상하수도학회지 (Journal of Korean Society of Water and Wastewater)

  • 제34권1호
  • /
  • Pages.35-43
  • /
  • 2020
  • /
  • 1225-7672(pISSN)
  • /
  • 2287-822X(eISSN)
대한상하수도학회 (The Korean Society of Water and Wastewater)
권호 표지
DOI QR코드

DOI QR Code

LSTM 모형을 이용한 하천 고탁수 발생 예측 연구

Prediction of high turbidity in rivers using LSTM algorithm

  • 박정수 (국립한밭대학교 건설환경공학과) ;
  • 이현호 (한국수자원공사 데이터센터)
  • Park, Jungsu (Department of Civil and Environmental Engineering, Hanbat National University) ;
  • Lee, Hyunho (Data Center, K-water)
  • 투고 : 2020.01.03
  • 심사 : 2020.01.31
  • 발행 : 2020.02.15
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Turbidity has various effects on the water quality and ecosystem of a river. High turbidity during floods increases the operation cost of a drinking water supply system. Thus, the management of turbidity is essential for providing safe water to the public. There have been various efforts to estimate turbidity in river systems for proper management and early warning of high turbidity in the water supply process. Advanced data analysis technology using machine learning has been increasingly used in water quality management processes. Artificial neural networks(ANNs) is one of the first algorithms applied, where the overfitting of a model to observed data and vanishing gradient in the backpropagation process limit the wide application of ANNs in practice. In recent years, deep learning, which overcomes the limitations of ANNs, has been applied in water quality management. LSTM(Long-Short Term Memory) is one of novel deep learning algorithms that is widely used in the analysis of time series data. In this study, LSTM is used for the prediction of high turbidity(>30 NTU) in a river from the relationship of turbidity to discharge, which enables early warning of high turbidity in a drinking water supply system. The model showed 0.98, 0.99, 0.98 and 0.99 for precision, recall, F1-score and accuracy respectively, for the prediction of high turbidity in a river with 2 hour frequency data. The sensitivity of the model to the observation intervals of data is also compared with time periods of 2 hour, 8 hour, 1 day and 2 days. The model shows higher precision with shorter observation intervals, which underscores the importance of collecting high frequency data for better management of water resources in the future.

