•  
  •  
 
Information Sciences Letters

Information Sciences Letters

Abstract

Leaks in water distribution networks impose several impacts on economy, freshwater resources, water quality, health and safety. Fast leak detection and reparation is a key for lowering its negative impacts and associated costs with conventional detection techniques. This study has been used a pressure-based model to detect leaks events and its coordinates based on pressure and flow measurements. Pressure and flow data for systems that having leaks in their structure were analyzed and compared with data generated from non-leaking systems using EPANET software packages. An extension package of EPANET software (EpanetWaterGen) has been used as it has the advantage of its ability to better simulate leaks. The results show the ability of the model to detect leaks in a small and large water distribution networks with uncertainty level associated with low pressure change. The developed leak detection model utilizes pressure and flow sensors and enables the network managers and administrators to optimally place the sensors in a manner to increase efficiency and optimize cost. The system allows operators to detect leak location and volume of lost water, thus enabling a better and more efficient response to leaks, such that the network managers can address and respond to most urgent leaks and optimize the time end efforts of technical and maintenance personnel.

Share

COinS