scholarly journals Cyber—Physical Attack Detection in Water Distribution Systems with Temporal Graph Convolutional Neural Networks

Water ◽  
2021 ◽  
Vol 13 (9) ◽  
pp. 1247
Author(s):  
Lydia Tsiami ◽  
Christos Makropoulos
Keyword(s):  
Neural Networks ◽  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Attack Detection ◽  
Convolutional Network ◽  
Irreversible Damage ◽  
Physical Attack ◽  
Temporal Graph ◽  
Graph Neural Networks

Prompt detection of cyber–physical attacks (CPAs) on a water distribution system (WDS) is critical to avoid irreversible damage to the network infrastructure and disruption of water services. However, the complex interdependencies of the water network’s components make CPA detection challenging. To better capture the spatiotemporal dimensions of these interdependencies, we represented the WDS as a mathematical graph and approached the problem by utilizing graph neural networks. We presented an online, one-stage, prediction-based algorithm that implements the temporal graph convolutional network and makes use of the Mahalanobis distance. The algorithm exhibited strong detection performance and was capable of localizing the targeted network components for several benchmark attacks. We suggested that an important property of the proposed algorithm was its explainability, which allowed the extraction of useful information about how the model works and as such it is a step towards the creation of trustworthy AI algorithms for water applications. Additional insights into metrics commonly used to rank algorithm performance were also presented and discussed.

Modeling Water Distribution Systems with Graph Neural Networks

2021 ◽  
Author(s):  
Alexander Garzón ◽  
Roberto Bentivoglio ◽  
Elvin Isufi ◽  
Zoran Kapelan ◽  
Riccardo Taormina
Keyword(s):  
Neural Networks ◽  
Learning Process ◽  
Distribution Systems ◽  
Water Distribution ◽  
Water Distribution Systems ◽  
Learning Tools ◽  
Essential Information ◽  
Benchmark System ◽  
Processing Architectures ◽  
Graph Neural Networks

<p>Water management has recently explored data-driven models to improve the adaptability of Water Distribution Systems (WDS) and strengthen decision making under uncertain conditions. The focus on these tools is motivated by the increasing availability of information and their proven performance in other fields. Modeling WDS with these techniques has been demonstrated to be useful; however, the traditional machine learning tools do not account for the graph structure present in the WDS. Considering this essential information offers the possibility to increase performance and to help the learning process. In this work, we introduce Graph Neural Networks (GNNs) for modeling WDS. GNNs are processing architectures to perform neural network tasks for data related to a graph. We first present the definitions and interpretations for using this framework in water networks. Then we compare the GNN approach against standard neural networks to predict an overall resilience metric in a benchmark system. The benefits of including the network structure in the learning process by the GNN are shown in the analysis of the obtained results.</p>

scholarly journals Application of genetic programming to modeling pipe failures in water distribution systems

2010 ◽  
Vol 13 (3) ◽  
pp. 419-428 ◽  
Author(s):  
Qiang Xu ◽  
Qiuwen Chen ◽  
Weifeng Li
Keyword(s):  
Genetic Programming ◽  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Pipe Failure ◽  
Beijing City ◽  
Failure Data ◽  
Processing Strategies ◽  
Gis Data ◽  
Failure Models

The water loss from a water distribution system is a serious problem for many cities, which incurs enormous economic and social loss. However, the economic and human resource costs to exactly locate the leakage are extraordinarily high. Thus, reliable and robust pipe failure models are demanded to assess a pipe's propensity to fail. Beijing City was selected as the case study area and the pipe failure data for 19 years (1987–2005) were analyzed. Three different kinds of methods were applied to build pipe failure models. First, a statistical model was built, which discovered that the ages of leakage pipes followed the Weibull distribution. Then, two other models were developed using genetic programming (GP) with different data pre-processing strategies. The three models were compared thereafter and the best model was applied to assess the criticality of all the pipe segments of the entire water supply network in Beijing City based on GIS data.

scholarly journals A Head Formulation for the Steady-State Analysis of Water Distribution Systems Using an Explicit and Exact Expression of the Colebrook–White Equation

Water ◽  
2021 ◽  
Vol 13 (9) ◽  
pp. 1163
Author(s):  
Mengning Qiu ◽  
Avi Ostfeld
Keyword(s):  
Steady State ◽  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Solution Method ◽  
Water Distribution System ◽  
Academic Research ◽  
Exact Expression ◽  
Contamination Source

