Anomaly Detection using Machine Learning Models in Water Supply Systems

TEchMA 2025 – new frontiers in mechanical engineering International Conference on Technologies for the Wellbeing and Sustainable Manufacturing Solutions Anomaly Detection using Machine Learning Models in Water Supply Systems Ana Luís Sousa (a), (b)*, Eugénio Rocha(c), António Andrade-Campos (a), (b) (a) Department of Mechanical Engineering, TEMA, University of Aveiro; (b) – Intelligent Systems Associate Laboratory (LASI); (c) Department of Mathematics, CIDMA, University of Aveiro (a) Aveiro, Portugal; (b) Guimarães, Portugal; (c) Aveiro, Portugal (a) [email protected], [email protected]; (c) [email protected] * Corresponding author Abstract — Water loss remains a critical global concern, particularly within the present-day scarcity of water resources. This problem constitutes a significant challenge faced by water supply systems (WSS) utilities, as these deal with water leakage, which can persist undetected for extended periods of time and have a significant impact on the system efficiency. The occurrence of water leakage in these systems can range from 3% to over 50% depending on the level of system network maintenance performed, since it happens in pipe and/or junctions by uncontrolled actions [1]. Moreover, and according to the Portuguese regulator ERSAR in the RASARP 2024 [2], actual water leakage in Portugal in 2023 was 5.5 m3/(km day) for the bulk side, which corresponds to a loss of more than 21 billion m3/year. On the other hand, on the distribution side the value was 2.4 m3/(km day) representing 4.6 billion m3/year. To address this problem, several leakage management measures can be implemented, including preventive measures, detection and localization techniques, and repair initiatives. The detection and localization techniques comprise hardwareand software-based methods, which can integrate Machine Learning (ML) models and digital twins technologies. This data analytics integration can become a powerful tool in automated data analysis and hydraulic simulation, leading to significant advancements to a faster and more accurate leakage detection and localization. This work aims to present a novel sub-framework employing ML techniques to detect anomalies in pressure time series, as small discrepancies in the values may represent potential water leakage scenarios in the water system. This approach is implemented on a benchmark dataset, the BattLeDIM network [3], in which different ML-based models were evaluated and then compared with prior baseline results. Keywords — Water Leakage, Water Supply System, Machine Learning, Anomaly Detection, BattLeDIM Benchmark TOPIC 2) b.: Technologies for the Wellbeing - Innovative Technologies for Smart Cities”. ACKNOWLEDGMENTS This work is supported by the doctoral grant (Ref. 2023.02917.BDANA) financed by the Portuguese Foundation for Science and Technology (FCT), by the project UID/00481 Centre for Mechanical Technology and Automation (TEMA), and through the FEDER and Regional Operational Program of the Center Region (CENTRO2030) within project I-ReTiS-LeaksD&Op nº 17304 (CENTRO2030-FEDER-01177300) and through the Portuguese Foundation for Science and Technology (FCT) within project IReTiS-Leaks (2024.07270.IACDC). REFERENCES [1] R. Puust, Z. Kapelan, D. A. Savic, and T. Koppel. A review of methods for leakage management in pipe networks. Urban Water Journal, 7(1): 25–45, February 2010. Doi: 10.1080/15730621003610878 [2] ERSAR. Edições anuais do RASARP – Vol 1, 2024. https://www.ersar.pt/informacao-relevante-setor/ [3] Vrachimis, S. G., Eliades, D. G., Taormina, R., Kapelan, Z., Ostfeld, A., Liu, S., Kyriakou, M., Pavlou, P., Qiu, M., and Polycarpou, M. M. (2022). “Battle of the leakage detection and isolation methods.” Journal of Water Resources Planning and Management, 148(12), 04022068. Doi: 10.1061/(ASCE)WR.19435452.0001601