Journal of Korea Water Resources Association (한국수자원학회논문집)

Korea Water Resources Association (한국수자원학회)
권호 표지
DOI QR코드

DOI QR Code

Application of recurrent neural network for inflow prediction into multi-purpose dam basin

다목적댐 유입량 예측을 위한 Recurrent Neural Network 모형의 적용 및 평가

  • Park, Myung Ky (Water Data Collection and Analysis Department, K-water) ;
  • Yoon, Yung Suk (Water Data Collection and Analysis Department, K-water) ;
  • Lee, Hyun Ho (Water Data Collection and Analysis Department, K-water) ;
  • Kim, Ju Hwan (K-water Convergence Institute, K-water)
  • 박명기 (한국수자원공사 물정보종합센터) ;
  • 윤영석 (한국수자원공사 물정보종합센터) ;
  • 이현호 (한국수자원공사 물정보종합센터) ;
  • 김주환 (한국수자원공사 K-water융합연구원)
  • Received : 2018.10.05
  • Accepted : 2018.10.23
  • Published : 2018.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

This paper aims to evaluate the applicability of dam inflow prediction model using recurrent neural network theory. To achieve this goal, the Artificial Neural Network (ANN) model and the Elman Recurrent Neural Network(RNN) model were applied to hydro-meteorological data sets for the Soyanggang dam and the Chungju dam basin during dam operation period. For the model training, inflow, rainfall, temperature, sunshine duration, wind speed were used as input data and daily inflow of dam for 10 days were used for output data. The verification was carried out through dam inflow prediction between July, 2016 and June, 2018. The results showed that there was no significant difference in prediction performance between ANN model and the Elman RNN model in the Soyanggang dam basin but the prediction results of the Elman RNN model are comparatively superior to those of the ANN model in the Chungju dam basin. Consequently, the Elman RNN prediction performance is expected to be similar to or better than the ANN model. The prediction performance of Elman RNN was notable during the low dam inflow period. The performance of the multiple hidden layer structure of Elman RNN looks more effective in prediction than that of a single hidden layer structure.

본 연구에서는 순환신경망을 이용한 댐 유입량 예측모형의 적용성 검토를 목적으로 하고 있으며, 이를 위해 소양강댐 유역 및 충주댐 유역을 대상으로 그간 댐 운영을 통해 축적된 기상 및 수문 빅데이터를 활용하여 인공신경망 모형과 엘만 순환신경망 모형을 구축하였다. 모형의 학습과 예측을 위하여 유역별 유입량, 강우량, 기온, 일조시간, 풍속자료가 입력자료로 사용되었고 10일간 일별 댐유입량 자료가 모델의 출력자료로 구조화 하여 학습을 진행한 후 검증을 목적으로 2016년 7월 ~ 2018년 6월까지 2개년에 대한 댐 유입량 예측을 수행하였다. 학습된 모형의 유입량 예측 결과를 비교분석한 결과, 소양강댐 유역에서는 인공신경망 모형과 순환신경망 모형 간 예측성능은 큰 차이를 보이지 않았으며, 충주댐 유역에서는 순환신경망 모형의 예측 결과가 인공신경망 모형에 비해 비교적 우수한 성능을 보임에 따라 엘만 순환신경망을 이용하여 댐 유입량 예측모형을 구축할 경우 예측성능은 기존의 인공신경망 모형과 비슷하거나 다소 우수할 것으로 판단된다. 또한 엘만 순환신경망은 갈수기 댐 유입량 예측에 있어서 인공신경망에 비해 예측결과의 재현성이 우수한 것으로 나타났으며, 엘만 순환신경망 학습에 있어 다중 은닉층 구조가 단일 은닉층 구조보다 예측 성능 향상에 효과적인 것으로 분석되었다.

