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.

Resilience has been increasingly pursued in the management of water distribution systems (WDSs) such that a system can adapt to and rapidly recover from potential failures in face of a deep uncertain and unpredictable future. Topology has been assumed to have a great impact on resilience of WDSs, and is the basis of many studies on assessing and building resilience. However, this fundamental assumption has not been justified and requires investigation. To address this, a novel framework for mapping between resilience performance and network topological attributes is proposed. It is applied to WDSs here but can be adaptable to other network systems. In the framework, resilience is comprehensively assessed using stress-strain tests which measure system performance on six metrics corresponding to system resistance, absorption and restoration capacities. Six key topological attributes of WDSs (connectivity, efficiency, centrality, diversity, robustness and modularity) are studied by mathematical abstraction of WDSs as graphs and measured by eight statistical metrics in graph theory. The interplay between resilience and topological attributes is revealed by the correlations between their corresponding metrics, based on 85 WDSs with different sizes and topological features. Further, network variants from a single WDS are generated to uncover the value of topological attribute metrics in guiding the extension/rehabilitation design of WDSs towards resilience. Results show that only certain aspects of resilience performance, i.e. spatial and temporal scales of failure impacts, are strongly influenced by some (not all) topological attributes, i.e. network connectivity, efficiency, modularity and centrality. Metrics for describing the topological attributes of WDSs need to be carefully selected; for example, clustering coefficient is found to be weakly correlated with resilience performance compared to other metrics of network connectivity (due to the grid-like structures of WDSs). Topological attribute metrics alone are not sufficient to guide the design of resilient WDSs and key details such as the location of water sources also need to be considered.

Urban flooding has become a global issue due to climate change, urbanization and limitation in the capacity of urban drainage infrastructures. To tackle the growing threats, it is crucial to understand urban surface flood resilience, i.e., how urban drainage catchments can resist against and recover from flooding. This study proposes a grid cell based resilience metric to assess urban surface flood resilience at the urban drainage catchment scale. The new metric is defined as the ratio of the number of unflooded grid cells to the total grid cell number in an urban drainage catchment. A two-dimensional Cellular Automata based model CADDIES is used to simulate urban surface flooding. This methodology is demonstrated using a case study in Dalian, China, which is divided into 31 urban drainage catchments for flood resilience analysis. Results show the high resolution resilience assessment identifies vulnerable catchments and helps develop effective adaptation strategies to enhance urban surface flood resilience. Comparison of the new metric with an existing metric reveals that new metric has the advantage of fully reflecting the changing process of system performance. Effectiveness of adaptation strategies for enhancing urban surface flood resilience is discussed for different catchments. This study provides a new way to characterize urban flood resilience and an in-depth understanding of flood resilience for urban drainage catchments of different characteristics, and thus help develop effective intervention strategies for sustainable sponge city development.

The main focus of this paper is the novel use of Artificial Intelligence (AI) in natural disaster, more specifically flooding, to improve flood resilience and preparedness. Different types of flood have varying consequences and are followed by a specific pattern. For example, a flash flood can be a result of snow or ice melt and can occur in specific geographic places and certain season. The motivation behind this research has been raised from the Building Resilience into Risk Management (BRIM) project, looking at resilience in water systems. This research uses the application of the state-of-the-art techniques i.e., AI, more specifically Machin Learning (ML) approaches on big data, collected from previous flood events to learn from the past to extract patterns and information and understand flood behaviours in order to improve resilience, prevent damage, and save lives. In this paper, various ML models have been developed and evaluated for classifying floods, i.e., flash flood, lakeshore flood, etc. using current information i.e., weather forecast in different locations. The analytical results show that the Random Forest technique provides the highest accuracy of classification, followed by J48 decision tree and Lazy methods. The classification results can lead to better decision-making on what measures can be taken for prevention and preparedness and thus improve flood resilience.