Analysis of the impact of leaks and valve maneuvers on pressure transient data in water distribution networks

Abstract

Swift detection and repair of leaks in water distribution networks are essential to ensure the operational efficiency of the system. It’s not viable, both financially and technically, to scour the entire extension of the system to look for leaks, therefore it’s important to narrow down areas prone to leakage. Hydraulic techniques are useful to that end, since hydraulic data can be used to identify anomalies and guide efforts on field campaigns. Among these techniques, those based on the transient state are used to infer about the actual condition of the system by causing controlled disturbances, such as closing a valve, as to reveal anomalies like leaks, deposits or illegal connections. In this paper, the software OpenFlows Hammer, was used to model the transient state and generate the transient state data. This data was then analyzed based on signal theory. The results show that it’s possible to identify the presence of leaks, even minor ones, and their location in relation to the monitoring point. However, further studies are needed to better establish metrics relevant to characterize the behavior of the data in the frequency domain.