All Issue

2025 Vol.58, Issue 10 Preview Page

Special Issue: 지능형 도시홍수 예측

Development of a spatio-temporal deep learning model for urban drainage outflow prediction

우수관망 유출량 예측을 위한 시공간 딥러닝 모델 개발

Hyunjung Kima
,
Donghwi Jungb*
김 현정a
,
정 동휘b*
aResearch Professor, Future and Fusion Lab of Architectural, Civil and Environmental Engineering, Korea University, Seoul, Korea
bAssociate Professor, School of Civil, Environmental and Architectural Engineering, Korea University, Seoul, Korea
a고려대학교 건축사회환경공학부 미래건설환경융합연구소 연구교수
b고려대학교 건축사회환경공학부 부교수
*Corresponding Author
31 October 2025. pp. 927-937
PDF XML Citation
Abstract

This study developed a spatio-temporal deep learning-based model for accurate prediction of stormwater outflow in urban drainage systems. In order to reflect temporal and spatial characteristics affecting runoff simultaneously, CNN-RNN model was designed by combining Convolutional Neural Networks (CNNs) which extract spatial features, with Recurrent Neural Networks (RNNs), especially Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) models, which handle temporal sequences. The proposed model was applied two urban drainage basins in Seoul : Gasan 1 rainwater pump station basin and Hagye basin. Its performance was evaluated by comparing the synthetically generated data from an urban drainage network model, using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Nash-Sutcliffe Efficiency (NSE) as evaluation metrics. In the results, CNN-LSTM and CNN-GRU models outperformed the respective single models (CNN, RNN) demonstrating lower prediction errors. These findings suggest that stormwater runoff in urban drainage systems is influenced by both spatial and temporal factors, and support the possibility of utilizing spatiotemporal deep learning-based models. Furthermore, the model is expected to be applicable for constructing real-time prediction systems that utilize smart sensor data, maintaining high prediction accuracy even with limited observational data.

Keywords
Urban drainage system
Outflow prediction
Spatio-temporal deep learning model
Artificial neural network
Urban flooding

본 연구에서는 도시유역 우수관망의 유출량을 정밀하게 예측하기 위한 시공간 딥러닝 기반 예측 모델을 개발하였다. 유출량에 영향을 미치는 시간적 및 공간적 특성을 동시에 반영하기 위해, 시계열 정보를 처리하는 순환신경망(Recurrent Neural Network, RNN) 계열인 장단기 메모리(Long-Short Term Memory, LSTM) 및 게이트 순환 유닛(Gated Recurrent Unit, GRU)과 공간 정보를 학습하는 합성곱신경망(Convolutional Neural Network, CNN)을 결합한 CNN-RNN 구조를 설계하였다. 개발된 모델은 서울시의 가산1 빗물펌프장 배수구역과 하계 배수구역에 적용되었으며, 총 27개의 강우사상(관측강우 3건, 확률강우 24건)을 Storm Water Management Model (SWMM)에 적용하여 생성한 1분 간격의 관망 노드 수위 및 유량 시계열 데이터를 학습 및 검증에 활용하였다. 모델의 예측 정확도 평가는 SWMM 모의 결과를 기준으로 하여 평균제곱근오차(Root Mean Squared Error, RMSE), 평균절대오차(Mean Absolute Error, MAE), 나쉬-서틀리프 효율계수(Nash-Sutcliffe Efficiency, NSE) 를 평가지표로 하여 비교·분석을 수행하였다. 그 결과, CNN-LSTM 및 CNN-GRU 모델은 각각의 단일 모델(CNN, RNN) 대비 더 낮은 오차를 보이며 우수한 예측 성능을 나타냈다. 이는 우수관망 내 유출 흐름이 시공간적으로 복합적인 영향을 받는다는 점을 시사하며, 시공간 딥러닝 기반 모델의 활용 가능성을 뒷받침한다. 또한 본 모델은 관측자료 확보가 제한적인 유역에서도 일부 수문 및 기상 관측 지점을 활용하여 실용적인 수준의 예측 성능을 확보할 수 있어, 향후 스마트 계측 기반의 실시간 예측 시스템 구축에도 적용 가능성이 높을 것으로 기대된다.

