Benchmark Water Distribution Networks Research Articles

A stepwise fast leakage localization method applying the strategy of dynamic area narrowing down for large-scale water distribution network

ABSTRACT Urban water distribution networks (WDNs) are facing serious leakage problems. As cities expand, the leakage localization burden on large WDNs gradually increases. Although methods have been widely researched, there is a lack of studies and successful applications for large-scale WDNs. To deal with this, a stepwise fast leakage localization Method, SFLLM, utilizing the `dynamic area narrowing down (DAND)' strategy and coupled leakage features (CLF) is proposed. The SFLLM includes a fast-and-dynamic stage using the DAND strategy to reduce the potential leakage area and an accurate localization stage. Only partial representative candidate locations are required to simulate leakages by DAND, and meantime CLF is used to analyze the leakage similarities so that the localization accuracy and efficiency can be improved. SFLLM was tested on a benchmark WDN, saving more than 88% of simulation by DAND strategy and achieving localization in 11 seconds. The results also proved the enhanced performance of CLF in ensuring the stability of the accuracy against various types of uncertainties that may occur in real WDNs. Moreover, three real burst leaks in an actual large-scale WDN were localized within 205 m in about 22 minutes by SFLLM, showing the method's reliable applicability in guiding field leak exploration.

Simultaneous Pipe Leak Detection and Localization Using Attention-Based Deep Learning Autoencoder

Water distribution networks are often susceptible to pipeline leaks caused by mechanical damages, natural hazards, corrosion, and other factors. This paper focuses on the detection of leaks in water distribution networks (WDN) using a data-driven approach based on machine learning. A hybrid autoencoder neural network (AE) is developed, which utilizes unsupervised learning to address the issue of unbalanced data (as anomalies are rare events). The AE consists of a 3DCNN encoder, a ConvLSTM decoder, and a ConvLSTM future predictor, making the anomaly detection robust. Additionally, spatial and temporal attention mechanisms are employed to enhance leak localization. The AE first learns the expected behavior and subsequently detects leaks by identifying deviations from this expected behavior. To evaluate the performance of the proposed method, the Water Network Tool for Resilience (WNTR) simulator is utilized to generate water pressure and flow rate data in a water supply network. Various conditions, such as fluctuating water demands, data noise, and the presence of leaks, are considered using the pressure-driven demand (PDD) method. Datasets with and without pipe leaks are obtained, where the AE is trained using the dataset without leaks and tested using the dataset with simulated pipe leaks. The results, based on a benchmark WDN and a confusion matrix analysis, demonstrate that the proposed method successfully identifies leaks in 96% of cases and a false positive rate of 4% compared to two baselines: a multichannel CNN encoder with LSTM decoder (MC-CNN-LSTM) and a random forest and model based on supervised learning with a false positive rate of 8% and 15%, respectively. Furthermore, a real case study demonstrates the applicability of the developed model for leak detection in the operational conditions of water supply networks using inline sensor data.

Leakage detection in water distribution networks via 1D CNN deep autoencoder for multivariate SCADA data

Leakages are undesirable events in water distribution networks (WDNs) with mostly unknown incident time making it extremely difficult to accurately label leaks in distribution systems. This study proposes a one-dimensional convolutional neural network (1D CNN) deep autoencoder (AE) that is trained in a semi-supervised fashion to detect and localize leaks using multivariate timeseries data in a bid to mitigate the adverse impact of random noise. A leakage identification framework based on Pruned Exact Linear Time (PELT) multiple change point detector and alarm correlation is proposed. Additionally, a novel localization framework is proposed based on the relative percentage deviation of 1D CNN deep AE reconstruction error of pressure sensors in a district metering area (DMA). The leak localization framework outputs a sensor priority list based on the proximity of the pressure sensor to the leaky pipe thereby significantly reducing the search area for a leakage. The proposed leakage identification and localization framework is validated on a benchmark WDN, the L-TOWN. The results indicate the ability of the proposed method to identify 16 of the 19 leaks in 2019 and accurately localize 13 of the 16 identified leaks within the maximum allowable leak search radius of 300 m without recourse to a hydraulic model. Minimal identification lag was recorded for abrupt leakages whiles incipient leaks require longer identification time due to gradual leak growth rate.

