[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122706-en":3,"doc-seo-122706-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122706,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Estimating irregular water demands with physics-informed machine learning to inform leakage detection - Abstract and introduction","Leakages in drinking water distribution networks create cascading operational, environmental, and financial impacts, making timely detection and accurate localisation essential for utilities. Existing leakage-detection methods are often constrained by the need for calibrated hydraulic models or by the availability of large training datasets. Physics-informed machine learning uses hydraulic information to bypass both limitations. The proposed approach estimates unknown irregular demands from pressure data via a fully connected neural network grounded in the Bernoulli equation, improving leakage identification on the L-Town benchmark.","Estimating irregular water demands with physics-informed machine learning to inform leakage detection  \nIvo Daniel a,b,*, Andrea Cominola a,b  \na Chair of Smart Water Networks, Technische Universität Berlin, Straße des 17. Juni 135, 10623 Berlin, Germany  \nb Einstein Center Digital Future, Wilhelmstraße 67, 10117 Berlin, Germany  \n*  \nEmail: [ivo.daniel@tu-berlin.de](ivo.daniel@tu-berlin.de)  \nABSTRACT  \nLeakages in drinking water distribution networks pose significant challenges to water utilities, leading to infrastructure failure, operational disruptions, environmental hazards, property damage, and economic losses. The timely identification and accurate localisation of such leakages is paramount for utilities to mitigate these unwanted effects. However, implementation of algorithms for leakage detection is limited in practice by requirements of either hydraulic models or large amounts of training data. Physics-informed machine learning can utilise hydraulic information thereby circumventing both limitations. In this work, we present a physics-informed machine learning algorithm that analyses pressure data and therefrom estimates unknown irregular water demands via a fully connected neural network, ultimately leveraging the Bernoulli equation and effectively linearising the leakage detection problem. Our algorithm is tested on data from the L-Town benchmark network, and results indicate a good capability for estimating most irregular demands, with R2 larger than 0.8. Identification results for leakages under the presence of irregular demands could be improved by a factor of 5.3 for abrupt leaks and a factor of 3.0 for incipient leaks when compared the results disregarding irregular demands.  \n1 Introduction  \nLeakages are the most common cause of infrastructure failure in drinking water distribution networks (WDN) (Arregui et al., 2018) and entail a variety of unintended consequences, ranging from operational disruptions (Misiunas et al., 2006), environmental hazards (Wan et al., 2022), property damage (Mansour-Rezaei and Naser, 2013), and sanitary issues (Shortridge and Guikema, 2014) to increased economic cost through non-revenue water (NRW) and  \ninfrastructure maintenance (Puust et al., 2010) . In 2019, the World Bank estimated global water losses to about 120 million m3 p.a., resulting in an associated economic loss of 39 billion USD p.a. (Liemberger and Wyatt, 2019) .  \nWhile up to 35% of water supplied to the WDN is lost as NRW in developed countries (Levinaset al., 2021), water losses amount to more than 50% in developing countries (Puust et al., 2010) . Additionally, urbanisation (Flörke et al., 2018) and climate change induced drought increase (Konapala et al., 2020) further exacerbate challenges faced by water utilities regarding the sustainability and security of their drinking water provision. Furthermore, leakages add up to an already heavy energy balance of the water sector, which contributes up to 2.7% of total global energy use (Liu et al., 2016), thus requiring drastic climate mitigation solutions (Parkinson, 2021) .  \nIn light of these challenges, leakage management arises as a crucial task for water utilities to efficiently sustain and maintain their WDN (Barton et al., 2019) . As possible strategies within leakage management, leakage control focuses on the implementation of policies and management strategies to prevent and mitigate leakage impacts, while leakage detection aims at timely identifying and precisely locating occurring leaks (Puust et al., 2010) . Methods for leakage detection can be further divided into (i) model-based approaches that employ hydraulic models to compare simulated to sensor data, (ii) data-driven approaches that analyse timeseries data using mathematical models, and (ii) hybrid approaches combining elements from both former categories (Zaman et al., 2020) .  \nIn recent years, the leakage detection problem has received increased attention from the scientific community with a st","cbCaiveSb356rjV3","https://ap.wps.com/l/cbCaiveSb356rjV3","pdf",3334675,1,23,"English","en",105,"# Abstract\n# Introduction\n## Challenges of leakage management\n## Leakage detection strategies and method categories\n## Limits of model-based and data-driven approaches","[{\"question\":\"Why is leakage detection difficult in real drinking water distribution networks?\",\"answer\":\"Detection is limited by either the need for well-calibrated hydraulic models or by the requirement of large amounts of training data, both of which can hinder practical implementation.\"},{\"question\":\"What does the proposed physics-informed machine learning method use as inputs and physical constraints?\",\"answer\":\"It analyses pressure data and estimates unknown irregular water demands, using a fully connected neural network that leverages the Bernoulli equation to linearise the leakage detection problem.\"},{\"question\":\"How is the algorithm evaluated and what performance improvements are reported?\",\"answer\":\"It is tested on the L-Town benchmark network, showing an ability to estimate most irregular demands with R² above 0.8. Leakage identification improves by 5.3× for abrupt leaks and 3.0× for incipient leaks compared with results ignoring irregular demands.\"}]","Estimating irregular water demands with physics-informed machine learning to inform leakage detection - Abstract and introduction | PDF",1785812409,58,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"estimating-irregular-water-demands-with-physics-informed-machine-learning-to-inform-leakage-detection-abstract-and-introduction","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/estimating-irregular-water-demands-with-physics-informed-machine-learning-to-inform-leakage-detection-abstract-and-introduction/122706/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is leakage detection difficult in real drinking water distribution networks?","Question",{"text":75,"@type":76},"Detection is limited by either the need for well-calibrated hydraulic models or by the requirement of large amounts of training data, both of which can hinder practical implementation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does the proposed physics-informed machine learning method use as inputs and physical constraints?",{"text":80,"@type":76},"It analyses pressure data and estimates unknown irregular water demands, using a fully connected neural network that leverages the Bernoulli equation to linearise the leakage detection problem.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the algorithm evaluated and what performance improvements are reported?",{"text":84,"@type":76},"It is tested on the L-Town benchmark network, showing an ability to estimate most irregular demands with R² above 0.8. Leakage identification improves by 5.3× for abrupt leaks and 3.0× for incipient leaks compared with results ignoring irregular demands.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]