Investigating the Suitability of Concept Drift Detection for Detecting Leakages in Water Distribution Networks
Valerie Vaquet, Fabian Hinder, Barbara Hammer
TL;DR
Leakages in water distribution networks are framed as concept drift in streaming data, and the authors compare model-loss-based and distribution-based drift detectors for real-time leakage detection using limited pressure data. With L-Town simulations, they show that distribution-based detectors, when paired with windowing strategies to remove temporal patterns, detect leaks of all sizes with practical delays, while model-loss-based methods mainly detect larger leaks due to model-generalization effects. A first drift-detection-based leakage localization approach is proposed, relying on sensor-level drift signals to identify proximity to the leak. The work demonstrates the potential of drift-aware monitoring for critical infrastructure and outlines directions for improving leakage localization and applying the approach to other domains.
Abstract
Leakages are a major risk in water distribution networks as they cause water loss and increase contamination risks. Leakage detection is a difficult task due to the complex dynamics of water distribution networks. In particular, small leakages are hard to detect. From a machine-learning perspective, leakages can be modeled as concept drift. Thus, a wide variety of drift detection schemes seems to be a suitable choice for detecting leakages. In this work, we explore the potential of model-loss-based and distribution-based drift detection methods to tackle leakage detection. We additionally discuss the issue of temporal dependencies in the data and propose a way to cope with it when applying distribution-based detection. We evaluate different methods systematically for leakages of different sizes and detection times. Additionally, we propose a first drift-detection-based technique for localizing leakages.