Sequence to Sequence based LSTM (LSTM-s2s)모형을 이용한 댐유입량 예측에 대한 연구 = Application of sequence to sequence learning based LSTM model (LSTM-s2s) for forecasting dam inflow

다국어 초록 (Multilingual Abstract)

Forecasting dam inflow based on high reliability is required for efficient dam operation. In this study, deep learning technique, which is one of the data-driven methods and has been used in many fields of research, was manipulated to predict the dam inflow. The Long Short-Term Memory deep learning with Sequence-to-Sequence model (LSTM-s2s), which provides high performance in predicting time-series data, was applied for forecasting inflow of Soyang River dam. Various statistical metrics or evaluation indicators, including correlation coefficient (CC), Nash-Sutcliffe efficiency coefficient (NSE), percent bias (PBIAS), and error in peak value (PE), were used to evaluate the predictive performance of the model. The result of this study presented that the LSTM-s2s model showed high accuracy in the prediction of dam inflow and also provided good performance for runoff event based runoff prediction. It was found that the deep learning based approach could be used for efficient dam operation for water resource management during wet and dry seasons.

국문 초록 (Abstract)

효율적인 댐 운영을 위해서는 높은 신뢰도를 기반으로 하는 유입량 예측이 요구된다. 본 연구에서는 최근 다양한 분야에서 사용되고 있는 데이터 기반의 예측 방법 중 하나인 딥러닝을 댐 유입량 예측에 활용하였다. 그 중 시계열 자료 예측에 높은 성능을 보이는 Sequence-to-Sequence 구조 기반의 Long Short-Term Memory 딥러닝 모형(LSTM-s2s)을 이용하여 소양강 댐의 유입량을 예측하였다. 모형의 예측 성능을 평가하기 위해 상관계수, Nash–Sutcliffe 효율계수, 평균편차비율, 그리고 첨두값 오차를 이용하였다. 그 결과, LSTM-s2s 모형은 댐 유입량 예측에 대한 높은 정확도를 보였으며, 단일 유량 수문곡선 기반의 예측 성능에서도 높은 신뢰도를 보였다. 이를 통해 홍수기와 이수기에 수자원 관리를 위한 효율적인 댐 운영에 딥러닝 모형의 적용 가능성을 확인할 수 있었다.

참고문헌 (Reference)

1 방윤현, "소하천의 초기계획하폭 산정식 개발에 관한 연구: 강원영서 지역을 대상으로" 한국방재학회 18 (18): 357-367, 2018

2 김영일, "미계측유역의 수문모형 매개변수 추정을 위한 하이브리드 지역화모형의 개발" 한국수자원학회 51 (51): 677-686, 2018

3 이기하, "메콩강 유출모의를 위한 물리적 및 데이터 기반 모형의 비교 ․ 분석" 한국수자원학회 51 (51): 503-514, 2018

4 강나래, "레이더 강우 자료를 이용한 도시유역의 유출 모의" 한국습지학회 15 (15): 413-422, 2013

5 김동현, "딥러닝을 활용한 산지습지 수위 예측 모형 개발" 한국습지학회 22 (22): 106-112, 2020

6 목지윤, "딥러닝을 활용한 다목적댐 유입량 예측" 한국수자원학회 53 (53): 97-105, 2020

7 정성호, "딥러닝 오픈 라이브러리를 이용한 하천수위 예측" 한국방재학회 18 (18): 1-11, 2018

8 박명기, "다목적댐 유입량 예측을 위한 Recurrent Neural Network 모형의 적용 및 평가" 한국수자원학회 51 (51): 1217-1227, 2018

9 한정호, "기후변화에 따른 미래 극한호우사상이 소양강댐 유역의 유량 및 유사량에 미치는 영향" 한국물환경학회 33 (33): 160-169, 2017

10 정의상, "가뭄이 농업유역의 하천 유량에 미치는 영향 분석: 준분포형 유역모델 STREAM의 논산천 유역 적용 사례" 한국해양환경·에너지학회 21 (21): 398-410, 2018

11 배영혜, "VARMA와 머신러닝 모형을 이용한 소양강댐 월유입량 예측" 기후연구소 14 (14): 183-198, 2019

12 Assem, H., "Urban water flow and water level prediction based on deep learning" Springer 3 : 317-329, 2017

13 Yan, J., "Urban flash flood forecast using support vector machine and numerical simulation" 20 (20): 221-231, 2018

14 Granata, F., "Support vector regression for rainfall-runoff modeling in urban drainage: A comparison with the EPA’s storm water management model" 8 (8): 69-81, 2016

