Using electrical resistance networks to enhance performance assessments of water distribution networks

Pressurized fluid-distribution networks are strategic elements of civil infrastructure. In the case of fresh-water distribution networks, where leaks are common and where real performance is unknown, advanced sensor-based diagnostic methodologies have the potential to provide enhanced management support. A structural identification methodology that has been used successfully for bridges is adapted to a study of a diagnostic methodology for leak detection in water distribution networks. This methodology is based on error-domain model falsification. Using analogies between Ohm’s law and the Hazen-Williams relationship, an electric network model is built to show similarities with hydraulic networks. The first step is to compare simulated values to show similarity of the network behaviour and then to show similarities throughout the diagnostic process. This study establishes the similarity of the behaviour for hydraulic and electrical networks. It also shows that results obtained with electrical networks are relevant for performance assessments of hydraulic networks. These results present to practicality of studying generic electrical networks of varying size and shape to illustrate the usefulness of the diagnostic methodology for general cases.