Leakage detection in water distribution networks using machine-learning strategies. Issue 3 (21st February 2023)

• Record Type: Journal Article
• Authors: Sousa, Diego Perdigão; Du, Rong; Mairton Barros da Silva Jr, José; Cavalcante, Charles Casimiro; Fischione, Carlo
• Is Part Of: Water Supply. Volume 23:Issue 3(2023)
• Page Start: 1115
• Page End: 1126
• Publication Date: 2023-02-21
• Subjects: clustering -- leakage detection -- machine-learning -- supervised learning -- unsupervised learning -- water distribution network
• DOI: 10.2166/ws.2023.054
• Languages: English
• ISSNs: 1606-9749

Abstract

Abstract: This work proposes a reliable leakage detection methodology for water distribution networks (WDNs) using machine-learning strategies. Our solution aims at detecting leakage in WDNs using efficient machine-learning strategies. We analyze pressure measurements from pumps in district metered areas (DMAs) in Stockholm, Sweden, where we consider a residential DMA of the water distribution network. Our proposed methodology uses learning strategies from unsupervised learning (K-means and cluster validation techniques), and supervised learning (learning vector quantization algorithms). The learning strategies we propose have low complexity, and the numerical experiments show the potential of using machine-learning strategies in leakage detection for monitored WDNs. Specifically, our experiments show that the proposed learning strategies are able to obtain correct classification rates up to 93.98%. HIGHLIGHTS: Leakage detection in water distribution networks using efficient machine-learning strategies. We analyze pressure measurements from pumps in district-metered areas in Stockholm, Sweden, where we consider a monitored subarea of the water distribution network. Our proposal can be applied to leakage detection scenarios where we have access to water pressure measurements at different points of the WDN.