KSCE Journal of Civil and Environmental Engineering Research (대한토목학회논문집)

Korean Society of Civil Engeneers (대한토목학회)
권호 표지
DOI QR코드

DOI QR Code

A SVR Based-Pseudo Modified Einstein Procedure Incorporating H-ADCP Model for Real-Time Total Sediment Discharge Monitoring

실시간 총유사량 모니터링을 위한 H-ADCP 연계 수정 아인슈타인 방법의 의사 SVR 모형

  • 노효섭 (서울대학교 건설환경공학부) ;
  • 손근수 (한국수자원조사기술원 첨단인프라실) ;
  • 김동수 (단국대학교 토목환경공학과) ;
  • 박용성 (서울대학교 건설환경공학부.건설환경종합연구소)
  • Received : 2023.02.23
  • Accepted : 2023.03.29
  • Published : 2023.06.01
Download PDF
⟨ Previous Next ⟩

Abstract

Monitoring sediment loads in natural rivers is the key process in river engineering, but it is costly and dangerous. In practice, suspended loads are directly measured, and total loads, which is a summation of suspended loads and bed loads, are estimated. This study proposes a real-time sediment discharge monitoring system using the horizontal acoustic Doppler current profiler (H-ADCP) and support vector regression (SVR). The proposed system is comprised of the SVR model for suspended sediment concentration (SVR-SSC) and for total loads (SVR-QTL), respectively. SVR-SSC estimates SSC and SVR-QTL mimics the modified Einstein procedure. The grid search with K-fold cross validation (Grid-CV) and the recursive feature elimination (RFE) were employed to determine SVR's hyperparameters and input variables. The two SVR models showed reasonable cross-validation scores (R2) with 0.885 (SVR-SSC) and 0.860 (SVR-QTL). During the time-series sediment load monitoring period, we successfully detected various sediment transport phenomena in natural streams, such as hysteresis loops and sensitive sediment fluctuations. The newly proposed sediment monitoring system depends only on the gauged features by H-ADCP without additional assumptions in hydraulic variables (e.g., friction slope and suspended sediment size distribution). This method can be applied to any ADCP-installed discharge monitoring station economically and is expected to enhance temporal resolution in sediment monitoring.

자연하천에서의 유사량 계측은 하천공학적으로 중요한 의미를 가지지만 계측 방법의 비용 문제로 유사량 실측에 어려움이 따른다. 특히 소류사량 계측의 어려움으로 인해 주기적인 유사량 모니터링의 대부분이 부유사 농도 계측에만 제한되어 있는 실정이다. 본 연구에는 자동유량관측소에 설치된 횡방향 도플러 유속계(H-ADCP)의 후방산란값과 부유사 농도의 상관관계를 이용해 실시간으로 부유사 농도를 산정하고 총유사량을 산정하는 서포트벡터회귀 모형을 제안한다. 제안하는 실시간 총유사량 모니터링 시스템은 부유사 농도 모형과 수정 아인슈타인 방법을 모사하는 총유사량 산정 모형으로 구성된다. 각 모형의 매개변수와 입력변수는 K겹 교차검증 기반 격자검색 방법과 재귀적 특징 제거법을 이용해 결정되었다. 교차검증에서 부유사 농도 모형과 총유사량 산정 모형의 R2가 각각 0.885와 0.860으로 유사량-유량 관계곡선에 비해 정확한 것으로 나타났다. 시계열 유사량 관측을 통해 새로 제시되는 실시간 총유사량 관측 시스템이 자연하천에서 발달하는 유사량-유량 이력관계와 미세한 유량 변화에서 나타나는 유사량 변화를 성공적으로 관측할 수 있음을 확인했다. 본 연구에서 제안하는 방법은 마찰경사나 부유사 입도 등의 수리 조건을 가정할 필요 없이 H-ADCP의 원시자료만으로 부유사 농도와 총유사량을 산정할 수 있어 기존 방법에 비해 불확도가 적으며 경제적이다. 본 방법은 H-ADCP가 설치된 유사량 관측소에 광범위하게 적용 가능해 유사량 모니터링의 시간적 해상도를 경제적으로 크게 줄일 수 있을 것으로 기대된다.

