Abstract
Basically, leaks and faults in water distribution pipelines beget fairly severe water loss and affects largely its potability. For this reason, leakage detection is extremely significant for the preservation of water resources and quality. This paper introduces a novel unsupervised RNN model for leakage detection and location. The elaborated approach relies upon a multivariate LSTM autoencoder, as well as a multithresholding to monitor all water distribution network zones. A threshold for each measurement point of water distribution network is determined to identify anomaly in hydraulic data and detect leak events. Furthermore, a statistical study is conducted to estimate the leak locations’ area. Both flow and pressure data from different realistic water demands scenarios of the LeakDB benchmark are assessed. Experiment results corroborate the effectiveness and reliability of the proposed system for both data types. Detection sensitivity achieved 97% using pressure data and 100 % using flow data, with true leak zone identification for 95% of scenarios.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data Availability
The used LeakDB dataset is an open access benchmark available at Stelios et al. (2018). However, the system codes and model are confidential and addressed to the financiers as indicated in the acknowledgment and may only be provided with their acceptations.
References
Abiodun OI, Jantan A, Omolara AE, Dada KV, Mohamed NA, Arshad H (2018) State-of-the-art in artificial neural network applications: a survey. Heliyon 4(11):e00938
Ahmadullah R, Dongshik K (2016) Designing of hydraulically balanced water distribution network based on GIS and Epanet. Int J Adv Comput Sci Appl 7(2)
Banovec P, Domadenik P (2018) Defining economic level of losses in shadow: Identification of parameters and optimization framework. Multidiscipl Digital Publish Inst Proc 2(11):599
Bjerke M (2019) Leak detection in water distribution networks using gated recurrent neural networks. Maste's Thesis, NTNU
Chalapathy R, Chawla S (2019) Deep learning for anomaly detection: a survey. Preprint at http://arxiv.org/abs/1901.03407
Chen S, Brokhausen F, Wiesner P, Thamsen L, Cominola A (2021) Assessing the resilience of water distribution networks under different sensor network architectures and data sampling frequencies
Cheng J, Peng S, Cheng R, Wu X, Fang X (2022) Burst area identification of water supply network by improved DenseNet algorithm with attention mechanism. Water Resour Manage 1–18
Chuang W-Y, Tsai Y-L, Wang L-H (2019) Leak detection in water distribution pipes based on CNN with MEL frequency cepstral coefficients. In: Proceedings of the 2019 3rd International Conference on Innovation in Artificial Intelligence. p 83–86
Fujiwara O, Khang DB (1990) A two-phase decomposition method for optimal design of looped water distribution networks. Water Resour Res 26(4):539–549
Gwenzi W, Nyamadzawo G (2014) Hydrological impacts of urbanization and urban roof water harvesting in water-limited catchments: a review. Environ Process 1(4):573–593
Hawkins S, He H, Williams G, Baxter R (2002) Outlier detection using replicator neural networks. In: International Conference on Data Warehousing and Knowledge Discovery. Springer, p 170–180
Hu X, Han Y, Yu B, Geng Z, Fan J (2021) Novel leakage detection and water loss management of urban water supply network using multiscale neural networks. J Clean Prod 278:123611
Kammoun M, Kammoun A, Abid M (2021) Experiments based comparative evaluations of machine learning techniques for leak detection in water distribution systems. Water Supply
Kammoun M, Kammoun A, Abid M (2022) Leak detection methods in water distribution networks: a comparative survey on artificial intelligence applications. J Pipeline Syst Eng Pract 13(3):04022024
Kanakoudis V, Tsitsifli S, Cerk M, Banovec P, Samaras P, Zouboulis AI (2015) Basic principles of a DSS tool developed to prioritize NRW reduction measures in water pipe networks. Water Qual Exposure Health 7(1):39–51
Qi Z, Zheng F, Guo D, Zhang T, Shao Y, Yu T, Zhang K, Maier HR (2018) A comprehensive framework to evaluate hydraulic and water quality impacts of pipe breaks on water distribution systems. Water Resour Res 54(10):8174–8195
Rahimi M, Alghassi A, Ahsan M, Haider J (2020) Deep learning model for industrial leakage detection using acoustic emission signal. In: Informatics (vol. 7). Multidisciplinary Digital Publishing Institute, p 49
Reehuis E, Kruisselbrink J, Deutz A, Bäck T, Emmerich M (2011) Multiobjective optimization of water distribution networks using SMS-EMOA. Evolutionary Methods for Design, Optimisation and Control with Application to Industrial Problems (EUROGEN 2011), p 269–279
Sadeghioon AM, Metje N, Chapman D, Anthony C (2018) Water pipeline failure detection using distributed relative pressure and temperature measurements and anomaly detection algorithms. Urban Water J 15(4):287–295
Song H, Jiang Z, Men A, Yang B (2017) A hybrid semi-supervised anomaly detection model for high-dimensional data. Comput Intell Neurosci 2017
Stelios G, Vrachimis DGE, Marios S, Kyriakou M, Polycarpou M (2018) Leakdb: a benchmark dataset for leakage diagnosis in water distribution networks. In: WDSA/CCWI Joint Conference Proceedings (vol. 1)
Wang, X., Guo, G., Liu, S., Wu, Y., Xu, X., & Smith, K. (2020). Burst detection in district metering areas using deep learning method. Journal of Water Resources Planning and Management, 146(6), 04020031
Williams G, Baxter R, He H, Hawkins S, Gu L (2002) A comparative study of RNN for outlier detection in data mining. In: Proceedings of 2002 IEEE International Conference on Data Mining. IEEE, p 709–712
Wu Y, Liu S (2017) A review of data-driven approaches for burst detection in water distribution systems. Urban Water J 14(9):972–983
Acknowledgements
This research work was accomplished in collaboration between Sofia Technologies Company and CES-lab in the National Engineering School of Sfax. This project is carried out under the MOBIDOC scheme, funded by the EU through the EMORI program and managed by ANPR.
Funding
This work is funded by the EU through the EMORI program and Sofia Technologies Company, and managed by ANPR.
Author information
Authors and Affiliations
Contributions
All authors contributed to this work. The Deeplearning analysis approach was performed by Maryam KAMMOUN and Amina KAMMOUN. The first draft of the manuscript was written by Maryam KAMMOUN and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing Interests
Author Maryam KAMMOUN has received a research grant from EMORI program.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kammoun, M., Kammoun, A. & Abid, M. LSTM-AE-WLDL: Unsupervised LSTM Auto-Encoders for Leak Detection and Location in Water Distribution Networks. Water Resour Manage 37, 731–746 (2023). https://doi.org/10.1007/s11269-022-03397-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11269-022-03397-6