키워드
우수관망
유출량 예측
시공간 딥러닝 모델
인공신경망
도시홍수
References
1

Ahn, S.M. (2016) “Deep learning architectures and applications.” Korea intelligent information systems society, Vol. 22, No. 2, pp. 127-142.

10.13088/jiis.2016.22.2.127
2

Elshorbagy, A., Simonovic, S.P., and Panu, U.S. (2000). “Performance evaluation of artificial neural networks for runoff prediction.” Journal of Hydrologic Engineering, Vol. 5, No. 4, pp. 424-427.

10.1061/(ASCE)1084-0699(2000)5:4(424)
3

Halff, A.H., Halff, H.M., and Azmoodeh, M. (1993). “Predicting runoff from rainfall using neural networks.” Engineering Hydrology, ASCE, pp. 760-765.

4

Hochreiter, S., and Schmidhuber, J. (1997). “Long short-term memory.” Neural Computation, Vol. 9, No. 8, pp. 1735-1780.

10.1162/neco.1997.9.8.1735
5

Hu, C., Wu, Q., Li, H., Jian, S., Li, N., and Lou, Z. (2018). “Deep learning with a long short-term memory networks approach for rainfall-runoff simulation.” Water, Vol. 10, No. 11, 1543.

10.3390/w10111543
6

Jung, J., Mo, H., Lee, J., Yoo, Y., and Kim, H.S. (2021). “Flood stage forecasting at the Gurye-Gyo station in Sumjin River using LSTM-based deep learning models.” Journal of Korean Society of Hazard Mitigation, Vol. 21, No. 3, pp. 193-201.

10.9798/KOSHAM.2021.21.3.193
7

Kim, D., and Kang, S. (2021). “Data collection strategy for building rainfall-runoff LSTM model predicting daily runoff.” Journal of Korea Water Resources Association, Vol. 54, No. 10, pp. 795-805.

10.3741/JKWRA.2021.54.10.795
8

Kim, H.I., Han, K.Y., and Lee, J.Y. (2020). “Prediction of urban flood extent by LSTM model and logistic regression.” KSCE Journal of Civil and Environmental Engineering Research, Vol. 40, No. 3, pp. 273-283.

9

Kim, Y.H., Hwang, Y.K., Kang, T.G., and Jung, K.M. (2016). “LSTM language model based Korean sentence genera.” Journal of the Korea Institute of Construction Safety, Vol. 41, No. 5, pp. 592-601.

10.7840/kics.2016.41.5.592
10

LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). “Gradient- based learning applied to document recognition.” Proceedings of the Institute of Electrical and Electronics Engineers, Vol. 86, No. 11, pp. 2278-2324.

10.1109/5.726791
11

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, Vol. 17, No. 6, pp. 724-741.

10.1061/(ASCE)HE.1943-5584.0000506
12

Park, H.S. (2020). Machine learning & deep learning. Hanbitmedia.

13

Song, C.M. and Lee, K.H. (2020). “Applicability evaluation for discharge model using curve number and convolution neural network.” Ecology and Resilient Infrastructure, Vol. 7, No. 2, pp. 114-125.

14

Toth, E., and Brath, A. (2007) “Multistep ahead streamflow forecasting: Role of calibration data in conceptual and neural network modeling.” Water Resources Research, Vol. 43, No. 11, W11405.

10.1029/2006WR005383
15

Van, S.P., Le, H.M., Thanh, D.V., Dang, T.D., Loc, H.H., and Anh, D.T. (2020). “Deep learning convolutional neural network in rainfall-runoff modelling.” Journal of Hydroinformatics, Vol, 22, No. 3, pp. 541-561.

10.2166/hydro.2020.095
Information
  • Publisher :KOREA WATER RESOURECES ASSOCIATION
  • Publisher(Ko) :한국수자원학회
  • Journal Title :Journal of Korea Water Resources Association
  • Journal Title(Ko) :한국수자원학회 논문집
  • Volume : 58
  • No :10
  • Pages :927-937
  • Received Date : 2025-07-07
  • Revised Date : 2025-08-19
  • Accepted Date : 2025-09-04
  • DOI :https://doi.org/10.3741/JKWRA.2025.58.10.927
Journal Informaiton Journal of Korea Water Resources Association Journal of Korea Water Resources Association
Online Submission
Archives
  • scopus
  • KCI-수자원
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close