Keywords

SJOHCI_2018_v51n12_1217_f0001.png 이미지

Fig. 1. Comparison of artificial neural network (a) and elman recurrent neural network (b)

SJOHCI_2018_v51n12_1217_f0002.png 이미지

Fig. 2. Location of the study area and the observation station

SJOHCI_2018_v51n12_1217_f0003.png 이미지

Fig. 3. Comparison of predicted and observed daily Soyanggang dam inflow of 2017 and 2018

SJOHCI_2018_v51n12_1217_f0004.png 이미지

Fig. 4. Comparison of predicted and observed daily Chungju dam inflow of 2017 and 2018

SJOHCI_2018_v51n12_1217_f0005.png 이미지

Fig. 5. Comparison of cumulative distribution between predicted and observed inflow

SJOHCI_2018_v51n12_1217_f0006.png 이미지

Fig. 6. Comparison of standardized residuals for different inflow regimes (Soyanggang)

SJOHCI_2018_v51n12_1217_f0007.png 이미지

Fig. 7. Comparison of standardised residuals for different inflow regimes (Chungju)

SJOHCI_2018_v51n12_1217_f0008.png 이미지

Fig. 8. Comparison of performance for number of hidden layers (Soyanggang)

SJOHCI_2018_v51n12_1217_f0009.png 이미지

Fig. 9. Comparison of performance for number of hidden layers (Chungju)

SJOHCI_2018_v51n12_1217_f0010.png 이미지

Fig. 10. Comparison of performance for number of hidden layer PEs (Soyanggang)

SJOHCI_2018_v51n12_1217_f0011.png 이미지

Fig. 11. Comparison of performance for number of hidden layer PEs (Chungju)

Table 1. Information of the Input data set

SJOHCI_2018_v51n12_1217_t0001.png 이미지

Table 2. Period of training and validation data set

SJOHCI_2018_v51n12_1217_t0002.png 이미지

Table 3. Hyper-parameter of artificial neural network and elman recurrent neural network

SJOHCI_2018_v51n12_1217_t0003.png 이미지

Table 4. Top 10 models prediction performance results (Soyanggang)

SJOHCI_2018_v51n12_1217_t0004.png 이미지

Table 5. Top 10 models prediction performance results (Chungju)

SJOHCI_2018_v51n12_1217_t0005.png 이미지

Table 6. Daily best model prediction performance results (Soyanggang)

SJOHCI_2018_v51n12_1217_t0006.png 이미지

Table 7. Daily best model prediction performance results (Chungju)

SJOHCI_2018_v51n12_1217_t0007.png 이미지

Table 8. Quantiles of dam inflow

SJOHCI_2018_v51n12_1217_t0008.png 이미지

Table 9. Peaks of standardized residuals for different inflow regimes

SJOHCI_2018_v51n12_1217_t0009.png 이미지

Table 10. Best model prediction performance results (RMSE) of each number of hidden layers

SJOHCI_2018_v51n12_1217_t0010.png 이미지

Table 11. Best model prediction performance results (RMSE) of each number of hidden layer PEs