A Bayesian Approach for Active Fault Isolation With an Application to Leakage Localization in Water Distribution Networks

This article proposes an active fault isolation method for application to water distribution networks (WDNs) to localize leaks. The method relies on the classification of observed outputs to a discrete set of hypothetical faults. Due to parametric uncertainties, the outputs are random vectors that follow unknown probability distribution functions (PDFs). The output PDFs corresponding to the considered faults are approximated using smooth kernel density estimation (SKDE). They are used to calculate the posterior probability of each hypothesis, given the observed outputs, by applying Bayes’ rule. The difficulty to classify observed outputs to the right fault comes from the overlap between output PDFs. An active algorithm is introduced that proactively minimizes the joint overlap between the output PDFs by designing optimal control inputs. Due to physical limitations on control inputs and depending on the intensity of uncertainties, full separation, and hence fault isolation, cannot be guaranteed for a single observed output. Therefore, subsequent observations are used in an iterative framework, where the posterior probabilities of the previous time step serve as the prior probabilities for the next time step. The method is applied to locate leaks in a benchmark WDN for different levels of uncertainty in customer water demand and leakage magnitude. Improvements in the performance are observed in comparison to the best considered passive method from literature.

Chaotic differential evolution algorithms for optimal design of water distribution networks

ABSTRACT Recently, coupling chaos theory with evolutionary algorithms (EAs) elevated further scope for improving EAs’ performance efficiency. In this view, the present study emphasizes investigating the influence of chaotic force on convergence properties of differential evolution (DE) algorithm in designing water distribution networks (WDNs). To this end, two novel chaos-directed DE models, Chaotic-DE and Chaotic-Fm-DE, are proposed. The Chaotic-DE model is formulated to enhance DE’s searchability and faster convergence by replacing every random phenomenon with a chaotic force. The Chaotic-Fm-DE model with a dynamic, chaotic mutation factor is developed to improve DE’s exploitation behavior. Essentially, these models differ from the previous chaos-directed EA models in how chaos ergodicity is simulated in DE mechanism. A novel scheme of non-sequential approach is used for this purpose. Further, their respective elitist models are formulated to promote the search in promising areas. Importantly, the elitism scheme developed saves the elite trial vectors to pass through the next generations. The results of proposed algorithms validated on five (new and rehabilitated) benchmark WDNs (whose dimensions vary from 8 to 454) demonstrate the enhanced search behavior of the chaotic models with solution precision and remarkable reduction in computational effort over the non-chaotic ones.

Leak detection in water distribution networks using deep learning

ABSTRACT Two deep learning algorithms, namely, Convolutional Neural Network (CNN) and Recurrent Neural Network- Long Short Term Memory (LSTM), were used to classify the water distribution networks (WDN) as leaky or non-leaky. LeakDB dataset was employed to generate different leakage scenarios for Net 1 and Hanoi benchmark WDN. Three cases, (a) incipient leaks, (b) abrupt leaks, and (c) mixed leak situations, are employed for pressure and flow conditions. A total of 1000 scenarios have been employed, 80% for training and 20% for testing. Seven metrics for analyzing the performance of CNN and LSTM are training accuracy, testing accuracy, total accuracy, true positive rate, false positive rate, false negative rate & area under curve. The results obtained are compared with those of Kammoun, et al. (2021). CNN is performing slightly better than LSTM in several metrics for most scenarios. However, both CNN and LSTM performed most of the time with better accuracy than those used by Kammoun et al. (2021). Leak detection accuracy is in the range of 90.56–98.23 % for Net1 WDN, whereas it is 49–96.55 % for Hanoi WDN.

A Chaotic Sobol Sequence-based multi-objective evolutionary algorithm for optimal design and expansion of water networks

