The optimal deployment of pressure monitoring sensors in water distribution networks is crucial for leak detection, network calibration, and system diagnostics. Water utilities face increasing pressure to reduce non-revenue water losses while continuing to improve service quality under budget constraints, thus making the strategic deployment of sensors a critical priority. However, traditional optimization approaches come with various disadvantages including high computational complexity, limited scalability, or dependence on uncertain preliminary parameter estimates. This paper addresses these shortcomings by proposing an innovative integrated framework that balances topological and hydraulic considerations, and applying a flexible metric blending approach to enable robust sensor positioning across networks that differ in scales and topologies. The methodology has been validated through three case studies: a theoretical reference grid, an urban district network, and a large-scale multisource irrigation system. The results prove the methodology to be consistently effective in identifying optimal sensor configurations across all test cases.
Optimal Sensor Placement in Water Distribution Networks: An Integrated Approach for Leak Detection and Network Monitoring
Francesco Di Menna
;Giorgia Diglio;Nicola Fontana;Gustavo Marini
2026-01-01
Abstract
The optimal deployment of pressure monitoring sensors in water distribution networks is crucial for leak detection, network calibration, and system diagnostics. Water utilities face increasing pressure to reduce non-revenue water losses while continuing to improve service quality under budget constraints, thus making the strategic deployment of sensors a critical priority. However, traditional optimization approaches come with various disadvantages including high computational complexity, limited scalability, or dependence on uncertain preliminary parameter estimates. This paper addresses these shortcomings by proposing an innovative integrated framework that balances topological and hydraulic considerations, and applying a flexible metric blending approach to enable robust sensor positioning across networks that differ in scales and topologies. The methodology has been validated through three case studies: a theoretical reference grid, an urban district network, and a large-scale multisource irrigation system. The results prove the methodology to be consistently effective in identifying optimal sensor configurations across all test cases.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