SJOHCI_2018_v51n12_1217_t0011.png 이미지

References

  1. Bergmeir, C., and Benitez, J.M. (2012). "Neural Networks in R Using the Stuttgart Neural Network Simulator: RSNNS." Journal of Statistical Software, Vol. 46, No. 7, pp. 1-26.
  2. Chen, P.-A., Chang, L.-C., and Chang, F.-J. (2013). "Reinforced recurrent neural networks for multi-step-ahead flood forecasts." Journal of Hydrology, Vol. 497, pp. 71-79. https://doi.org/10.1016/j.jhydrol.2013.05.038
  3. Chiang, Y.-M., Hsu, K.-L., Chang, F.-J., Hong, Y., and Sorooshian, S. (2007). "Merging multiple precipitation sources for flash flood forecasting." Journal of Hydrology, Vol. 340, No. 3-4, pp. 183-196. https://doi.org/10.1016/j.jhydrol.2007.04.007
  4. Coulibaly P., and Anctil F. (1999). "Real-time short-term natural water inflows forecasting using recurrent neural networks." International Joint Conference On Neural Networks, Washington Dc, Proceedings Ijcnn'99, Vol. 6, pp. 3802-3805.
  5. Coulibaly, P., and Baldwin, C. K. (2005). "Nonstationary hydrological time series forecasting using nonlinear dynamic methods." Journal of Hydrology, Vol. 307, No. 1, pp. 164-174. https://doi.org/10.1016/j.jhydrol.2004.10.008
  6. Elman, Jeffrey L. (1990). "Finding Structure in Time." Cognitive Science, Vol. 14, No. 2, pp. 179-211. https://doi.org/10.1207/s15516709cog1402_1
  7. Hsu, K., Gupta, H. V., and Sorooshian, S. (1995). "Artificial Neural Network Modeling of the Rainfall-Runoff Process." Water Resources Research, Vol. 31, No. 10, pp. 2517-2530. https://doi.org/10.1029/95WR01955
  8. Jung, S. H., Lee, D. E., and Lee, K. S. (2018). "Prediction of River Water Level Using Deep-Learning Open Library." Journal of the Korean Society of Hazard Mitigation, Vol. 18, No. 1, pp. 1-11.
  9. Jun, H. D., and Lee, J. H. (2013). "A Methodology for Flood Forecasting and Warning Based on the Characteristic of Observed Water Levels Between Upstream and Downstream." Journal of the Korean Society of Hazard Mitigation, Vol. 13, No. 6, pp. 367-374. https://doi.org/10.9798/KOSHAM.2013.13.6.367
  10. Kim, J. H., Park, C. Y., and Kang K. W. (1992). "Nonlinear Prediction of Stream flows by Pattern Recognition Method." Journal of Korean Water Resources Association, Vol. 25, No. 3, pp. 105-113.
  11. Kim, J. H. (1993). A Study on Hydrologic Forecasting of Stream flows Based on Artificial Neural Network. Ph. D. dessertation, Inha University.
  12. Kwon, S. H., Lee, J. W., and Chung, G. H. (2017). "Snow Damages Estimation using Artificial Neural Network and Multiple Regression Analysis." Journal of the Korean Society of Hazard Mitigation, Vol. 17, No. 2, pp. 315-325. https://doi.org/10.9798/KOSHAM.2017.17.2.315
  13. Lee, C. Y., and Kim, J. H. (2018), "The Prediction and Analysis of the Power Energy Time Series by Using the Elman Recurrent Neural Network." Journal of Society of Korea Industrial and Systems Engineering, Vol. 41, No. 1, pp. 84-93.
  14. Lee, K. H., Jung, S. H., and Lee, D. E. (2018). "Comparison of physics-based and data-driven models for streamflow simulation of the Mekong river." Journal of Korea Water Resources Association, Vol. 51, No. 6, pp. 503-513. https://doi.org/10.3741/JKWRA.2018.51.6.503
  15. Maier, H. R., and Dandy, G. C. (2000). "Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications." Environmental Modelling & Software, Vol. 15, No. 1, pp. 101-124. https://doi.org/10.1016/S1364-8152(99)00007-9
  16. McCulloch, W.S., and Pitts, W. (1943). "A Logical Calculusof the Ideas Immanent in Nervous Activity." The Bulletin of Mathematical Biophysics, Vol. 5, No. 4, pp. 115-133. https://doi.org/10.1007/BF02478259
  17. Rong, X. (2014). deepnet: deep learning toolkit in R, accessed 1 May 2018, .
  18. Rosenblatt, F. (1958). "The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain." Psychological Review, Vol. 65, No. 6, pp. 386-408. https://doi.org/10.1037/h0042519
  19. Rumelhart, D. E., Hintont, G. E., and Williams, R. J. (1986). "Learning Representations by Back-propagating Errors." NATURE, Vol. 323, No. 9, pp. 533-536. https://doi.org/10.1038/323533a0
  20. Shoaib, M., Shamseldin, A. Y., Melville, B. W., and Khan, M. M. (2016). "A comparison between wavelet based static and dynamic neural network approaches for runoff prediction." Journal of Hydrology, Vol. 535, pp. 211-225. https://doi.org/10.1016/j.jhydrol.2016.01.076
  21. Yaseen, Z. M., El-shafie, A., Jaafar, O., Afan, H. A., and Sayl, K. N. (2015). "Artificial intelligence based models for stream-flow forecasting: 2000-2015." Journal of Hydrology, Vol. 530, pp. 829-844. https://doi.org/10.1016/j.jhydrol.2015.10.038
  22. Zhang, D., Holland, E. S., Lindholm, G., and Ratnaweera, H. (2017). "Hydraulic modeling and deep learning based flow forecasting for optimizing inter catchment wastewater transfer." Journal of Hydrology. Vol. In Press, Corrected Proof, .
  23. Zhang, D., Lindholm, G., and Ratnaweera, H. (2018). "Use long short-term memory to enhance Internet of Things for combined sewer overflow monitoring." Journal of Hydrology, Vol. 556, pp. 409-418. https://doi.org/10.1016/j.jhydrol.2017.11.018
  24. Zhang, J., Zhu, Y., Zhang, X., Ye, M., and Yang, J. (2018). "Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas." Journal of Hydrology, Vol. 561, pp. 918-929. https://doi.org/10.1016/j.jhydrol.2018.04.065