Journal of Korea Water Resources Association (한국수자원학회논문집)

Korea Water Resources Association (한국수자원학회)
권호 표지
DOI QR코드

DOI QR Code

Prediction of real-time data in water distribution systems using LSTM

LSTM을 이용한 상수도 시스템 실시간 데이터 예측

  • Cho, Eunyoung (Department of Civil and Environmental Engineering, Incheon National University) ;
  • Choi, Seonhong (Department of Civil and Environmental Engineering, Incheon National University) ;
  • Jung, Hanna (Department of Civil and Environmental Engineering, Incheon National University) ;
  • Jang, Dongwoo (Department of Civil and Environmental Engineering, Incheon National University)
  • 조은영 (인천대학교 건설환경공학과) ;
  • 최선홍 (인천대학교 건설환경공학과) ;
  • 정한나 (인천대학교 건설환경공학과) ;
  • 장동우 (인천대학교 건설환경공학과)
  • Received : 2023.10.03
  • Accepted : 2024.11.07
  • Published : 2024.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The domestic water supply coverage rate exceeds 99%, and as aging water pipes increase, the risk of hydraulic and water quality incidents in water pipelines is rising, making it urgent to establish countermeasures. To prepare for various incidents in water supply systems, such as red water and larval accidents, real-time monitoring devices have recently been installed in pipelines. Consequently, the importance of research utilizing the measured data is gradually increasing. In this study, the Long Short-Term Memory (LSTM) method was used to predict future hydraulic and water quality data within water pipelines. Two months of hourly real-time measurement data, including flow discharge, hydraulic pressure, residual chlorine, and pH, were collected from a small block in Seo-gu, Incheon. The predictive accuracy of the LSTM technique was evaluated using the coefficient of determination (R2) and Root Mean Square Error (RMSE). The results of predicting water supply factors over the next seven days showed a high correlation, with R2 reaching up to 0.91 for flow discharge. Electrical conductivity, water temperature, and residual chlorine also demonstrated high accuracy, with R2 values exceeding 0.8. However, hydraulic pressure showed a low R2 value of 0.019, suggesting that the predictive accuracy of the LSTM model is influenced by the continuity of the data used. The prediction results of real-time measurement data through LSTM can serve as crucial information for water supply management and incident prevention, and are expected to improve the ability to respond to incidents effectively.

국내 수도관의 보급률은 99% 이상이며, 이에 노후 수도관이 증가하고 있고, 상수관로 내 수리 및 수질 사고의 발생 위험을 높여 대책 마련이 시급한 실정이다. 적수 및 유충 사고 등 상수도에서 발생할 수 있는 다양한 사고에 대비하기 위하여 최근 실시간 모니터링 계측기가 관로에 설치되고 있고, 계측된 데이터를 활용한 연구의 중요성이 점차 증가하고 있다. 본 연구에서는 LSTM (Long Short-Term Memory) 방법을 이용하여 미래 상수도관 내 수리 및 수질 데이터를 예측하고자 하였다. 인천광역시 서구 소블럭의 유량, 수압, 잔류염소, pH 등 2개월의 시 단위 실시간 계측 데이터를 수집하였고, 결정계수와 RMSE (Root Mean Square Error)를 이용하여 LSTM 기법의 예측 정확도를 평가하였다. 장래 7일간의 상수도 인자별 데이터를 예측한 결과, 유량의 경우 R2가 최대 0.91로 높은 상관성을 보이는 것으로 나타났으며, 전기전도도, 수온, 잔류염소의 경우 0.8 이상의 값으로 높은 정확도를 보여주었다. 반면 수압의 경우 0.019의 낮은 값을 보여주었는데, 이는 LSTM 모델은 활용하는 데이터의 연속성 유무에 따라 예측 정확도에 영향을 받는 것으로 시사된다. LSTM을 통한 실시간 계측 데이터의 예측 결과는 상수도 관리와 사고 예방에 중요한 정보로 활용될 수 있으며, 사고 발생 시 대응 능력을 향상시키는 데 도움이 될 것으로 기대된다.

Keywords

Acknowledgement

본 과제는 행정안전부 지역맞춤형 재난안전 연구개발 사업의 지원을 받아 수행된 연구입니다(20019334).

