Article ID Journal Published Year Pages File Type
246255 Automation in Construction 2016 9 Pages PDF
Abstract

•Multitier method for detecting and locating leaks in water distribution networks•GPR is utilized to detect pipe location and IR to detect leaks.•An image segmentation algorithm is developed to estimate leakage area and its centroid.•The method was successfully applied with small margin of error (2.9–5.6%).•The developed method can aid operators in detecting and locating water leaks.

This paper presents newly developed method for detecting and locating leaks in water distribution networks utilizing two detection techniques; ground penetrating radar (GPR) and infrared photography (IR). The experimental work and field investigation were carried out over 2 years in three locations in City of Doha, Qatar to capture 115 IR image frames and 23 GPR image frames. Firstly, GPR technology is utilized to accurately define location of buried pipes. After locating these pipes, IR images are collected for simulated and actual leaks. The developed algorithm segments each image into leakage and non-leakage areas and the centroid of each leakage is calculated using Green's theorem. Subsequently, GPR images are introduced as a second layer and overlaid with IR images to compare pipes location with leak location. The method was successfully applied to detect simulated and actual leaks in summer and winter seasons with small margin of error (2.9–5.6%) in estimating leakage areas. When examining the investigated four operating conditions, it was found that the developed method can predict leaks in a more reliable way if the camera height is 2 m and the speed is 1.65 m/s in both simulated and actual leaks. The newly developed method is robust and can aid operators and city engineers in detecting and locating water leaks with high accuracy.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slide

Related Topics
Physical Sciences and Engineering Engineering Civil and Structural Engineering
Authors
, , , , ,