The design of a water distribution network (WDN) is an optimization problem that is computationally challenging with conflicting objectives. This study offers an enhanced Chaotic Sobol Sequence-based Multi-Objective Self-Adaptive Differential Evolution (CS-MOSADE) algorithm for multi-objective WDN design. The CS-MOSADE algorithm was tested on two benchmark WDNs, and a real WDN. Optimization results indicate that the CS-MOSADE algorithm converged two to three times faster than the MOSADE and NSGA-IIalgorithms and led to better output in terms of even distribution of solutions and convergence towards the true Pareto-optimal front. Smaller spacing metric indicated better uniformity in the obtained solutions; and larger hyper-area and coverage function values depicted better convergence towards the true Pareto-optimal front for the CS-MOSADE algorithm compared to the other algorithms. The CS-MOSADE algorithm was then applied to solve a WDN expansion problem for optimal pump scheduling and minimization of Life Cycle Cost, maximization of reliability and minimization of Green House Gas (GHG) emissions. A significant reduction in GHG emissions of 2.17 x 106 kg was achieved at an additional cost of $0.55 x 107 when optimal pump scheduling was incorporated in the model of the real WDN over service life of 50 years.

State Estimation and Leakage Detection in Water Distribution Networks Using Constrained Zonotopes

State estimation in water distribution networks (WDNs) is a challenging task due to the scarcity of measurements and the presence of several modeling uncertainties. In the literature, pseudo-measurements are often required to obtain an observable model for state estimation, which are highly uncertain and mostly represented by bounded sets. This article proposes two new methods for set-based state estimation in WDNs, considering unknown-but-bounded measurement and model uncertainties to calculate bounds on the system states. A new interval method is proposed based on iterative computation of tight enclosures of the nonlinear head-loss functions, and bounding the solution of the algebraic equations using rescaling. This method, referred to as iterative slope approximation and rescaling (ISAR), is mildly more conservative than the iterative hydraulic interval state estimation (IHISE) method previously proposed by the authors, but significantly more computationally efficient. However, since intervals are not capable of capturing the dependencies between state variables, we propose the use of constrained zonotopes (CZs) as an additional step to both IHISE and ISAR. This yields two new algorithms for set-based state estimation of WDNs, capable of capturing the dependencies between hydraulic states, which result in sets with significantly smaller volumes than intervals. The benefits of CZs are also highlighted when the new enclosures are used for leakage detection. These provide higher leakage detection rates compared to intervals, as demonstrated in two case studies using benchmark WDNs.

Dynamic segment criticality analysis: A precursor to scheduling of maintenance routines in water distribution networks

This study proposes a framework for dynamic segment criticality analysis and its implications on scheduling of some reactive maintenance routines in water distribution networks (WDNs). The framework harmonises multilayer networks which are capable of ascertaining the dynamics of WDNs in discrete time and a novel hybrid segment criticality measure (SCM) to achieve dynamic segment criticality analysis. The novel hybrid SCM utilises reachability theory from complex networks, which accounts for multiple sources, cyclic paths in WDNs, and demand shortfall attributed to segment isolation is computed using Pressure Driven Analysis (PDA) courtesy EPANET. Using a benchmark WDN, North Marin Water District Network, as a case study we have demonstrated the efficacy of the proposed framework and its ability to capture the dynamics of WDNs and evaluate the criticality of any segment given any temporal window. The results indicate the framework accurately evaluates the criticality of all segments within any temporal window and establishes the fact that segment criticality varies with time. Based on the results of this study reactive maintenance routines could be scheduled during temporal periods where the segment criticality measure is minimal to avoid excessive service disruption and customer displeasure. This study supports managers of water utilities with a decision support system to accurately schedule maintenance routines and evaluate segment criticality.

Improved MOSADE algorithm incorporating Sobol sequences for multi-objective design of Water Distribution Networks

Design of Water Distribution Networks (WDNs) is a tremendously hard optimization problem, and consideration of reliability further adds to the complexity, which may involve huge computational effort. As several past studies stressed that Evolutionary Algorithms (EAs) could be efficient tools for WDN design, so this study, presents an effective methodology, Multi-Objective Self Adaptive Differential Evolution (MOSADE) algorithm using Sobol sequences for random number generation, termed as (S-MOSADE) for WDN design. The efficacy of the S-MOSADE framework is evaluated by application on a few benchmark WDNs, by considering cost minimization and mechanical reliability maximization and comparing the results with that of NSGA-II algorithm. The results illustrated that S-MOSADE algorithm leads to a better Pareto-optimal front than NSGA-II with respect to uniformly spaced and wide range of non-dominated solutions, and converges faster as compared to other algorithms. To further reduce the computational burden, minimizing the cost and maximizing the network resilience is carried out to generate the initial population for S-MOSADE algorithm. This has reduced the computational burden by almost three times as compared to random initialization of population, thus saving a lot of computational time. The study concludes that the proposed S-MOSADE algorithm with the strategy of solutions initialized with minimum cost and maximum network resilience could be used effectively for speeding-up the multi-objective design of WDNs.

