Ecology and Resilient Infrastructure

Korean Society of Ecology and Infrastructure Engineering (응용생태공학회)
권호 표지
DOI QR코드

DOI QR Code

Applicability Evaluation for Discharge Model Using Curve Number and Convolution Neural Network

Curve Number 및 Convolution Neural Network를 이용한 유출모형의 적용성 평가

  • Song, Chul Min (Department of Policy for Watershed Management, The Policy Council for Paldang Watershed) ;
  • Lee, Kwang Hyun (Department of Policy for Watershed Management, The Policy Council for Paldang Watershed)
  • 송철민 (특별대책지역 수질보전정책협의회 정책국) ;
  • 이광현 (특별대책지역 수질보전정책협의회 정책국)
  • Received : 2020.05.12
  • Accepted : 2020.06.10
  • Published : 2020.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Despite the various artificial neural networks that have been developed, most of the discharge models in previous studies have been developed using deep neural networks. This study aimed to develop a discharge model using a convolution neural network (CNN), which was used to solve classification problems. Furthermore, the applicability of CNN was evaluated. The photographs (pictures or images) for input data to CNN could not clearly show the characteristics of the study area as well as precipitation. Hence, the model employed in this study had to use numerical images. To solve the problem, the CN of NRCS was used to generate images as input data for the model. The generated images showed a good possibility of applicability as input data. Moreover, a new application of CN, which had been used only for discharge prediction, was proposed in this study. As a result of CNN training, the model was trained and generalized stably. Comparison between the actual and predicted values had an R2 of 0.79, which was relatively high. The model showed good performance in terms of the Pearson correlation coefficient (0.84), the Nash-Sutcliffe efficiency (NSE) (0.63), and the root mean square error (24.54 ㎥/s).

본 연구는 유출모형 연구를 위해 주로 사용되었던 DNN에서 벗어나, 다양한 신경망을 이용하여 유출모형을 개발하고 모형의 적합성을 나타내고자 하였다. 이를 위해 분류문제에만 사용되었던 CNN을 활용하였는데, 본 모형의 입력자료로 일반적으로 CNN에서 사용하는 사진을 이용할 수 없으며, 연구의 특성상 유역조건 및 강우 등의 영향이 반영된 수치적(numerical) 이미지(image)를 사용해야 하는 난해점이 있다. 이를 해결하고자 NRCS의 CN을 사용하여 이미지를 생성했으며, CNN 모형의 입력자료로 충분히 활용 가능함을 나타냈다. 이에 더하여, 유출 추정을 위해서만 사용되어왔던 CN의 새로운 용도를 제시할 수 있었다. 모형의 학습 및 검정 결과, 전반적으로 안정적으로 모형의 학습 및 일반화가 이루어졌으며, 관측값과 산정값간의 관계를 나타내는 R2는 0.79로 비교적 높은 값이 나타났다. 또한, 모형의 평가결과는 Pearson 상관계수, NSE, 및 RMSE 등이 각각 0.84, 0.65 및 24.54 ㎥/s으로 나타나, 전반적으로 양호한 모형의 산정성능을 보인것으로 나타났다.

