Journal of Korea Water Resources Association (한국수자원학회논문집)

Korea Water Resources Association (한국수자원학회)
권호 표지
DOI QR코드

DOI QR Code

Prediction of multipurpose dam inflow using deep learning

딥러닝을 활용한 다목적댐 유입량 예측

  • Mok, Ji-Yoon (Department of Civil Engineering, University of Seoul) ;
  • Choi, Ji-Hyeok (Department of Civil Engineering, University of Seoul) ;
  • Moon, Young-Il (Department of Civil Engineering, University of Seoul)
  • 목지윤 (서울시립대학교 토목공학과) ;
  • 최지혁 (서울시립대학교 토목공학과) ;
  • 문영일 (서울시립대학교 토목공학과)
  • Received : 2019.08.08
  • Accepted : 2019.12.30
  • Published : 2020.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, Artificial Neural Network receives attention as a data prediction method. Among these, a Long Shot-term Memory (LSTM) model specialized for time-series data prediction was utilized as a prediction method of hydrological time series data. In this study, the LSTM model was constructed utilizing deep running open source library TensorFlow which provided by Google, to predict inflows of multipurpose dams. We predicted the inflow of the Yongdam Multipurpose Dam which is located in the upper stream of the Geumgang. The hourly flow data of Yongdam Dam from 2006 to 2018 provided by WAMIS was used as the analysis data. Predictive analysis was performed under various of variable condition in order to compare and analyze the prediction accuracy according to four learning parameters of the LSTM model. Root mean square error (RMSE), Mean absolute error (MAE) and Volume error (VE) were calculated and evaluated its accuracy through comparing the predicted and observed inflows. We found that all the models had lower accuracy at high inflow rate and hourly precipitation data (2006~2018) of Yongdam Dam utilized as additional input variables to solve this problem. When the data of rainfall and inflow were utilized together, it was found that the accuracy of the prediction for the high flow rate is improved.

최근 데이터 예측 방법으로 인공신경망(Artificial Neural Network, ANN)분야에 대한 관심이 높아졌으며, 그 중 시계열 데이터 예측에 특화된 LSTM(Long Short-Term Memory)모형은 수문 시계열자료의 예측방법으로도 활용되고 있다. 본 연구에서는 구글에서 제공하는 딥러닝 오픈소스 라이브러리인 텐서플로우(TensorFlow)를 활용하여 LSTM모형을 구축하고 금강 상류에 위치한 용담다목적댐의 유입량을 예측하였다. 분석 자료로는 WAMIS에서 제공하는 용담댐의 2006년부터 2018년까지의 시간당 유입량 자료를 사용하였으며, 예측된 유입량과 관측 유입량의 비교를 통하여 평균제곱오차(RMSE), 평균절대오차(MAE), 용적오차(VE)를 계산하고 모형의 학습변수에 따른 정확도를 평가하였다. 분석결과, 모든 모형이 고유량에서의 정확도가 낮은 것으로 나타났으며, 이와 같은 문제를 해결하기 위하여 용담댐 유역의 시간당 강수량 자료를 추가 학습 자료로 활용하여 분석한 결과, 고유량에 대한 예측의 정확도가 높아지는 것을 알 수 있었다.

Keywords

References

  1. Bae, D.H., Jeong, D.M., and Kim, K.S. (2010). "Monthly dam inflow forecasts using weather forecasting information and neuro-fuzzy technique." Hydrological Sciences Journal, Taylor & Francis, Vol. 50, No. 1, pp. 99-113.
  2. Choi, J.H., Kim, J.S., Kwon, J.H., and Moon, Y.I. (2016). "Weighting assessment on evaluation indicators of dam rehabilitation using the AHP analysis." Journal of Korea Water Resources Association, Korea Water Resources Association, Vol. 49, No. 5, pp. 381-389. https://doi.org/10.3741/JKWRA.2016.49.5.381
  3. Eom, J.I., and Jung, K.S. (2019). "Estimation of hourly dam inflow using time series data." Journal of the Korean Society of Hazard Mitigation, Korean Society of Hazard Mitigation, Vol. 19, No. 2, pp. 163-168. https://doi.org/10.9798/kosham.2019.19.2.163
  4. Jeong, D.I., Kim, Y.O., Cho, S.J., and Shin, H.J. (2003). "A study on rainfall-runoff models for improving ensemble streamflow prediction." Journal of Korean Society of Civil Engineers, Korean Society of Civil Engineers, Vol. 23, No. 6B, pp. 521-530.
  5. 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, Korean Society of Hazard Mitigation, Vol. 18, No. 1, pp. 1-11. https://doi.org/10.9798/KOSHAM.2018.18.1.1
  6. Kang, S.K., Lee, D.R., Moon, J.W., and Choi, S.J. (2010). "Effects of dams and water use on flow regime alteration of the Geum River basin." Journal of Korea Water Resources Association, Korea Water Resources Association, Vol. 43, No. 4, pp. 325-336. https://doi.org/10.3741/JKWRA.2010.43.4.325
  7. Kim K.S. (2010). A study on the real time forecasting for monthly inflow Daecheong Dam using hydrologic time series analyses. Master's thesis, Seokyeong University, pp. 32-54.
  8. Kim, K.S., and Yang, J.H. (2004). "A study on rainwater storage methodology for rainwater utilization and preventing flood damage." Seoul Studies, The Seoul Institute, Vol. 5, No. 4, pp. 73-92.
  9. Kwon, O.I., and Shim, M.P. (1997). "Determination scheme of variable restricted water level during flood period of multipurpose dam." Journal of Korea Water Resources Association, Korea Water Resources Association, Vol. 30, No. 6, pp. 409-720.
  10. Park, M.K., Yoon, Y.S., Lee, H.H., and Kim, J.H. (2018). "Application of recurrent neural network for inflow prediction into multipurpose dam basin." Journal of Korea Water Resources Association, Korea Water Resources Association, Vol. 51, No. 12, pp. 1217-1227.
  11. Tran, Q.K., and Song, S.G. (2017). "Water level forecasting based on deep learning: A use case of trinity river-Texas-The United States." Journal of KIISE, Korean Institute of Information Scientists and Engineers, Vol. 44, No. 6, pp. 607-612. https://doi.org/10.5626/JOK.2017.44.6.607
  12. Wang, W., Jin, J., and Li, Y. (2009). "Prediction of inflow at three gores dam in Yangtze River with wavelet network model." Water Resources Management, Springer, Vol. 23, No. 13, pp. 2791-2803. https://doi.org/10.1007/s11269-009-9409-2