References

  1. Abbasimehr, H., and Paki, R. (2022). “Improving time series forecasting using LSTM and attention models.” Journal of Ambient Intelligence and Humanized Computing, Vol. 13, No. 1, pp. 673-691. https://doi.org/10.1007/s12652-020-02761-x
  2. Caissie, D., Satish, M.G., and El‐Jabi, N. (2005). “Predicting river water temperatures using the equilibrium temperature concept with application on Miramichi River catchments (New Brunswick, Canada).” Hydrological Processes: An International Journal, Vol. 19, No. 11, pp. 2137-2159. https://doi.org/10.1002/hyp.5684
  3. Dong, Q., Lin, Y., Bi, J., and Yuan, H. (2019). “An integrated deep neural network approach for large-scale water quality time series prediction.” 2019 IEEE International Conference on Systems, Man, and Cybernetics (SMC), IEEE, Bari, Italy, pp. 3537-3542.
  4. Gers, F.A., Schmidhuber, J., and Cummins, F. (1999). “Continual prediction using LSTM with forget gates.” In Neural Nets WIRN Vietri-99: Proceedings of the 11th Italian Workshop on Neural Nets, Vietri Sul Mare, Salerno, Italy, Springer London, pp. 133-138.
  5. Ghimire, S., Yaseen, Z.M., Farooque, A.A., Deo, R.C., Zhang, J., and Tao, X. (2021). “Streamflow prediction using an integrated methodology based on convolutional neural network and long short-term memory networks.” Scientific Reports, Vol. 11, No. 1, 17497. https://doi.org/10.1038/s41598-021-96751-4
  6. Han, H., Choi, C., Jung, J., and Kim, H.S. (2021). “Application of sequence to sequence learning based LSTM model (LSTMs2s) for forecasting dam inflow.” Journal of Korea Water Resources Association, Vol. 54, No. 3, pp. 157-166. https://doi.org/10.3741/JKWRA.2021.54.3.157
  7. Hochreiter, S., and Schmidhuber, J. (1997). “Long short-term memory.” Neural Computation, Vol. 9, No. 8, pp. 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  8. 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, Vol. 19, No. 6, 1420. https://doi.org/10.3390/s19061420
  9. Jiang, Y., Li, C., Sun, L., Guo, D., Zhang, Y., and Wang, W. (2021). “A deep learning algorithm for multi-source data fusion to predict water quality of urban sewer networks.” Journal of Cleaner Production, Vol. 318, 128533. https://doi.org/10.1016/j.jclepro.2021.128533
  10. Jung, S., Cho, H., Kim, J., and Lee, G. (2018). “Prediction of water level in a tidal river using a deep-learning based LSTM model.” Journal of Korea Water Resources Association, Vol. 51, No. 12, pp. 1207-1216. https://doi.org/10.3741/JKWRA.2018.51.12.1207
  11. Kong, Y., Wang, Z., Nie, Y., Zhou, T., Zohren, S., Liang, Y., and Wen, Q. (2024). “Unlocking the power of LSTM for long term time series forecasting.” arXiv preprint, arXiv:2408.10006.
  12. Li, W., Kiaghadi, A., and Dawson, C. (2021). “Exploring the best sequence LSTM modeling architecture for flood prediction.” Neural Computing and Applications, Vol. 33, pp. 5571-5580. https://doi.org/10.1007/s00521-020-05334-3
  13. Lin, T.Y., Goyal, P., Girshick, R., He, K., and Dollar, P. (2017). “Focal loss for dense object detection.” Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy, pp. 2980-2988.
  14. Liu, P., Wang, J., Sangaiah, A.K., Xie, Y., and Yin, X. (2019). “Analysis and prediction of water quality using LSTM deep neural networks in IoT environment.” Sustainability, Vol. 11, No. 7, 2058. https://doi.org/10.3390/su11072058
  15. Mehedi, M.A.A., Khosravi, M., Yazdan, M.M.S., and Shabanian, H. (2022). “Exploring temporal dynamics of river discharge using univariate long short-term memory (LSTM) recurrent neural network at East Branch of Delaware River.” Hydrology, Vol. 9, No. 11, 202. https://doi.org/10.3390/hydrology9110202
  16. Mok, J.-Y. (2019). Construction of LSTM model and prediction of dam inflow using deep learning. Master’s Dissertation, University of Seoul.
  17. Moon, T., Choi, J., Kim, S., Cha, J., Yoom, H., and Kim, C. (2008). “Prediction of influent flow rate and influent components using artificial neural network (ANN).” Journal of Korean Society on Water Environment, Vol. 24, No. 1, pp. 91-98.
  18. Park, K., Jung, Y., Seong, Y., and Lee, S. (2022). “Development of deep learning models to improve the accuracy of water levels time series prediction through multivariate hydrological data.” Water, Vol. 14, No. 3, 469. https://doi.org/10.3390/w14030469
  19. Park, M.K., Yoon, Y.S., Lee, H.H., and Kim, J.H. (2018). “Application of recurrent neural network for inflow prediction into multi-purpose dam basin.” Journal of Korea Water Resources Association, Vol. 51, No. 12, pp. 1217-1227. https://doi.org/10.3741/JKWRA.2018.51.12.1217
  20. Smith, S.L. (2017). “Don't decay the learning rate, increase the batch size.” arXiv preprint, arXiv:1711.00489.
  21. Tran, Q.K., and Song, S.K. (2017). “Water level forecasting based on deep learning: A use case of Trinity River-Texas-The United States.” Journal of the Korean Institute of Information Scientists and Engineers, Vol. 44, No. 6, pp. 607-612.
  22. Van Houdt, G., Mosquera, C., and Nápoles, G. (2020). “A review on the long short-term memory model.” Artificial Intelligence Review, Vol. 53, No. 8, pp. 5929-5955. https://doi.org/10.1007/s10462-020-09838-1
  23. Wang, H.,, Mu, L.,, Shi, F., and Dou, H. (2020). “Production prediction at ultra-high water cut stage via Recurrent Neural Network.” Petroleum Exploration and Development, Vol. 47, No. 5, pp. 1084-1090. https://doi.org/10.1016/S1876-3804(20)60119-7
  24. Zhang, Y., Li, C., Jiang, Y., Sun, L., Zhao, R., Yan, K., and Wang, W. (2022). “Accurate prediction of water quality in urban drainage network with integrated EMD-LSTM model.” Journal of Cleaner Production, Vol. 354, 131724. https://doi.org/10.1016/j.jclepro.2022.131724
  25. Zhou, J., Wang, Y., Xiao, F., Wang, Y., and Sun, L. (2018). “Water quality prediction method based on IGRA and LSTM.” Water, Vol. 10, No. 9, 1148. https://doi.org/10.3390/w10091148
  26. Zhu, S., Luo, X., Yuan, X., and Xu, Z. (2020). “An improved long short-term memory network for streamflow forecasting in the upper Yangtze River.” Stochastic Environmental Research and Risk Assessment, Vol. 34, pp. 1313-1329. https://doi.org/10.1007/s00477-020-01766-4