Keywords

References

  1. Benzineb, K. and Remaoun, M. 2016. Daily rainfall-runoff modelling by neural networks in semi-arid zone: case of Wadi Ouahrane's basin. Journal of Fundamental and Applied Sciences 8(3): 956-970. https://doi.org/10.4314/jfas.v8i3.17
  2. Chen, Z. and Ho, P.H. 2019. Global-connected network with generalized ReLU activation. Pattern Recognition 96: 106961. https://doi.org/10.1016/j.patcog.2019.07.006
  3. EGIS. 2010. Environmental Geographic Information Service. egis.me.go.kr.
  4. Farias, C.A., Santos, C.A., Lourenco, A.M. and Carneiro, T.C. 2013. Kohonen neural networks for rainfall-runoff modeling: case study of pianco river basin. Journal of Urban and Environmental Engineering 7(1): 176-182. https://doi.org/10.4090/juee.2013.v7n1.176182
  5. Ide, H. and Kurita, T. 2017. Improvement of learning for CNN with ReLU activation by sparse regularization. 2017 International Joint Conference on Neural Networks. Anchorage, AK. pp. 2684-2691.
  6. Kalteh, A.M. 2008. Rainfall-runoff modelling using artificial neural networks (ANNs): modelling and understanding Caspian. Journal of Environmental Science 6(1): 53-58.
  7. Kasiviswanathan, K.S, Sudheer, K.P. 2013. Quantification of the predictive uncertainty of artificial neural network based river flow forecast models. Stoch Environ Res Risk Assess 27(1): 137-146. https://doi.org/10.1007/s00477-012-0600-2
  8. Keras. 2019. www.keras.io.
  9. KMA. 2020. Korea Meteorological Administration. www.kma.go.kr.
  10. Lim, H., Kim, J., Kwon, D. and Han, Y. 2017. Comparison analysis of TensorFlow's optimizer based on MNIST's CNN model. Journal of Advanced Technology Research 2(1): 6-14.
  11. Maca, P., Pech, P., and Pavlasek, J. 2014. Comparing the selected transfer functions and local optimization methods for neural network flood runoff forecast. Mathematical Problems in Engineering 2014: 1-10.
  12. Maier, H.A., Jain, G., Dandy, and Sudheer, K.P. 2010. Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions. Environmental Modelling & Software 25(8): 891-909. https://doi.org/10.1016/j.envsoft.2010.02.003
  13. Mishra, P.K. and Karmakar, S. 2019. Performance of optimum neural network in rainfall-runoff modeling over a river basin. International Journal of Environmental Science & Technology 16(3): 1289-1302. https://doi.org/10.1007/s13762-018-1726-7
  14. MLTM. 2012. Design Flood Estimation Techniques, Ministry of Land Transport and Maritime Affairs. (in Korean)
  15. 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. American Society of Agricultural and Biological Engineers 50(3): 885-900.
  16. Nourani, V., Komasi, M., and Alami, M.T. 2012. Hybrid wavelet-genetic programming approach to optimize ANN modeling of rainfall-runoff process. Journal of Hydrologic Engineering 17(6): 724-741. https://doi.org/10.1061/(ASCE)HE.1943-5584.0000506
  17. Othman, F. and Naseri, M. 2011. Reservoir inflow forecasting using artificial neural network. International Journal of the Physical Sciences 6(3): 434-440.
  18. Patel, B. and Joshi, G.S. 2017. Civil modeling of rainfallrunoff correlations using artificial neural network - A case study of Dharoi watershed of a Sabarmati river basin, India. Ajay Engineering Journal 3(2): 78-87. (online).
  19. Python 3.7. 2018, www.python.org. Released 27 June 2018.
  20. Rallison, R.E. 1980. Origin and evolution of the SCS runoff equation. In Symposium on Watershed Management 1980, ASCE. pp. 912-924.
  21. Shoaib, M., Shamseldin, A.Y., Melville, B.W. and Khan, M.M. 2016. A comparison between wavelet based static and dynamic neural network approaches for runoff prediction. Journal of Hydrology 535: 211-225. https://doi.org/10.1016/j.jhydrol.2016.01.076
  22. Singh, P.V., Akhilesh, K., Rawat, J.S., and Devendra, K. 2013. Artificial neural networks based daily rainfall-runoff model for an agricultural hilly watershed. International Journal of Engineering, Management & Sciences 4(2):108-112.
  23. Tensorflow. 2019. www.tensorflow.org.
  24. WAMIS. 2003. Water Resource Management Information System. www.wamis.go.kr.
  25. Wu, C.L. and Chau, K.W. 2011. Rainfall-runoff modeling using artificial neural network coupled with singular spectrum analysis. Journal of Hydrology 399(3-4): 394-409. https://doi.org/10.1016/j.jhydrol.2011.01.017
  26. Zhang, B. and Govindaraju, R.S. 2000. Prediction of watershed runoff using Bayesian concepts and modular neural networks. Water Resources Research 36(3): 753-762. https://doi.org/10.1029/1999WR900264

Cited by

  1. 활성화 함수에 따른 유출량 산정 인공신경망 모형의 성능 비교 vol.63, pp.1, 2020, https://doi.org/10.5389/ksae.2021.63.1.103