Pipe bursts in water distribution networks lead to considerable water loss and pose risks of bacteria and pollutant contamination. Pipe burst localisation methods help water service providers repair the burst pipes and restore water supply timely and efficiently. Although methods have been reported on burst detection and localisation, there is a lack of studies on accurate localisation of a burst within a potential district by accessible meters. To address this, a novel Burst Location Identification Framework by Fully-linear DenseNet (BLIFF) is proposed. In this framework, additional pressure meters are placed at limited, optimised places for a short period (minutes to hours) to monitor system behaviour after the burst. The fully-linear DenseNet (FL-DenseNet) newly developed in this study modifies the state-of-the-art deep learning algorithm to effectively extract features in the limited pressure signals for accurate burst localisation. BLIFF was tested on a benchmark network with different parameter settings, which showed that accurate burst localisation results can be achieved even with high model uncertainties. The framework was also applied to a real-life network, in which 57 of the total 58 synthetic bursts in the potential burst district were correctly located when the top five most possible pipes are considered and among them, 37 were successfully located when considering only the top one. Only one failed because of the very small pipe diameter and remote location. Comparisons with DenseNet and the traditional fully linear neural network demonstrate that the framework can effectively narrow the potential burst district to one or several pipes with good robustness and applicability. Codes are available at https://github.com/wizard1203/waternn.

Pipe bursts in water distribution networks cause considerable water losses and lead to potential environmental hazards. Effective burst detection methods enable water companies to repair broken pipes in a timely manner and minimize damage and disruption. A data‐driven detection method is developed and proven for real‐time leak and burst detection in water distribution networks. The method uses a unique integration of disturbance extraction and isolation forest techniques to enable detection of subtle burst signals from normally noisy pressure data. Verification and validation progress from synthetic data to the application to real‐life data from large‐scale open networks. The method is shown to generate comparable levels of detection, with low false positive rates, to other published techniques but without the need to introduce closed valves and expensive flow meters of district meter area structures and the loss of resilience these cause. The method offers the potential to effectively detect bursts from pressure data alone, helping to better manage the distribution of increasingly stressed water resources through aging pipe networks.

Real‐time simulation of water distribution systems (WDSs) has been applied to water resource problems ranging from engineering optimization design and operation planning and management. However, accurate implementation of real‐time simulation of WDS is still a challenging task due to the limited knowledge of the large amount of varying nodal demands and pipe characteristics over the entire network system. This paper presents a self‐adaptive calibration method based on Kalman filter (KF) that takes advantage of the long‐term monitoring data for dual calibration of nodal demands and pipe roughness. Inferential measurements are introduced to avoid linear assumptions of the WDS system and link the hydraulics of WDS with KF. Hence, it is able to employ KF to solve the nonlinear problems in looped water distribution network. By assimilating the long‐term monitoring data and adapting the calibrated parameters to various operating conditions, the framework can reduce the uncertainties caused by measurement errors and quantify the uncertainties by covariance matrixes. In addition, the presented method can help to identify abnormal WDS conditions. Three case studies have been conducted to illustrate the validity of the proposed method and its applications. The results have shown that the proposed framework is reliable and effective in practical applications.