Journal of Korean Society of Coastal and Ocean Engineers (한국해안·해양공학회논문집)

Korean Society of Coastal and Ocean Engineers (한국해안·해양공학회)
권호 표지
DOI QR코드

DOI QR Code

Prediction of Salinity of Nakdong River Estuary Using Deep Learning Algorithm (LSTM) for Time Series Analysis

시계열 분석 딥러닝 알고리즘을 적용한 낙동강 하굿둑 염분 예측

  • Woo, Joung Woon (Department of Civil and Environmental Engineering (Institute of Construction Technology Center), Inje University) ;
  • Kim, Yeon Joong (Department of Civil and Environmental Engineering, Inje University) ;
  • Yoon, Jong Sung (Department of Civil and Environmental Engineering, Inje University)
  • 우정운 (인제대학교 건설기술연구소) ;
  • 김연중 (인제대학교 건설환경공학부) ;
  • 윤종성 (인제대학교 건설환경공학부)
  • Received : 2022.07.11
  • Accepted : 2022.08.24
  • Published : 2022.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Nakdong river estuary is being operated with the goal of expanding the period of seawater inflow from this year to 2022 every month and creating a brackish water area within 15 km of the upstream of the river bank. In this study, the deep learning algorithm Long Short-Term Memory (LSTM) was applied to predict the salinity of the Nakdong Bridge (about 5 km upstream of the river bank) for the purpose of rapid decision making for the target brackish water zone and prevention of salt water damage. Input data were constructed to reflect the temporal and spatial characteristics of the Nakdong River estuary, such as the amount of discharge from Changnyeong and Hamanbo, and an optimal model was constructed in consideration of the hydraulic characteristics of the Nakdong River Estuary by changing the degree according to the sequence length. For prediction accuracy, statistical analysis was performed using the coefficient of determination (R-squred) and RMSE (root mean square error). When the sequence length was 12, the R-squred 0.997 and RMSE 0.122 were the highest, and the prior prediction time showed a high degree of R-squred 0.93 or more until the 12-hour interval.

낙동강 하굿둑은 올해 2022년 해수 유입기간을 매월 대조기마다로 확대, 하굿둑 상류 15 km 이내로 기수역 조성을 목표로 운영되고 있다. 목표 기수역 조성구간 및 염수피해 방지를 위한 신속한 의사결정을 위해 본 연구에서는 딥러닝 알고리즘 Long Short-Term Memory(LSTM)을 적용하여 낙동대교(하굿둑 상류 약 5 km)지점의 염분 예측을 수행하였다. 창녕·함안보 방류량 등 낙동강 하구역의 시·공간적 특성을 반영하기 위한 입력데이터를 구축하였으며, Sequence length에 따른 정도 변화를 통해 낙동강 하구역의 수리학적 특성을 고려한 최적모델을 구축하였다. 예측 정확도는 결정계수(R-squred)와 RMSE(root mean square error) 이용하여 통계분석을 실시하였으며. Sequence length가 12일 때 R-squred 0.997, RMSE 0.122로 가장 정도가 높았으며, 선행 예측시간은 12시간 간격까지 R -squred 0.93 이상으로 높은 정도를 보였다.

Keywords

Acknowledgement

본 연구는 2020년 한국연구재단의 이공분야기초연구사업 (NRF-2020R1I1A3A0403784313)의 재원으로 수행된 연구결과 중 일부임을 밝히며, 연구비 지원에 감사드립니다.

References

  1. Ahn, J.M., Lee, K. and Lyu, S. (2020). Effect of changes in watershed runoff characteristics on salinity intrusion in estuary using EFDC. KSCE J. Civ. Eng., 24(1), 87-98. https://doi.org/10.1007/s12205-020-1306-5
  2. Barzegar, R., Aalami, M.T. and Adamowski, J. (2020). Short-term water quality variable prediction using a hybrid CNN-LSTM deep learning model. Stochastic Environmental Research and Risk Assessment, 34(2), 415-433. https://doi.org/10.1007/s00477-020-01776-2
  3. Blumberg, A.F. and Mellor, G.L. (1987). A description of a three-dimensional coastal ocean circulation model. Three-dimensional Coastal Ocean Models, 4, 1-16. https://doi.org/10.1029/CO004p0001
  4. Chen, W., Liu, W., Huang, W. and Liu, H. (2017). Prediction of salinity variations in a tidal estuary using artificial neural network and three-dimensional hydrodynamic models. Comput. Water Energy Environ. Eng., 06, 107-128. https://doi.org/10.4236/cweee.2017.61009
  5. Han, C.-S., Park, S.-K., Jung, S.-W. and Roh, T.-Y. (2011). The study of salinity distribution at nakdong river estuary. J. Korean Soc. Coast. Ocean Eng., 23(1), 101-108 (in Korean). https://doi.org/10.9765/KSCOE.2011.23.1.101
  6. Hochreiter, S. and chmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  7. Jeong, S., Lee, S., Hur, Y.T., Kim, Y. and Kim, H.Y. (2022). Development of seawater inflow equations considering density difference between seawater and freshwater at the Nakdong River estuary. J. Korea Water Resources Association, 55(5), 383-392 (in Korean).
  8. Kim, T. (2020). Study on th Behavior Analysis of Salt Water According to the Operation of Nakdong River Estuary Barrage. Inje University.
  9. Kim, T. and Lee, W.-D. (2022). Review on applications of machine learning in coastal and ocean engineering. J. Ocean Eng. Technol., 36(3), 194-210. https://doi.org/10.26748/KSOE.2022.007
  10. Lee, S.-M., Park, K.-D. and Kim, I.-K. (2020). Comparison of machine learning algorithms for Chl-a prediction in the middle of Nakdong River (focusing on water quality and quantity factors). J. Korean Soc. Water Wastewater, 34, 277-288 (in Korean). https://doi.org/10.11001/jksww.2020.34.4.277
  11. Melesse, A.M., Khosravi, K., Tiefenbacher, J.P., Heddam, S., Kim, S., Mosavi, A. and Pham, B.T. (2020). River water salinity prediction using hybrid machine learning models. Water, 12, 2951.
  12. Park, S. and Kim, K. (2021). Prediction of DO concentration in nakdong river estuary through case study based on long short term memory model. J. Korean Soc. Coast. Ocean Eng., 33(6), 238-245 (in Korean). https://doi.org/10.9765/KSCOE.2021.33.6.238
  13. Parsa, J., Etemad-Shahidi, A. and Hosseiny, S. (2007). Evaluation of computer and empirical models for prediction of salinity intrusion in the bahmanshir estuary. J. Coast. Res., 50(SI), 658-662.
  14. Seo, I.W. and Yun, S.H. (2016). Forcasting water quality by ANN model at the downstream of cheongpyeong dam. KSCE 2016 CONVENTION PROGRAM, 41-42.
  15. Yang, H., Lee, K., Choo, Y. and Kim, K. (2020). Underwater acoustic research trends with machine learning: Ocean parameter inversion applications. J. Ocean Eng. Technol., 34(5), 371-376. https://doi.org/10.26748/KSOE.2020.016