Keywords

Acknowledgement

본 연구는 환경부의 재원으로 한국환경산업기술원의 수요대응형 물공급서비스 연구사업의 지원을 받아 수행되었으며 이에 감사드립니다(2020002650001). 또한, 서울대학교 공학연구원의 지원에도 감사드립니다.

References

  1. Al-Mejibli, I. S., Alwan, J. K. and Abd, D. H. (2020). "The effect of gamma value on support vector machine performance with different kernels." International Journal of Electrical and Computer Engineering, Vol. 10, No. 5, 5497.
  2. Boser, B. E., Guyon, I. M. and Vapnik, V. N. (1992). "A training algorithm for optimal margin classifiers." Proceedings of the Fifth Annual Workshop on Computational Learning Theory, pp. 144-152.
  3. Chaudhry, M. H. (2008). Open-channel flow, Springer, NY, U.S., pp. 481.
  4. Colby, B. R. and Hembree, C. H. (1954). "Computations of total sediment discharge, Niobrara River near Cody, Nebraska." Science, Vol. 119, No. 3097, pp. 657-658. https://doi.org/10.1126/science.119.3097.657.b
  5. Cortes, C. and Vapnik, V. (1995). "Support-vector networks." Machine Learning, Vol. 20, No. 3, pp. 273-297. https://doi.org/10.1007/BF00994018
  6. Downing, A., Thorne, P. D. and Vincent, C. E. (1995). "Backscattering from a suspension in the near field of a piston transducer." The Journal of the Acoustical Society of America, Vol. 97, No. 3, pp. 1614-1620, doi: 10.1121/1.412100.
  7. Drucker, H., Burges, C. J. C., Kaufman, L., Smola, A. and Vapnik, V. (1997). "Support vector regression machines." Advances in Neural In-Formation Processing System, Vol. 9, pp. 155-161.
  8. Flammer, G. H. (1962). Ultrasonic measurement of suspended sediment, US Government Printing Office, Vol. 1141.
  9. Gellis, A. C. (2013). "Factors influencing storm-generated suspended-sediment concentrations and loads in four basins of contrasting land use, humid-tropical Puerto Rico." Catena, Vol. 104, pp. 39-57. https://doi.org/10.1016/j.catena.2012.10.018
  10. Guerrero, M., Ruther, N. and Szupiany, R. N. (2012). "Laboratory validation of acoustic Doppler current profiler (ADCP) techniques for suspended sediment investigations." Flow Measurement and Instrumentation, Vol. 23, No. 1, pp. 40-48. https://doi.org/10.1016/j.flowmeasinst.2011.10.003
  11. Guyon, I., Weston, J., Barnhill, S. and Vapnik, V. (2002). "Gene selection for cancer classification using support vector machines." Machine Learning, Vol. 46, No. 1-3, pp. 389-422. https://doi.org/10.1023/A:1012487302797
  12. Haught, D., Venditti, J. G. and Wright, S. A. (2017). "Calculation of in situ acoustic sediment attenuation using off-the-shelf horizontal ADCPs in low concentration settings." Water Resources Research, Vol. 53, No. 6, pp. 5017-5037. https://doi.org/10.1002/2016WR019695
  13. Hoitink, A. J. F. and Hoekstra, P. (2005). "Observations of suspended sediment from ADCP and OBS measurements in a mud-dominated environment." Coastal Engineering, Vol. 52, No. 2, pp. 103-118. https://doi.org/10.1016/j.coastaleng.2004.09.005
  14. Landers, M. N. (2012). Fluvial suspended sediment characteristics by high-resolution, surrogate metrics of turbidity, laser-diffraction, acoustic backscatter, and acoustic attenuation, Ph. D. Thesis, Dept. of Civil and Environ. Eng., Georgia Inst. of Technol., Atlanta, Ga.
  15. Landers, M. N., Straub, T. D., Wood, M. S. and Domanski, M. M. (2016). Sediment acoustic index method for computing continuous suspended-sediment concentrations, No. 3-C5. US Geological Survey.
  16. Ministry of Environment (MoE) (2019). Hydrological annual report in Korea (in Korean).