Optimization of Water Distribution Networks Using Genetic Algorithm Based SOP–WDN Program

Water distribution networks are vital hydraulic infrastructures, essential for providing consumers with sufficient water of appropriate quality. The cost of construction, operation, and maintenance of such networks is extremely large. The problem of optimization of a water distribution network is governed by the type of water distribution network and the size of pipelines placed in the distribution network. This problem of optimal diameter allocation of pipes in a distribution network has been heavily researched over the past few decades. This study describes the development of an algorithm, ‘Smart Optimization Program for Water Distribution Networks’ (SOP–WDN), which applies genetic algorithm to the problem of the least-cost design of water distribution networks. SOP–WDN demonstrates the application of an evolutionary optimization technique, i.e., genetic algorithm, linked with a hydraulic simulation solver EPANET, for the optimal design of water distribution networks. The developed algorithm was applied to three benchmark water distribution network optimization problems and produced consistently good results. SOP–WDN can be utilized as a tool for guiding engineers during the design and rehabilitation of water distribution pipelines.

An investigation on the performance of different reliability criteria for design of water distribution networks

ABSTRACT Due to the high construction costs of a water distribution network, the least cost design with the most reliability is of great importance. In this paper, a survey was conducted on the performance of different reliability metrics including nodal pressure (PIP), pipe velocity (PIV), pressure difference (ΔH), combined index of pipe velocity and nodal pressure (PI) as well as the network resilience index (NRI), minimum surplus head (MSH) and entropy reliability index (ERI) using NSGA-II optimization algorithm and EPANET hydraulic solver. Reliability metrics were tested on three benchmark water distribution networks: Hanoi, Pescara and Modena and the results were analyzed in three scenarios of 5%, 10%, and 20% pipe failure. Experimental results showed that NRI generally outperforms other metrics and the PIP index is in the second score. These metrics increase the average of network pressure and thus provide higher reliability in the situation of unexpected events like pipe breaks.

A systematic framework for dynamic nodal vulnerability assessment of water distribution networks based on multilayer networks

Nodal demands vary throughout the day, as such any vulnerability analysis based on static networks, which considers daily average demands cannot realistically represent the criticality of nodes in the network. This study presents a systematic framework, which couples multilayer networks, structural reducibility and a Demand Adjusted Vulnerability Measure for dynamic nodal vulnerability assessment of water distribution networks (WDNs) under extended period simulation. Within this framework, we present the novel idea of characterizing the dynamics of WDNs with multi-slice networks, which captures the state of the network within a predefined temporal window taking into consideration the directional flow in pipes and the operational status of pumps, valves etc. Using a benchmark WDN, Net 3, as a case study we have demonstrated the importance of demand variations and operational status of various components, no matter how minuscule their operational time, on nodal vulnerability assessment in WDNs. The results indicated that the framework evaluates the criticality of all types of nodes, even intermediary nodes with zero base demand, within any temporal window much more realistically than conventional vulnerability analysis methods based on single (static) networks. Structural reducibility unearthed correlations between the operational status of source nodes and pumps on the general dynamics of the distribution system. The multilayer framework opens a new frontier in vulnerability analysis of WDNs and could serve as a tool for stakeholders in accessing node criticality, impact of various failure scenarios and optimal scheduling of maintenance routines.

Performance-based multi-objective design and expansion of water distribution networks considering life cycle costs and future demands

