A hybrid river flow forecasting model combining hydrodynamic and machine learning models

Jae Gyeong Lee; Kyung Soo Jun

doi:10.3741/JKWRA.2024.57.11.859

All Issue

2024 Vol.57, Issue 11 Preview Page Next Page

Research Article

A hybrid river flow forecasting model combining hydrodynamic and machine learning models 동수역학적 모형과 기계학습을 모형을 결합한 하천 홍수 예측모형

30 November 2024. pp. 859-872

PDF XML

Abstract

A hybrid approach combining a distributed parameter hydrodynamic flow model and a deep learning based error correction model is proposed to enhance the forecasting accuracy of flood water levels. The hydrodynamic model for the downstream reach of the Han River was calibrated by considering the spatial distribution and temporal variation of Manning’s roughness coefficient in unsteady flow conditions. In the following error correction procedure, the systematic errors of the hydrodynamic model were captured using long short term memory (LSTM) model. To assess the model accuracy, performances of the hydrodynamic model, two LSTM based models (LSTM1 and LSTM2), and the hybrid model were compared. Regarding inputs, LSTM1 considers the historical observations only, and LSTM2 considers not only the historical observations that used in LSTM1 but also the prescribed boundary conditions required for the hydrodynamic forecast model. The root-mean-square errors were used for the evaluation of model performance. The hybrid model significantly improved the forecasting accuracy of flood water levels compared to individual models, which indicates that the hybrid model is able to take advantage of complementary strengths of both the hydrodynamic model and the LSTM model. The LSTM-based hybrid model also outperformed the hybrid model based on artificial neural network, resulting in less than 10 cm of root-mean-square error for 3-hour lead time forecasting of the flood water level.

Keywords

Flood water level forecasting

Hydrodynamic model

Machine learning model

Hybrid model

홍수위 예측의 정확도를 향상시키기 위하여 분포형 매개변수를 갖는 동수역학적 흐름모형과 심화학습 기반의 오차보정 모형을 결합한 혼합모형을 제시하였다. 한강 하류구간을 대상으로 동수역학적 계산모형을 수립하였으며, 부정류 상태에서 Manning 조도계수의 시간적, 공간적 변동성을 고려하여 모형의 보정을 수행하였다. 이어지는 오차보정 과정에서는 장단기 기억 신경망(LSTM) 모형을 사용하여 동수역학적 모형의 시스템적 오차를 찾아내고자 하였다. 모형의 정확성을 평가하기 위하여 동수역학적 모형, 두 개의 LSTM 기반 모형(LSTM1, LSTM2) 및 혼합모형 등 4개 모형의 성능을 비교하였다. LSTM1은 현재까지의 관측 자료들만을 입력변수로 사용하는 모형이며, LSTM2는 여기에 이후 시간에서의 동수역학적 모형 경계조건들을 입력변수로 추가하여 수위를 예측하도록 수립된 모형이다. 평균제곱근 오차를 평가 지표로 하여 모형들의 성능을 평가한 결과, 혼합모형이 개별 모형들에 비하여 홍수위의 예측 정확성을 현저하게 개선하는 것으로 나타났다. 이는 혼합모형이 동수역학적 모형과 LSTM 모형의 상호 보완적인 장점들을 모두 취함으로써 가능한 것이라 할 수 있다. 또한 LSTM 기반의 혼합모형은 인공신경망 기반의 혼합모형보다도 우수한 성능을 보여, 예측 선행시간이 3시간인 경우에도 제곱평균제곱근 오차 10 cm 이내로 홍수위를 예측하는 것으로 나타났다.

키워드

홍수위 예측

동수역학적 모형

기계학습 모형

혼합모형

References

Abebe, A.J., and Price, R.K. (2003). "Managing uncertainty in hydrological models using complementary models." Journal of Hydrological Sciences, Vol. 48, pp. 679-692.

10.1623/hysj.48.5.679.51450

Azad, A.S., Sokkalingam, R., Daud, H., Adhikary, S.K., Khurshid, H., Mazlan, S.N.A., and Rabbani, M.B.A. (2022). "Water level prediction through hybrid SARIMA and ANN models based on time series analysis: Red Hills reservoir case study." Journal of Sustainability, Vol. 14, 1843.

10.3390/su14031843

Cho, K., and Kim, Y. (2022). "Improving stream prediction in the WRF-Hydro model with LSTM networks." Journal of Hydrology, Vol. 605, 127297.

10.1016/j.jhydrol.2021.127297

Cunge, J.A., Holly, F.M., and Verwey, A. (1980). Practical aspects of computational river hydraulics. Pitman, Toronto, Ontario, Canada, pp. 65-72.

Doherty, J. (2000). Visual PEST: Model-independent parameter estimation. Watermark Computing & Waterloo Hydrogeologic, Waterloo, Ontario, Canada.

