A GIS-Based Tool for Leak Detection in Water Distribution Networks (WDNs)

Leakages in the water distribution networks (WDNs) are real problems for water utility companies. In the United Arab Emirates (UAE), 10% of the country’s water is lost due to leaks. Typically, leaks in WDNs are detected manually through physical inspection, but recently many technologies have been used for leak detection, for instance, radars, humidity sensors, and infrared cameras. However, timely leak detection and location identification has been a challenge. In this paper, a new method for robust leak detection is presented. Normally, water leakage causes drops in flow, pressure, and/or chlorine around leak locations. Our method uses simultaneous drops in these parameters to determine potential water leakage locations in the WDN using geographic information system (GIS) techniques. The method starts by creating GIS layers from the hourly values of these three parameters using a test database provided by Sharjah Electricity, Water and Gas Authority (SEWA). These layers are then analyzed where groups of locations with dropped values of each of the parameters are identified and overlaid with each other, which may result with one or more locations or with no locations at all. In the case where there are no overlaying locations between these two parameters the conclusion would be there are no leaks in the WDN. Our tool then examines if the locations with drops in flow and pressure also encountered drops in chlorine. This may also result with one or more locations or no overlaying locations at all. If there are no overlaying locations at this stage of the analysis that means no leaks. However, if overlaid locations are identified, then these are considered potential leak locations. This method as demonstrated on the WDN of SEWA can help water utility companies in the timely detection of leaks saving money, time, and effort.