Abstract Designing the Water Distribution Networks (WDNs) consists of finding out pipe sizes such that the demands are satisfied and the desired performance levels are achieved at minimum cost. However, WDNs are subject to many future changes such as an increase (or decrease) in demand due to population change and migration, changes in water availability due to seasonal and climatic change, etc. Thus, the capacity expansion of WDNs needs to be performed such that the cost of interventions made is minimum while satisfying the demand and performance requirements at various time periods. Therefore, the current study proposed a Dynamic Programming (DP) framework for capacity expansion of WDNs and solved using Multi-Objective Self Adaptive Differential Evolution (MOSADE). The methodology is tested on three benchmark WDNs, namely Two-loop (TL), GoYang, and Blacksburg (BLA) WDNs, and applied to a real case study of the Badlapur region, Maharashtra, India. The results show that the proposed methodology leads to effective Pareto optimal fronts, making it an efficient method for solving WDN expansion problems. Subsequently, an Analytical Hierarchy Process (AHP) based multi-criteria decision-making (MCDM) analysis was performed on the obtained Pareto-optimal solutions to determine the most suitable solution based on three criteria: Life Cycle Cost (LCC) of expansions, hydraulic reliability, and mechanical reliability. The main advantage of the proposed methodology is its capability to consider hydraulic performance as well as structural integrity and demand satisfaction in the face of hydraulic and mechanical failures.

Graph convolutional networks based contamination source identification across water distribution networks

Water distribution Networks (WDNs) are one of the most important infrastructures for modern society. Due to accidental or malicious reasons, water contamination incidents have been repeatedly reported all over the world, which not only disrupt the water supply but also endanger public health. To ensure the safety of WDNs, water quality sensors are deployed across the WDNs for real-time contamination detection and source identification. In the literature, various methods have been employed to improve the performance of contamination source identification (CSI) and recent studies show that there is a great potential to tackle the CSI problem by deep learning models. The success of deep learning based CSI methods often requires a large size of training samples being collected. In real-world situations, the number of contamination events occurring in a single WDN is rather small, especially for a newly built WDN. However, the existing CSI methods in the literature mostly focus on the study of training and applying models on the same WDNs and the knowledge of CSI gained from one WDN cannot be reused by a different WDN. To these ends, based on the application of graph convolutional networks, this paper provides a solution for cross-network CSI that can transfer the CSI knowledge learned from one WDN to a different WDN. Empirically, based on a benchmark WDN in the task of contamination source identification, we show that the proposed cross-network CSI method can achieve comparable accuracy even trained on a different WDN.

A Wasserstein distance based multiobjective evolutionary algorithm for the risk aware optimization of sensor placement

• A Multiobjective Evolutionary Algorithm for the sensor placement (SP) in a network. • A novel data structure for the spatio-temporal contamination dynamics. • The fitness of a SP is represented as the histogram of a discrete random variable. • The Wasserstein distance between SP's enables new genetic operators. In this paper we propose a new algorithm for the identification of optimal “sensing spots”, within a network, for monitoring the spread of “effects” triggered by “events”. This problem is referred to as “Optimal Sensor Placement” and many real-world problems fit into this general framework. In this paper sensor placement (SP) (i.e., location of sensors at some nodes) for the early detection of contaminants in water distribution networks (WDNs) will be used as a running example. Usually, we have to manage a trade-off between different objective functions, so that we are faced with a multi objective optimization problem. (MOP). The best trade-off between the objectives can be defined in terms of Pareto optimality. In this paper we model the sensor placement problem as a multi objective optimization problem with boolean decision variables and propose a Multi Objective Evolutionary Algorithm (MOEA) for approximating and analyzing the Pareto set. The evaluation of the objective functions requires the execution of a simulation model: to organize the simulation results in a computationally efficient way we propose a data structure collecting simulation outcomes for every SP which is particularly suitable for visualization of the dynamics of contaminant concentration and evolutionary optimization. This data structure enables the definition of information spaces, in which a candidate placement can be represented as a matrix or, in probabilistic terms as a histogram. The introduction of a distance between histograms, namely the Wasserstein (WST) distance, enables to derive new genetic operators, indicators of the quality of the Pareto set and criteria to choose among the Pareto solutions. The new algorithm MOEA/WST has been tested on two benchmark water distribution networks and a real world network. Preliminary results are compared with NSGA-II and show a better performance, in terms of hypervolume and coverage, in particular for relatively large networks and low generation counts.

