ABSTRACT
Managing pipe breaks in water supply networks has been a challenge for water utilities around the world. In order to transform from reactive to proactive management of pipe breaks, South Australia Water Corporation (SA Water) has invested more than 4 million dollars to set up a Smart Water Network (SWN), in the central business district (CBD) of Adelaide. The network includes 305 permanent accelerometers that continuously monitor the ‘noise level’ in the water network. When a new pipe crack forms, the baseline noise level of the continuous measurements increases (relative to ‘normal’ network and other ‘environmental’ diurnal noise variations). This research develops a technique for automated pipe crack detection by detecting baseline changes in continuous noise measurements from the accelerometers. The developed technique is validated by field data from Adelaide SWN, and results confirm that the developed technique can achieve cost-effective early pipe crack detection and uncontrolled main break prevention.
Acknowledgements
The research presented in this paper has been supported by the South Australian Water Corporation through a collaborative research project with the University of Adelaide (project ID: UA183151), and the Australian Research Council through a Linkage Project (LP180100569). The authors thank All Water staff Mr Goran Pazeski-Nikoloski, Mr Matthew Maresca and Mr Adrian Cavallaro for their support in the field investigation.
Disclosure statement
No potential conflict of interest was reported by the author(s).