He, J., Valeo, C., Chu, A., and Neumann, N.F. (2011). "Prediction of event-based stormwater runoff quantity and quality by ANNs devel-oped using PMI-based input selection." Journal of Hydrology, Vol. 400, pp.10-23.

10.1016/j.jhydrol.2011.01.024

Hochreiter, S., and Schmidhuber, J. (1997). "Long short-term memory." Journal of Neural Computation, Vol. 9, pp. 1735-1780.

10.1162/neco.1997.9.8.17359377276

Jung, S., Cho, H., Kim, J., and Lee, G. (2018). "Prediction of water level in a tidal river using a deep-learning based LSTM model." Journal of Korea Water Resources Association, Vol. 51, No. 12, pp. 1207-1216.

Konapala, G., Kao, S.C., Painter, S.L., and Lu, D. (2020). "Machine learning assisted hybrid models can improve streamflow simulation in diverse catchments across the conterminous US." Environmental Research Letters, Vol. 15, No. 10, 104022.

10.1088/1748-9326/aba927

Li, L., and Jun, K.S. (2018). "Distributed parameter unsteady flow model for the Han River." Journal of Hydro-Environment Research, Vol. 21, pp. 86-95.

10.1016/j.jher.2018.08.004

Li, L., and Jun, K.S. (2022). "A hybrid approach to improve flood forecasting by combining a hydrodynamic flow model and artificial neural networks." Water, Vol. 14, No. 1393.

10.3390/w14091393

Li, L., Chung, E.S., and Jun, K.S. (2018). "Robust parameter set selection for a hydrodynamic model based on multi-site calibration using multi-objective optimization and minimax regret approach." Water Resources Management, Vol. 32, No. 12, pp. 3979-3995.

10.1007/s11269-018-2031-4

Lipton, Z.C., Berkowitz, J., and Elkan, C. (2015). "A critical review of recurrent neural networks for sequence learning." arXiv, accessed 30 June 2024, <http://arxiv.org/abs/1506.00019>.

May, R.J., Dandy, G.C., Maier, H.R., and Nixon, J.B. (2008a). "Application of partial mutual information variable selection to ANN forecast-ing of water quality in water distribution systems." Environmental Modeling and Software, Vol. 23, pp. 1289-1299.

10.1016/j.envsoft.2008.03.008

May, R.J., Maier, H.R., Dandy, G.C., and Fernando, T.M.K. (2008b). "Non-linear variable selection for artificial neural networks using partial mutual information." Environmental Modeling and Software, Vol. 23, pp. 1312-1326.

10.1016/j.envsoft.2008.03.007

Ricci, S., Piacentini, A., Thual, O., Le Pape, E., and Jonville, G. (2011). "Correction of upstream flow and hydraulic state with data assimi-lation in the context of flood forecasting." Hydrology and Earth System Sciences, Vol. 15, pp. 3555-3575.

10.5194/hess-15-3555-2011

Torres-Rua, A.F., Ticlavilca, A.M., Walker, W.R., and McKee, M. (2012). "Machine learning approaches for error correction of hydraulic simulation models for canal flow schemes." Journal of Irrigation and Drainage Engineering, Vol. 138, pp. 999- 1010.

10.1061/(ASCE)IR.1943-4774.0000489

Xu, T., and Valocchi, A.J. (2015). "Data-driven methods to improve baseflow prediction of a regional groundwater model." Computers and Geosciences, Vol. 85, pp. 124-136.

10.1016/j.cageo.2015.05.016

Xu, T., Valocchi, A.J., Choi, J., and Amir, E. (2014). "Use of machine learning methods to reduce predictive error of groundwater models." Groundwater, Vol. 52, pp. 448-460.

10.1111/gwat.1206123647322

Yang, S., Yang, D., Chen, J., Santisirisomboon, J., Lu, W., and Zhao, B. (2020). "A physical process and machine learning combined hydro-logical model for daily streamflow simulations of large watersheds with limited observation data." Journal of Hydrology, Vol. 590, 125206.

10.1016/j.jhydrol.2020.125206

Yue, Z., Ai, P., Xiong, C., Hong, M., and Song, Y. (2020). "Mid-to long-term runoff prediction by combining the deep belief network and partial least-squares regression." Journal of Hydroinformatics, Vol. 22, No. 5, pp. 1283-1305.

10.2166/hydro.2020.022

Information

Publisher :KOREA WATER RESOURECES ASSOCIATION
Publisher(Ko) :한국수자원학회
Journal Title :Journal of Korea Water Resources Association
Journal Title(Ko) :한국수자원학회 논문집
Volume : 57
No :11
Pages :859-872
Received Date : 2024-10-08
Revised Date : 2024-10-29
Accepted Date : 2024-10-29
DOI :https://doi.org/10.3741/JKWRA.2024.57.11.859

Journal of Korea Water Resources Association ISSN:2799-8746(Print) 2799-8754(Online) 한국수자원학회 논문집

All Issue