As pipelines in water distribution networks (WDNs) age and deteriorate, pipe bursts have become increasingly frequent, wasting a significant amount of precious treated water. Deployment of real-time monitoring sensors has been advocated as a promising solution to ensure timely detection, identification, and response to pipe bursts. The detection and identification performance of the sensors relies heavily upon the strategic choices of sensor locations.

This paper focuses on the optimal sensor placement problem for the identification of pipe bursts. The problem involves selecting the optimal number and locations of sensors, such that maximum number of potential pipe bursts can be uniquely identified and localized. A robust representation approach is first proposed to describe the impacts of bursts by extracting the characteristic features of the pressure transients induced by the bursts, as modeled by hydraulic transient simulations. The extracted features are then summarized in the signature matrix, based on which two types of identification problems are then formulated: (1) when there is no constraint on the number of available sensors and the objective is to minimize the number of sensors required to identify all potential burst events and (2) when the number of available sensors is limited and the objective is to maximize the number of identified pipe bursts. The two problems are cast as minimum test cover (MTC) and maximum covering test (MCT) combinatorial optimization problems, which are subsequently formulated and solved as mixed integer linear programming (MILP). Moreover, identification tolerance is introduced into the MILP to account for data and modeling uncertainties. The proposed algorithm is applied to a medium-sized WDN. The results show that the incorporation of characteristic features into the signature matrix can improve the performance of burst identification, but sufficient accuracy of the extracted information is required to accrue the benefits.

Access the Conference Proceedings Paper on ASCE Library 

As pipelines in water distribution networks (WDNs) age and deteriorate, pipe bursts have become increasingly frequent, wasting a significant amount of precious treated water. Deployment of real-time monitoring sensors has been advocated as a promising solution to ensure timely detection, identification, and response to pipe bursts. The detection and identification performance of the sensors relies heavily upon the strategic choices of sensor locations.

This paper focuses on the optimal sensor placement problem for the identification of pipe bursts. The problem involves selecting the optimal number and locations of sensors, such that maximum number of potential pipe bursts can be uniquely identified and localized. A robust representation approach is first proposed to describe the impacts of bursts by extracting the characteristic features of the pressure transients induced by the bursts, as modeled by hydraulic transient simulations. The extracted features are then summarized in the signature matrix, based on which two types of identification problems are then formulated: (1) when there is no constraint on the number of available sensors and the objective is to minimize the number of sensors required to identify all potential burst events and (2) when the number of available sensors is limited and the objective is to maximize the number of identified pipe bursts. The two problems are cast as minimum test cover (MTC) and maximum covering test (MCT) combinatorial optimization problems, which are subsequently formulated and solved as mixed integer linear programming (MILP). Moreover, identification tolerance is introduced into the MILP to account for data and modeling uncertainties. The proposed algorithm is applied to a medium-sized WDN. The results show that the incorporation of characteristic features into the signature matrix can improve the performance of burst identification, but sufficient accuracy of the extracted information is required to accrue the benefits.

**[Access the Conference Proceedings Paper on ASCE Library](https://doi.org/10.1061/9780784482971.028)** 

Transient Model-based Sensor Placement for Burst Identification under Uncertainties in Water Distribution Networks

Created in 1999, the Environmental & Water Resources Institute (EWRI) is the recognized leader within ASCE for the integration of technical expertise and public policy in the planning, design, construction, and operation of environmentally sound and sustainable infrastructure impacting air, land and water resources.

[Paper Proceedings](https://doi.org/10.1061/9780784483466)

Conference DOI: https://doi.org/10.26226/m.6329c39b68fb0e00131f77d2