| description abstract | Water main breaks in city areas can cause significant economic loss, interrupt traffic, and result in negative publicity for utilities. The South Australian Water Corporation (SA Water) has taken a proactive approach to reduce water main breaks in the Adelaide, Australia, central business district (CBD) by detecting developing pipe cracks/leaks using distributed acoustic sensors (accelerometers) with customized analytics and internet of things (IoT) technologies. This paper reports the findings and the outcomes of this initiative 24 months after the first commissioning of the continuous leak detection system. The acoustic data analytics team (comprising the authors) has successfully interpreted acoustic data from numerous cracked pipes in the Adelaide CBD, with approximately 55% of all developing leaks associated with cracked pipes being able to be detected before uncontrolled failure (i.e., before unplanned interruption of customers and/or third party economic damage). Furthermore, the localization and repair of specific cracks (sometimes through sections of corroded cast iron pipe walls) can help justify extending the service life of the water main as points of weakness (faults) are specifically rectified. However, only 45% of the developing cracks have been able to be repaired, after detection, under controlled conditions (i.e., before the planned interruption of customers) due to practical limitations affecting operational responses. The results have confirmed the additional insight that can be obtained by continuously collecting acoustic data in terms of distinguishing environmental and crack noises as well as circumferential from longitudinal cracks. | |