A New Evolutionary Approach to Optimal Sensor Placement in Water Distribution Networks

The sensor placement problem is modeled as a multi-objective optimization problem with Boolean decision variables. A new multi objective evolutionary algorithm (MOEA) is proposed for approximating and analyzing the set of Pareto optimal solutions. The evaluation of the objective functions requires the execution of a hydraulic simulation model of the network. To organize the simulation results a data structure is proposed which enables the dynamic representation of a sensor placement and its fitness as a heatmap. This allows the definition of information spaces, in which the fitness of a placement can be represented as a matrix or, in probabilistic terms as a histogram. The key element in the new algorithm is this probabilistic representation which is embedded in a space endowed with a metric based on a specific notion of distance. Among several distances between probability distributions the Wasserstein (WST) distance has been selected: WST has enabled to derive new genetic operators, indicators of the quality of the Pareto set and criteria to choose among the Pareto solutions. The new algorithm has been tested on a benchmark water distribution network with two objective functions showing an improvement over NSGA-II, in particular for low generation counts, making it a good candidate for expensive black-box multi-objective optimization

An Integrated Bottom-Up Approach for Leak Detection in Water Distribution Networks Based on Assessing Parameters of Water Balance Model

Loss of water due to leakage is a common phenomenon observed practically in all water distribution networks (WDNs). However, the leakage volume can be reduced significantly if the occurrence of leakage is detected within minimal time after its occurrence. Based on the discriminative behavior of different consumption in water balance, an integrated bottom-up water balance model is presented for leak detection in WDNs. The adaptive moment estimation (Adam) algorithm is employed to assess the parameters in the model. By analyzing the current value and the rising rate of the assessed parameters, abnormal events (e.g., leak, illegal use, or metering inaccuracy) could be detected. Furthermore, a one-step-slower strategy is proposed to estimate the weighted coefficient of pressure sensors to provide approximate location information of leak. The method was applied in a benchmark WDN and an experimental WDN to evaluate its performance. The results showed that relatively small leak could be detected in near-real-time. In addition, the method was able to identify the pressure sensors near to the leak.

DLGEA: a deep learning guided evolutionary algorithm for water contamination source identification

Water distribution network (WDN) is one of the most essential infrastructures all over the world and ensuring water quality is always a top priority. To this end, water quality sensors are often deployed at multiple points of WDNs for real-time contamination detection and fast contamination source identification (CSI). Specifically, CSI aims to identify the location of the contamination source, together with some other variables such as the starting time and the duration. Such information is important in making an efficient plan to mitigate the contamination event. In the literature, simulation-optimisation methods, which combine simulation tools with evolutionary algorithms (EAs), show great potential in solving CSI problems. However, the application of EAs for CSI is still facing big challenges due to their high computational cost. In this paper, we propose DLGEA, a deep learning guided evolutionary algorithm to improve the efficiency by optimising the search space of EAs. Firstly, based on a large number of simulated contamination events, DLGEA trains a deep neural network (DNN) model to capture the relationship between the time series of sensor data and the contamination source nodes. Secondly, given a contamination event, DLGEA guides the initialisation and optimise the search space of EAs based on the top K contamination nodes predicated by the DNN model. Empirically, based on two benchmark WDNs, we show that DLGEA outperforms the CSI method purely based on EAs in terms of both the average topological distance and the accumulated errors between the predicted and the real contamination events.

Optimal Design of Pipe Networks Accounting for Future Demands and Phased Expansion using Integrated Dynamic Programming and Differential Evolution Approach

The water distribution network (WDN) design comprises determining optimal pipe sizes to achieve minimum cost pipe network to meet the required demands and performance levels. However, with time, the water demand for any region changes due to population, migration, and lifestyles, so interventions need to be made to the existing WDNs. Therefore, the capacity expansion problem consists of determining the suitable interventions at various stages such that the required demands are met at minimum cost. The present study proposed a novel methodology for planning such interventions based on Dynamic Programming (DP) formulation and presented a combined Self-Adaptive Differential Evolution and DP (SADE-DP) methodology for solving the WDN expansion problems considering life cycle costs. The methodology is applied and tested on three benchmark WDNs, namely New York Tunnel (NYT), Two loop (TL), and Blacksburg (BLA) networks, and also for a real case study of the Badlapur WDN in Maharashtra, India. The proposed model solutions are validated by comparing them with other WDN expansion methods taken from the literature. The results indicate that the proposed SADE-DP approach is computationally efficient and provides cost-effective solutions by meeting desired performance levels at various stages, and can serve as a potential alternative for solving real-world WDN expansion problems.