닝 기법을 이용한 내일강수 예측

  • Ha, Ji-Hun (Department of Embedded Software Engineering, Kwangwoon University) ;
  • Lee, Yong Hee (National Institute of Meteorological Science) ;
  • Kim, Yong-Hyuk (Department of Computer Science and Engineering, Kwangwoon University)
  • 하지훈 (광운대학교 임베디드SW공학과) ;
  • 이용희 (국립기상과학원) ;
  • 김용혁 (광운대학교 컴퓨터과학과)
  • Received : 2015.11.16
  • Accepted : 2015.12.28
  • Published : 2016.04.25

Abstract

For accurate precipitation forecasts the choice of weather factors and prediction method is very important. Recently, machine learning has been widely used for forecasting precipitation, and artificial neural network, one of machine learning techniques, showed good performance. In this paper, we suggest a new method for forecasting precipitation using DBN, one of deep learning techniques. DBN has an advantage that initial weights are set by unsupervised learning, so this compensates for the defects of artificial neural networks. We used past precipitation, temperature, and the parameters of the sun and moon's motion as features for forecasting precipitation. The dataset consists of observation data which had been measured for 40 years from AWS in Seoul. Experiments were based on 8-fold cross validation. As a result of estimation, we got probabilities of test dataset, so threshold was used for the decision of precipitation. CSI and Bias were used for indicating the precision of precipitation. Our experimental results showed that DBN performed better than MLP.

정확한 강수예측을 위해서는 예측인자 선정과 예측방법에 대한 선택이 매우 중요하다. 최근에는 강수예측 방법으로 기계학습 기법이 많이 사용되고 있으며, 그 중에서도 특히 인공신경망을 사용한 강수예측 방법은 좋은 성능을 보였다. 본 논문에서는 딥러닝 기법 중 하나인 DBN(deep belief network)를 이용한 새로운 강수예측 방법을 제안한다. DBN는 비지도 사전 학습을 통해 초기 가중치를 설정하여 기존 인공신경망의 문제점을 보완한다. 예측인자로는 기온, 전일-전주 강수일, 태양과 달 궤도 관련 자료를 선정하였다. 기온과 전일-전주 강수일은 서울에서의 1974년부터 2013년까지 총 40년간의 AWS(automatic weather system) 관측 자료를 사용하였고, 태양과 달의 궤도 관련 자료는 서울을 중심으로 계산한 결과를 사용하였다. 전체 기간에서 일부는 학습 자료로 사용하여 예측모델을 생성하였고, 나머지를 생성한 모델의 검증 자료로 사용하였다. 모델 검증 결과로 나온 예측값들은 확률값을 가지며 임계치를 이용하여 강수유무를 판별하였다. 강수 정확도의 척도로 양분예보기법 중 CSI(critical successive index)와 Bias(frequency bias)를 계산하였다. 이를 통해 DBN와 MLP(multilayer perceptron)의 성능을 비교한 결과 DBN의 강수 예측 정확도가 높았고, 수행속도 또한 2배 이상 빨랐다.

Keywords

References

  1. J.-H. Seo, Y. H. Lee, Y.-H. Kim, "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, 2012 https://doi.org/10.5391/JKIIS.2012.22.6.706
  2. R. J. Kuligowski, A. P. Barros, "Localized precipitation forecasts from a numerical weather prediction model using artificial neural networks", Weather and Forecasting, vol. 13, no. 4, pp. 1194-1204, 1998 https://doi.org/10.1175/1520-0434(1998)013<1194:LPFFAN>2.0.CO;2
  3. S. Lee, S. Cho, P. M. Wong, "Rainfall prediction using artificial neural networks", Journal of Geographic Information and Decision Analysis, vol. 2, no. 2, pp. 233-242, 1998
  4. T. Hall, H. E. Brooks, C. A. Doswell III, "Precipitation forcasting using a neural network", Weather and Forecasting, vol. 14, no. 3, pp. 338-345, 1999 https://doi.org/10.1175/1520-0434(1999)014<0338:PFUANN>2.0.CO;2
  5. K. C. Luk, J. E. Ball, A. Sharma, "An application of artificial neural networks for rainfall forecasting", Mathematical and Computer Modelling, vol. 33, no. 6-7, pp. 683-693, 2001 https://doi.org/10.1016/S0895-7177(00)00272-7
  6. M. C. V. Ramirez, H. F. C. Velho, N. J. Ferreira, "Artificial neural network technique for rainfall forecasting applied to the Sao Paulo region", Journal of Hydrology, vol. 301, nos. 1-4, pp. 146-162, 2005 https://doi.org/10.1016/j.jhydrol.2004.06.028
  7. H.-K. Kim, W.-Y. Choi, S.-K. Oh, "Design of summer very short-term precipitation forecasting pattern in metropolitan area using optimized RBFNNs", Journal of Korean Institute of Intelligent Systems, vol. 23, no. 6, pp. 533-538, 2013 https://doi.org/10.5391/JKIIS.2013.23.6.533
  8. H.-M. Kim, S.-K. Oh, Y.-H. Lee, "Design of heavy rain advisory decision model based on optimized RBFNNs using KLAPS reanalysis data", Journal of Korean Institute of Intelligent Systems, vol. 23, no. 5, pp. 473-478, 2013 https://doi.org/10.5391/JKIIS.2013.23.5.473
  9. G. E. Hinton, S. Osindero, Y.-W. Teh, "A fast learning algorithm for deep belief nets", Neural Computation, vol. 18, no. 7, pp. 1527-1554, 2006 https://doi.org/10.1162/neco.2006.18.7.1527
  10. A. Dai, K. E. Trenberth, T. R. Karl, "Effects of clouds, soil moisture, precipitation, and water vapor on diurnal temperature range", Journal of Cliamate, vol. 12, no. 8, pp. 2451-2473, 1999 https://doi.org/10.1175/1520-0442(1999)012<2451:EOCSMP>2.0.CO;2
  11. S. Samarasinghe, Neural Networks for Applied Sciences and Engineering: From Fundamentals to Complex Pattern Recognition, CRC Press, 2006
  12. I. T. Jolliffe, D. B. Stephenson, Forecast Verification: a Practitioner's Guide in Atmospheric Science, John Wiley & Sons, 2012
  13. R. J. Donaldson, R. M. Dyer, M. J. Kraus, "Objective evaluator of technique for predicting severe weather events", Bulletin of the American Meteorological Society, vol 56, no. 7, 1975
  14. P. J. Roebber, "Visualizing multiple measures of forecast quality", Weather and Forecasting, vol. 24, no. 2, pp. 601-608, 2009 https://doi.org/10.1175/2008WAF2222159.1

Cited by

  1. A Method for Correcting Air-Pressure Data Collected by Mini-AWS vol.26, pp.3, 2016, https://doi.org/10.5391/JKIIS.2016.26.3.182
  2. Study of Precipitation Forecast Based on Deep Belief Networks vol.11, pp.9, 2018, https://doi.org/10.3390/a11090132