Steady-state demand-driven water distribution system (WDS) solution is the bedrock for much research conducted in the field related to WDSs. WDSs are modeled using the Darcy–Weisbach equation with the Swamee–Jain equation. However, the Swamee–Jain equation approximates the Colebrook–White equation, errors of which are within 1% for ϵ/D∈[10−6,10−2] and Re∈[5000,108]. A formulation is presented for the solution of WDSs using the Colebrook–White equation. The correctness and efficacy of the head formulation have been demonstrated by applying it to six WDSs with the number of pipes ranges from 454 to 157,044 and the number of nodes ranges from 443 to 150,630. The addition of a physically and fundamentally more accurate WDS solution method can improve the quality of the results achieved in both academic research and industrial application, such as contamination source identification, water hammer analysis, WDS network calibration, sensor placement, and least-cost design and operation of WDSs.

scholarly journals Comparison of Searching Behaviour of Three Evolutionary Algorithms Applied to Water Distribution System Design Optimization

Water ◽  
2020 ◽  
Vol 12 (3) ◽  
pp. 695 ◽  
Author(s):  
Weiwei Bi ◽  
Yihui Xu ◽  
Hongyu Wang
Keyword(s):  
Evolutionary Algorithms ◽  
Real Time ◽  
Distribution System ◽  
Optimization Design ◽  
Distribution Systems ◽  
Water Distribution ◽  
Optimal Solutions ◽  
Great Ability ◽  
Searching Behaviour ◽  
Run Time

Over the past few decades, various evolutionary algorithms (EAs) have been applied to the optimization design of water distribution systems (WDSs). An important research area is to compare the performance of these EAs, thereby offering guidance for the selection of the appropriate EAs for practical implementations. Such comparisons are mainly based on the final solution statistics and, hence, are unable to provide knowledge on how different EAs reach the final optimal solutions and why different EAs performed differently in identifying optimal solutions. To this end, this paper aims to compare the real-time searching behaviour of three widely used EAs, which are genetic algorithms (GAs), the differential evolution (DE) algorithm and the ant colony optimization (ACO). These three EAs are applied to five WDS benchmarking case studies with different scales and complexities, and a set of five metrics are used to measure their run-time searching quality and convergence properties. Results show that the run-time metrics can effectively reveal the underlying searching mechanisms associated with each EA, which significantly goes beyond the knowledge from the traditional end-of-run solution statistics. It is observed that the DE is able to identify better solutions if moderate and large computational budgets are allowed due to its great ability in maintaining the balance between the exploration and exploitation. However, if the computational resources are rather limited or the decision has to be made in a very short time (e.g., real-time WDS operation), the GA can be a good choice as it can always identify better solutions than the DE and ACO at the early searching stages. Based on the results, the ACO performs the worst for the five case study considered. The outcome of this study is the offer of guidance for the algorithm selection based on the available computation resources, as well as knowledge into the EA’s underlying searching behaviours.

scholarly journals Modelling both the continual erosion and regeneration of discolouration material in drinking water distribution systems

2013 ◽  
Vol 14 (1) ◽  
pp. 81-90 ◽  
Author(s):  
W. R. Furnass ◽  
R. P. Collins ◽  
P. S. Husband ◽  
R. L. Sharpe ◽  
S. R. Mounce ◽  
...  
Keyword(s):  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Transport Model ◽  
Water Distribution Systems ◽  
Sensitivity Analyses ◽  
Drinking Water Distribution Systems ◽  
Proactive Management ◽  
Drinking Water Distribution ◽  
Mass Transport Model

The erosion of the cohesive layers of particulate matter that causes discolouration in water distribution system mains has previously been modelled using the Prediction of Discolouration in Distribution Systems (PODDS) model. When first proposed, PODDS featured an unvalidated means by which material regeneration on pipe walls could be simulated. Field and laboratory studies of material regeneration have yielded data that suggest that the PODDS formulations incorrectly model these processes. A new model is proposed to overcome this shortcoming. It tracks the relative amount of discolouration material that is bound to the pipe wall over time at each of a number of shear strengths. The model formulations and a mass transport model have been encoded as software, which has been used to verify the model's constructs and undertake sensitivity analyses. The new formulations for regeneration are conceptually consistent with field and laboratory observed data and have potential value in the proactive management of water distribution systems, such as evaluating change in discolouration risk and planning timely interventions.