키워드

참고문헌

  1. Alexandrov, Y., Laronne, J.B., and Reid, I. (2007). Intra-event and interseasonal behaviour of suspended sediment in flash floods of the semiarid Northern Negev, Israel, Geomorphol., 85(1-2), 85-91. https://doi.org/10.1016/j.geomorph.2006.03.013
  2. Asrafuzzaman, M., Fakhruddin, A., and Hossain, M.A. (2011). Reduction of turbidity of water using locally available natural coagulants, ISRN microbiol., 1-6.
  3. Bennett, N.D., Croke, B.F., Guariso, G., Guillaume, J.H., Hamilton, S.H., Jakeman, A.J., and Perrin, C. (2013). Characterising performance of environmental models, Environ. Modell. Softw., 40, 1-20. https://doi.org/10.1016/j.envsoft.2012.09.011
  4. Burnham, K., and Anderson, D. (2002). Model Selection and Multi-model Inference. 2nd Ed., Springer, New York.
  5. Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., and Bengio, Y. (2014). Learning phrase representations using RNN Encoder-Decoder for statistical machine translation, arXiv, 1406.1078.
  6. Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling, arXiv, 1412.3555.
  7. Droppo, I.G., Liss, S.N., Williams, D., Nelson, T., Jaskot, C., and Trapp, B. (2009). Dynamic existence of waterborne pathogens within river sediment compartments. Implications for water quality regulatory affairs, Environ. Sci. Technol., 43(6), 1737-1743. https://doi.org/10.1021/es802321w
  8. Gray, A., Warrick, J., Pasternack, G., Watson, E., and Goni, M. (2014). Suspended sediment behavior in a coastal dry-summer subtropical catchment: effects of hydrologic preconditions, Geomorphol., 214, 485-501. https://doi.org/10.1016/j.geomorph.2014.03.009
  9. Greff, K., Srivastava, R.K., Koutník, J., Steunebrink, B.R., and Schmidhuber, J. (2016). LSTM: A search space odyssey, IEEE Trans. Neural Netw., 28(10), 2222-2232.
  10. Hicks, D.M., Gomez, B., and Trustrum, N.A. (2000). Erosion thresholds and suspended sediment yields, Waipaoa River basin, New Zealand, Water Resour. Res., 36(4), 1129-1142. https://doi.org/10.1029/1999WR900340
  11. Hochreiter, S., and Schmidhuber, J. (1997). Long short-term memory, Neural Comput., 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  12. Hu, Z., Zhang, Y., Zhao, Y., Xie, M., Zhong, J., Tu, Z., and Liu, J. (2019). A water quality prediction method based on the deep LSTM network considering correlation in smart mariculture, Sensors, 19(6), 1420. https://doi.org/10.3390/s19061420
  13. Huang, J., Gao, J., and Zhang, Y. (2015). Combination of artificial neural network and clustering techniques for predicting phytoplankton biomass of Lake Poyang, China, Limnol., 16(3), 179-191. https://doi.org/10.1007/s10201-015-0454-7
  14. LeCun, Y., Bengio, Y., and Hinton, G. (2015). Deep learning, Nat., 521(7553), 436-444. https://doi.org/10.1038/nature14539
  15. Lee, S., and Lee, D. (2018). Improved prediction of harmful algal blooms in four Major South Korea's Rivers using deep learning models, Int. J. Environ. Res. Public Health, 15(7), 1322. https://doi.org/10.3390/ijerph15071322
  16. Lin, W., Sung, S., Chen, L., Chung, H., Wang, C., Wu, R., and Peng, X. (2004). Treating high-turbidity water using full-scale floc blanket clarifiers, J. Environ. Eng., 130(12), 1481-1487. https://doi.org/10.1061/(ASCE)0733-9372(2004)130:12(1481)
  17. McCulloch, W.S., and Pitts, W. (1943). A logical calculus of the ideas immanent in nervous activity, Bull. Math. Biophys., 5(4), 115-133. https://doi.org/10.1007/BF02478259
  18. MOE. (2019). Water environment monitoring network installation and operation plan.
  19. Moriasi, D.N., Arnold, J.G., Van Liew, M.W., Bingner, R.L., Harmel, R.D., and Veith, T.L. (2007). Model evaluation guidelines for systematic quantification of accuracy in watershed simulations, T. ASABE, 50(3), 885-900. https://doi.org/10.13031/2013.23153
  20. Nair, V., and Hinton, G. E. (2010). "Rectified linear units improve restricted boltzmann machines", Proceedings of the 27th international conference on machine learning, 21-24 June, Haifa, Israel.
  21. Nistor, C.J., and Church, M. (2005). Suspended sediment transport regime in a debris-flow gully on Vancouver Island, British Columbia, Hydrol. Process., 19, 861-885. https://doi.org/10.1002/hyp.5549
  22. Olah, C. (2015). Understanding LSTM Networks, GITHUB blog, Retrieved from http://colah.github.io/posts/2015-08-Understanding-LSTMs/.
  23. Park, H.S., Chung, S.W., and Choung, S.A. (2017a). Analyzing the effect of an extreme turbidity flow event on the dam reservoirs in North Han River basin, J. Korean Soc. Water Environ., 33(3), 282-290. https://doi.org/10.15681/KSWE.2017.33.3.282
  24. Park, J., and Hunt, J.R. (2017b). Coupling fine particle and bedload transport in gravel-bedded streams, J. Hydrol., 552, 532-543. https://doi.org/10.1016/j.jhydrol.2017.07.023
  25. Park, J., and Hunt, J.R. (2018). Modeling fine particle dynamics in gravel-bedded streams: Storage and re-suspension of fine particles, Sci. Total. Environ., 634, 1042-1053. https://doi.org/10.1016/j.scitotenv.2018.04.034
  26. Rosenblatt, F. (1958). The perceptron: a probabilistic model for information storage and organization in the brain, Psychol. Rev., 65(6), 386. https://doi.org/10.1037/h0042519
  27. Seo, S.D., Lee, J.Y., and Ha, S.R. (2011). Effect of hydroelectric power plant discharge on the turbidity distribution in Dae-Cheong dam reservoir, Environ. Impact Assess., 20(2), 225-232.
  28. Singer, M.B., Aalto, R., James, L.A., Kilham, N.E., Higson, J.L., and Ghoshal, S. (2013). Enduring legacy of a toxic fan via episodic redistribution of California gold mining debris, Proc. Natl. Acad. Sci., 110(46), 18436-18441. https://doi.org/10.1073/pnas.1302295110
  29. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting, J. Mach. Learn. Res., 15(1), 1929-1958.
  30. Suttle, K.B., Power, M.E., Levine, J.M., and McNeely, C. (2004). How fine sediment in riverbeds impairs growth and survival of juvenile salmonids, Ecol. Appl., 14(4), 969-974. https://doi.org/10.1890/03-5190
  31. Walling, D. (1974). Suspended sediment and solid yields from a small catchment prior to urbanization, Fluv. Process. Instrum. Watersheds, 6, 169-192.
  32. Walling, D. (1977). Assessing the accuracy of suspended sediment rating curves for a small basin, Water Resour. Res., 13(3), 531-538. https://doi.org/10.1029/WR013i003p00531
  33. Warrick, J., Madej, M., Goñi, M., and Wheatcroft, R. (2013). Trends in the suspended-sediment yields of coastal rivers of northern California, J. Hydrol., 489, 1955-2010
  34. Warrick, J.A. (2015). Trend analyses with river sediment rating curves, Hydrol. Process., 29(6), 936-949. https://doi.org/10.1002/hyp.10198
  35. Williams, G.P. (1989). Sediment concentration versus water discharge during single hydrologic events in rivers, J. Hydrol., 111, 89-106. https://doi.org/10.1016/0022-1694(89)90254-0
  36. Wu, N., Huang, J., Schmalz, B., and Fohrer, N. (2014). Modeling daily chlorophyll a dynamics in a German lowland river using artificial neural networks and multiple linear regression approaches, Limnol., 15(1), 47-56. https://doi.org/10.1007/s10201-013-0412-1
  37. Zhou, J., Wang, Y., Xiao, F., Wang, Y., and Sun, L. (2018). Water quality prediction method based on IGRA and LSTM, Water, 10(9), 1148. https://doi.org/10.3390/w10091148

피인용 문헌

  1. 딥러닝과 앙상블 머신러닝 모형의 하천 탁도 예측 특성 비교 연구 vol.35, pp.1, 2020, https://doi.org/10.11001/jksww.2021.35.1.083