Enhanced Water Leak Detection with Convolutional Neural Networks and One-Class Support Vector Machine

2025-11-10

AI Summary

Water loss due to leaks in water distribution networks (WDNs) poses a critical challenge for sustainable water management. To address this, this paper proposes an unsupervised leakage detection method requiring only normal-operation pressure time-series data from network nodes. The approach employs a convolutional neural network (CNN) to automatically extract topology-aware features from the pressure sequences, followed by a one-class support vector machine (One-Class SVM) for anomaly detection. Its key innovation lies in the deep integration of CNN-based feature learning with One-Class SVM—marking the first such fusion—thereby circumventing the scarcity of labeled leakage samples and substantially enhancing practical deployability and generalization capability. Experimental evaluation on the Modena WDN simulation dataset demonstrates superior performance: the method achieves lower false-negative rates and higher detection accuracy compared to state-of-the-art data-driven approaches, confirming its effectiveness and robustness.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Learning & Optimization for CV

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Web Mining and Content Analysis: Web data integration and cleaning

Abstract

Water is a critical resource that must be managed efficiently. However, a substantial amount of water is lost each year due to leaks in Water Distribution Networks (WDNs). This underscores the need for reliable and effective leak detection and localization systems. In recent years, various solutions have been proposed, with data-driven approaches gaining increasing attention due to their superior performance. In this paper, we propose a new method for leak detection. The method is based on water pressure measurements acquired at a series of nodes of a WDN. Our technique is a fully data-driven solution that makes only use of the knowledge of the WDN topology, and a series of pressure data acquisitions obtained in absence of leaks. The proposed solution is based on an feature extractor and a one-class Support Vector Machines (SVM) trained on no-leak data, so that leaks are detected as anomalies. The results achieved on a simulate dataset using the Modena WDN demonstrate that the proposed solution outperforms recent methods for leak detection.

Problem

• Detecting water leaks in distribution networks using pressure data analysis
• Developing data-driven leak detection with convolutional neural networks
• Identifying leaks as anomalies through one-class SVM classification

Innovation

• Uses convolutional neural networks for feature extraction
• Applies one-class SVM trained on no-leak data
• Detects water leaks as anomalies in pressure measurements