Dynamic updating of post-earthquake damage and functional restoration forecasts of water distribution systems using Bayesian inferencing

2021 ◽  
pp. 875529302110380
Author(s):  
Agam Tomar ◽  
Henry V Burton ◽  
Ali Mosleh
Keyword(s):  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Epistemic Uncertainty ◽  
Discrete Event ◽  
Complete Recovery ◽  
Current Data ◽  
Functional Restoration ◽  
Mitigation Measures ◽  
Total Recovery

A framework for dynamically updating post-earthquake functional recovery forecasts is presented to reduce the epistemic uncertainty in the predictive model. A Bayesian Network (BN) model is used to provide estimates of the total recovery time, and a process-based discrete event simulation (PBDES) model generates forecasts of the complete recovery trajectory. Both models rely on component damage and duration-based input parameters that are dynamically updated using Bayes’ theorem, as information becomes available throughout the recovery process. The effectiveness of the proposed framework is demonstrated through an application to the pipe network of the City of Napa water distribution system. More specifically, pipe damage and repair data from the 2014 earthquake are used as a point of comparison for the dynamic forecasts. It is shown that, over time, the mean value of the total recovery duration generated by the BN-based model converges to the observed value and the dispersion is reduced. Also, despite a crude initial estimate, the median trajectory generated by the PBDES model provides a reasonable approximation of the observed recovery within 30 days following the earthquake. The proposed framework can be used by emergency managers to investigate the efficacy of post-event mitigation measures (e.g. crew allocation, resource prioritization) utilizing the most current data and knowledge.

scholarly journals Multiscale Spatio-Temporal Graph Neural Networks for 3D Skeleton-Based Motion Prediction

2021 ◽  
Vol 30 ◽  
pp. 7760-7775
Author(s):  
Maosen Li ◽  
Siheng Chen ◽  
Yangheng Zhao ◽  
Ya Zhang ◽  
Yanfeng Wang ◽  
...  
Keyword(s):  
Neural Networks ◽  
Motion Prediction ◽  
Temporal Graph ◽  
3D Skeleton ◽  
Spatio Temporal ◽  
Graph Neural Networks

scholarly journals Design of Water Distribution System: Appropriateness of Structured System in Large Irrigation Projects in Nepal

2016 ◽  
Vol 19 ◽  
pp. 25-30
Author(s):  
Basistha Adhakari
Keyword(s):  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Irrigation System ◽  
Command Area ◽  
Irrigation Project ◽  
Irrigation Projects ◽  
Structured System ◽  
Large Irrigation ◽  
Area Development

Many large irrigation projects in Nepal operate under command area development works that emphasize on-farm water distribution and management. These projects have specific design characteristics that were planned to comply with available water resources, climatic conditions, soil type, and water distribution technology. The water distribution technologies differ based on the design needs of each individual project and the design preferences of various foreign consulting firms. This article focuses on the issues of planning and designing water distribution systems of large irrigation systems at the irrigation service delivery level. The layout planning of an irrigation system is an important aspect of design for water distribution, typically guided by hierarchical system. This article also highlights the existing canal hierarchy of these systems and their appropriateness for efficient water distribution. Furthermore, the appropriateness of the structured system is also examined in the Sunsari Morang Irrigation Project. The article concluded with some suggestions for planning and designing command area development works of forthcoming large irrigation projects such as the Sikta Irrigation Project, the Babai Irrigation Project, and the Mahakali Irrigation Project Stage-III.HYDRO Nepal JournalJournal of Water, Energy and EnvironmentIssue: 19Page: 25-30

scholarly journals Hydraulic Performance Analysis and Modeling of Water Supply Distribution System of Addis Ababa Science and Technology University, Ethiopia

2020 ◽  
Author(s):  
Chalchisa Milkecha ◽  
Habtamu Itefa
Keyword(s):  
Water Supply ◽  
Water Demand ◽  
Distribution System ◽  
Distribution Systems ◽  
Water Distribution ◽  
Addis Ababa ◽  
Science And Technology ◽  
Hydraulic Performance ◽  
Alternative Methods ◽  
The University

This study was conducted generally by aiming assessment of the hydraulic performance of water distribution systems of Addis Ababa Science and Technology University (AASTU). In line with the main objective, this study addressed, (1) pinpointing problems of existing water supply versus demand deficit (2) evaluating the hydraulic performance of water distribution system using water GEMS and (3) recommended alternative methods for improving water demand scenarios. The University’s water supply distribution network layout was a looped system and the flow of water derived by both gravity and pressurized system. The gravity flow served for the academic and administrative staffs whereas the pressurized system of the network fed the students dormitories, cafeteria’s etc. The study revealed the existence of unmet minimum pressure requirement around the student dormitories which accounts 25.64% below the country’s building code standard during the peak hour consumption. The result of the water demand projection showed an increment of 2.5 liter per capita demand (LPCD) in every five years. Hence, first, the university’s water demand was projected and then hydraulic parameters such as; pressure, head loss and velocity were modeled for both the existing and the improved water supply distribution. The finding of the study was recommended to the university’s water supply project and institutional development offices for its future modification and rehabilitation works.

Sign in / Sign up

Export Citation Format

Share Document