  17. Noh, H., Son, G., Kim, D. and Park, Y. S. (2021). "Clustering of sediment characteristics in South Korean Rivers and its expanded application strategy to H-ADCP based suspended sediment concentration monitoring technique." Journal of Korea Water Resources Association, Vol. 55, No. 1, pp. 43-57 (in Korean).
  18. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M. and Duchesnay, E. (2011). "Scikit-learn: Machine learning in python." Journal of Machine Learning Research, Vol. 12, pp. 2825-2830.
  19. Raghavendra, N. S. and Deka, P. C. (2014). "Support vector machine applications in the field of hydrology: A review." Applied Soft Computing, Vol. 19, pp. 372-386. https://doi.org/10.1016/j.asoc.2014.02.002
  20. Rajaee, T., Nourani, V., Zounemat-Kermani, M. and Kisi, O. (2011). "River suspended sediment load prediction: application of ANN and wavelet conjunction model." Journal of Hydrologic Engineering, Vol. 16, No. 8, pp. 613-627. https://doi.org/10.1061/(ASCE)HE.1943-5584.0000347
  21. Schulkin, M. and Marsh, H. W. (1962). "Sound absorption in sea water." The Journal of the Acoustical Society of America, Vol. 34, No. 6, pp. 864-865. https://doi.org/10.1121/1.1918213
  22. Sheng, J. and Hay, A. E. (1988). "An examination of the spherical scatterer approximation in aqueous suspensions of sand." The Journal of the Acoustical Society of America, Vol. 83, No. 2, pp. 598-610. https://doi.org/10.1121/1.396153
  23. Smola, A. (1996). Regression estimation with support vector learning machines, MSc Thesis, The Technical University, Munchen, Germany.
  24. Son, G. (2021). Riverine sediment load measurement technique using acoustic doppler current profilers considering hysteresis. Ph. D. Thesis, Dankook University (in Korean).
  25. Topping, D. J., Wright, S. A., Melis, T. S. and Rubin, D. M. (2006). "High-resolution monitoring of suspended-sediment concentration and grain size in the Colorado River using laser-diffraction instruments and a three-frequency acoustic system." Proceedings of the 8th Federal Interagency Sedimentation Conference, Reno, Nev., U.S. Geological Survey, pp. 539-546.
  26. Topping, D. J., Wright, S. A., Melis, T. S. and Rubin, D. M. (2007). "High-resolution measurements of suspended-sediment concentration and grain size in the Colorado River in Grand Canyon using a multi-frequency acoustic system." Proceedings, 10th International Symposium on River Sedimentation, August 1-4, 2007: Moscow, Russia, Vol. 3, pp. 330-339.
  27. Urick, R. J. (1948). "The absorption of sound in suspensions of irregular particles." The Journal of the Acoustical Society of America, Vol. 20, No. 3, pp. 283-289. https://doi.org/10.1121/1.1906373
  28. Urick, R. J. (1975). Principles of underwater sound for engineers, McGraw Hill, NY, U.S., p. 384.
  29. Venditti, J. G., Church, M., Attard, M. E. and Haught, D. (2016). "Use of ADCPs for suspended sediment transport monitoring: An empirical approach." Water Resources Research, Vol. 52, No. 4, pp. 2715-2736. https://doi.org/10.1002/2015WR017348
  30. Williams, G. P. (1989). "Sediment concentration versus water discharge during single hydrologic events in rivers." Journal of Hydrology, Vol. 111, pp. 89-106. https://doi.org/10.1016/0022-1694(89)90254-0
  31. Zounemat-Kermani, M., Mahdavi-Meymand, A., Alizamir, M., Adarsh, S. and Yaseen, Z. M. (2020). "On the complexities of sediment load modeling using integrative machine learning: Application of the great river of Loiza in Puerto Rico." Journal of Hydrology, Vol. 585, 124759.