A Particle Filter Based Leak Detection Technique for Water Distribution Systems

Abstract

Leak detection and localization plays an important role in the efficient management of Water Distribution Systems (WDS), as it will help in reducing water wastage. Automation of WDS (with water quantity and quality sensors) has helped the water authorities around the world to get near real time data of important system parameters. Various models developed on the basis of Extended Kalman Filter (EKF), Nonlinear Kalman Filter etc. have been widely used for leak detection in pipelines. These models are applied on flow, pressure and acoustic signals from the system. But the main disadvantage of most of these models is that, they require the non-linear pipeline model to be linearized. In this paper, we propose the use of a Particle Filter (PF) based technique for the detection of leaks in water pipelines, and the developed model is applied to a real world network in Mandya (Karnataka, India).