Robust Anomaly Detection in Water Distribution Networks using a Physics-Aware Graph Neural Network with Stratified Cross-Validation

Robust Anomaly Detection in Water Distribution Networks using a Physics-Aware Graph Neural Network with Stratified Cross-Validation | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 21 January 2026 V1 Latest version Share on Robust Anomaly Detection in Water Distribution Networks using a Physics-Aware Graph Neural Network with Stratified Cross-Validation Authors : Rai Ali Yar 0009-0004-3523-6520 [email protected] , Umaisa Lail , Anwar Shah , Muneeb Arif , and Zeeshan Haider Authors Info & Affiliations https://doi.org/10.22541/au.176901860.09145325/v1 109 views 65 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Modern water distribution systems have largely gone digital. While that gives us better oversight, it also opens the door to a broader range of cyber-physical attacks. The issue is that a lot of existing anomaly detection methods still lean on older machine learning techniques that struggle to model irregularities. Consequently, their performance tends to suffer when the system gets noisy or when an attack effectively hides behind normal daily fluctuations, a scenario that is unfortunately common in real-world facilities. In this study, we propose a graph-based learning framework in which each end device is treated as a node. Since datasets like BATADAL usually lack complete physical maps, we establish the connections (edges) based on statistical similarity rather than strict hydraulic theory. To capture how components naturally operate in clusters, we also introduced an "inter-intra" machine grouping scheme that helps build a more realistic interaction graph. We process the input time series using sliding windows, feeding them into a hybrid architecture that blends graph convolution layers with temporal sequence modeling. Before the training phase, we run a lightweight preprocessing step to strip out constant features and handle missing data, ensuring the windowing process captures short-term behavior without accidentally leaking future information. Our tests on the BATADAL benchmark show that this framework successfully flags abnormal patterns, even if the malicious behavior is subtle or short-lived. While performance Vol., No.,. does vary slightly depending on the window length and graph construction thresholds, the model beats standard baselines. kn Supplementary Material File (robust anomaly detection in water distribution.pdf) Download 1.57 MB Information & Authors Information Version history V1 Version 1 21 January 2026 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords anomaly detection cyber-physical systems graph neural networks water distribution networks Authors Affiliations Rai Ali Yar 0009-0004-3523-6520 [email protected] FAST NU View all articles by this author Umaisa Lail Riphah International University View all articles by this author Anwar Shah FAST NU View all articles by this author Muneeb Arif FAST NU View all articles by this author Zeeshan Haider FAST NU View all articles by this author Metrics & Citations Metrics Article Usage 109 views 65 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Rai Ali Yar, Umaisa Lail, Anwar Shah, et al. Robust Anomaly Detection in Water Distribution Networks using a Physics-Aware Graph Neural Network with Stratified Cross-Validation. Authorea . 21 January 2026. DOI: https://doi.org/10.22541/au.176901860.09145325/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176901860.09145325/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe60ad3b9658e2e',t:'MTc3OTIyNDU0Mw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();