15 Chen, L., "Streamflow forecast uncertainty evolution and its effect on real-time reservoir operation" 540 : 712-726, 2016

16 Neitsch, S. L., "Soil and water assessment tool theoretical documentation version 2009"

17 Sutskever, I., "Sequence to sequence learning with neural networks"

18 김보경, "SWAT 모형 매개변수의 적정 범위 추정방법개발 : 임하호 유입량 및 부유사량에의 적용" 대한토목학회 27 (27): 691-668, 2007

19 Ghumman, A. R., "Runoff forecasting by artificial neural network and conventional model" 50 (50): 345-350, 2011

20 Chen, P. A., "Reinforced recurrent neural networks for multi-step-ahead flood forecasts" 497 : 71-79, 2013

21 Zhang, Q., "Regionalization and spatial changing properties of droughts across the Pearl River basin, China" 472 : 355-366, 2012

22 Kratzert, F., "Rainfall-runoff modelling using long short-term memory (LSTM) networks" 22 : 6005-6022, 2018

23 Lee, K., "Prediction of heavy rain damage using deep learning" 12 (12): 1942-1959, 2020

24 Wilcox, B. P., "Predicting runoff from rangeland catchments: A comparison of two models" 26 (26): 2401-2410, 1990

25 Riad, S., "Predicting catchment flow in a semi‐arid region via an artificial neural network technique" 18 (18): 2387-2393, 2004

26 Cho, K., "On the properties of neural machine translation: Encoderdecoder approaches"

27 Abrahart, R., "Neural networks for hydrological modeling" CRC Press 2004

28 Gholami, V. C. K. W., "Modeling of groundwater level fluctuations using dendrochronology in alluvial aquifers" 529 : 1060-1069, 2015

29 Moriasi, D. N., "Model evaluation guidelines for systematic quantification of accuracy in watershed simulations" 50 (50): 885-900, 2007

30 Noh, H., "Long-term simulation of daily streamflow using radar rainfall and the SWAT model: A case study of the Gamcheon basin of the Nakdong River, Korea" 2016 : 431-442, 2016

31 Palmer, M., "Linkages between flow regime, biota, and ecosystem processes: Implications for river restoration" 365 (365): eaaw2087-, 2019

32 Cho, K., "Learning phrase representations using RNN encoder-decoder for statistical machine translation"

33 Tian, Y., "Integration of a parsimonious hydrological model with recurrent neural networks for improved streamflow forecasting" 10 (10): 1655-, 2018

34 Bicknell, B. R., "Hydrological simulation programFORTRAN. User’s manual for release 11"

35 Montanari, A., "Fractionally differenced ARIMA models applied to hydrologic time series:Identification, estimation, and simulation" 33 (33): 1035-1044, 1997

36 Mosavi, A., "Flood prediction using machine learning models: Literature review" 10 (10): 1536-, 2018

37 Legates, D. R., "Evaluating the use of “goodness-of-fit” measures in hydrologic and hydroclimatic model validation" 35 (35): 233-241, 1999

38 Feng, D., "Enhancing streamflow forecast and extracting insights using long‐short term memory networks with data integration at continental scales" 56 (56): e2019WR026-, 2020

39 Adeyemi, O., "Dynamic neural network modelling of soil moisture content for predictive irrigation scheduling" 18 (18): 3408-3429, 2018

40 Choi, C., "Development of water level prediction models using machine learning in wetlands: A case study of Upo Wetland in South Korea" 12 (12): 93-110, 2020

41 Zhang, J., "Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas" 561 : 918-929, 2018

42 Hu, C., "Deep learning with a long short-term memory networks approach for rainfall-runoff simulation" 10 (10): 1543-1558, 2018

43 Xu, J., "Dadu River runoff forecasting via Seq2Seq" 494-498, 2019

44 Fan, H., "Comparison of long short term memory networks and the hydrological model in runoff simulation" 12 (12): 175-189, 2020

45 Blöschl, G., "Changing climate both increases and decreases European river floods" 573 (573): 108-111, 2019

46 Kim, K. S., "A study on the real time forecasting for monthly inflow Daecheong dam using hydrologic time series analyses" Seokyeong University 2010

47 Devia, G. K., "A review on hydrological models" 4 : 1001-1007, 2015

48 Xiang, Z., "A rainfall-runoff model with LSTM-based sequence‐to‐sequence learning" 56 : e2019WR025-, 2020

49 Shoaib, M., "A comparison between wavelet based static and dynamic neural network approaches for runoff prediction" 535 : 211-225, 2016