Detection of Water Leaks in Suburban Distribution Mains with Lift and Shift Vibro-Acoustic Sensors

Leaks in Water Distribution Networks (WDNs) account for a large proportion of Non-Revenue (NRW) utilities worldwide. Typically, leak is only confirmed once water surfaces, allowing the to be traced; however, high percentage leaks may never surface, incurring losses and costs utilities. Active Leak Detection (ALD) methods can used detect hidden leaks; success such highly dependent on available detection instrumentation experience operator. To aid both surfacing leaks, deployment vibro-acoustic sensors being increasingly explored by temporary structural health monitoring. In this paper, data were collected curated from range Lift Shift (L&S) sensor deployments across suburban Sydney. Time-frequency frequency-domain features generated assess performance suitability two state-of-the-art binary classification models detection. The results drawn extensive field sets are shown provide reliable outcomes, with accuracies at least 97% low false positive rates. Through use system, streamline their repair processes, effectively mitigating NRW reducing customer disruptions.