This session aims to provide an active forum to discuss and exchange knowledge on state-of-the-art and emerging tools, frameworks, and methodologies for planning and management of modern urban water infrastructure, with a particular focus on digitalization and/or interconnections with other systems, looking at the bigger picture. Topics and applications may cover any area of urban water network analysis, modeling, and management, including intelligent sensors and advanced metering, digital twins, asset management, decision making, novel applications of IoT, and challenges to their implementation or risk of lock-in of rigid system designs. Methods and approaches may also include big-data analytics and information retrieval, data-driven behavioral analysis, graph theory, ontologies and artificial intelligence for water applications (including large language models and physics-informed machine learning), descriptive and predictive models of, e.g., water demand, sewer system flow/flood extent, experimental approaches to demand management, water demand and supply optimization, energy recovery from urban water networks, real-time control of urban drainage systems, anomaly identification in hydraulic and water quality sensor data (e.g., for leak detection, identification of contamination events). Investigations on interconnected systems could explore emerging areas such as cyber-physical security of urban water systems (i.e., communication infrastructure), combined reliability and assessment studies on urban metabolism, or minimization of flood impacts on urban networks and energy usage optimization.
PICO: Thu, 7 May, 08:30–10:15 | PICO spot 4
Two emerging paradigms are redefining AI for physical systems: foundational models and intelligent agents. In the context of urban drainage systems, both offer the opportunity to move beyond task-specific tools toward more general, scalable, and operationally meaningful AI systems. This contribution discusses the state of the art, presents recent advances from our research, and outlines a realistic pathway toward AI-based urban drainage intelligence.
The first paradigm focuses on foundational, physics-aware models that aim to replace computationally expensive numerical simulators while fully exploiting real-world measurements. Such models have the potential to significantly strengthen digital twins by enabling fast, differentiable, and transferable representations of system dynamics. We present recent developments in graph neural network (GNN)–based surrogate models for urban drainage simulation, including autoregressive architectures capable of emulating both the dynamics of 1D numerical models and 2D shallow-water equation solvers. These results demonstrate that accurate and scalable learning-based simulators are now feasible for complex hydraulic processes.
Looking ahead, a key research challenge lies in integrating drainage network models with surface flow representations, enabling unified 1D/2D modelling of the coupled behaviour of sewers, floodplains, and urban catchments during extreme events. Another critical opportunity lies in exploiting the differentiable nature of AI models, which opens the door to assimilating real-world observations directly into model parameters. This offers a principled alternative to traditional calibration workflows, while also enabling continuous adaptation as new data become available. At the same time, the scarcity of high-quality real-world flood observations implies that pretraining on large-scale simulations will likely remain essential to developing robust and transferable models.
The second paradigm concerns AI agents for complex engineering tasks, where systems are designed not only to make predictions but also to reason, plan, and take actions within operational workflows. In the context of urban drainage, such agents could ultimately support activities ranging from model building and calibration to decision support, monitoring, and infrastructure management. As a concrete example of this broader direction, we present our ongoing work on automatic sewer defect detection, where we evaluate the limitations of current general-purpose vision–language models for infrastructure inspection. Our results indicate that meaningful progress will require domain-specific multimodal models, tailored to sewer imagery and engineering semantics. These models can naturally evolve toward vision–language–action systems, enabling compact, efficient agents suitable for deployment on robotic platforms and edge devices, with appropriate safety constraints.
How to cite: Taormina, R.: Foundational Models and Intelligent Agents for Urban Drainage Systems: The Road Ahead, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12269, https://doi.org/10.5194/egusphere-egu26-12269, 2026.
Urban pluvial flooding, caused when local precipitation intensity outpaces the capacity of natural and engineered drainage systems, is among the costliest, most dangerous, and most prevalent forms of natural disaster. While flood inundation extent maps form the basis for federal management of coastal and riverine flood risk management, there is no existing technology that allows for real-time spatially continuous monitoring of urban pluvial flooding, which critically limits equitable and effective planning, response, and mitigation. The proposed research enables a low-power distributed sensor network that synchronizes storm sewer stage monitoring with a camera-based flood mapping platform that will provide automatic monitoring and alerts of urban surface inundation and stormwater backups, and autonomously generate spatially complete maps of peak flooding extent for comprehensive and equitable pluvial risk characterization and mitigation. To realize this vision, applied research will be conducted to leverage edge-AI (Artificial Intelligence) to identify flooding in multispectral images representing diverse urban and peri-urban contexts. Efficient methods for georeferencing images using photogrammetric models of the neighborhoods where cameras are installed will be developed and evaluated. By synchronizing cameras with LoRaWAN-networked USGS storm sewer stage sensors, we will demonstrate how edge-AI enables spatially distributed inundation measurements from cameras with low energy and communication costs and without collecting raw imagery that could contribute to unnecessary surveillance in host communities. Furthermore, we will demonstrate how networked sensor platforms can get smarter over time.
How to cite: Shergill, M. and Carter, E.: Harnessing Computer Vision for Advanced Flood Forecasting in Urban Environments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14862, https://doi.org/10.5194/egusphere-egu26-14862, 2026.
Pluvial flooding on roadways poses a significant threat to drivers. In addition, frequency of such pluvial floods is increasing due to climate change, urbanization, impervious surfaces, and aging stormwater infrastructure. While physics-based hydrodynamic models provide detailed results, their computational demand and reliance on often-uncertain drainage layouts make them impractical for rapid, real-time urban risk assessment.
SaferPlaces (saferplaces.co) addresses this with a web-platform that integrates high-resolution geospatial and climate data from sources like GEE, OSM, and Copernicus to automate the construction of digital-twins for flood risk modelling in urban areas. As part of the SaferPlaces platform, Safer_RAIN, a fast-processing DEM based algorithm that combines a pixel-based Green-Ampt infiltration model with a Hierarchical Filling-and-Spilling Algorithm (HFSA) approach to enable building-level risk assessments. By leveraging cloud-based infrastructure, the platform delivers high-resolution, real-time results for urban planning. However, when compared to 2D hydrodynamic models, Safer_RAIN showed some limitations, including the underestimation of flooded extents due to the absence of hydraulic backwater effects and single flow-path constraints. Therefore, this study introduces an enhancement to Safer_RAIN, termed Kinematic Safer_RAIN, by incorporating a kinematic approach to simulate flooding beyond depressions and integrating a flow-path inundation extension feature. This enhancement utilizes the Height Above Nearest Drainage (HAND) approach, coupled with Manning’s equation, to represent flood inundation along flow paths (e.g. essential for assessing risks to urban road networks).
Kinematic Safer_RAIN, featuring flow-path flood extension, was benchmarked against HEC-RAS 2D Rain-on-Grid hydrodynamic simulations using 1m-resolution LiDAR DEMs in two case studies. In the Cottonwood Lake Study Area (USA), Kinematic Safer_RAIN produced maximum flooding extents and water depth distributions that closely match HEC-RAS results. The model was further validated in Pamplona (Spain), using the extreme storm event of July 2010. Kinematic Safer_RAIN successfully identified flood-prone depressions and flooding along flow paths (primarily main roads and lanes), yielding high True Positive Rates and aligning with flood evidence from local newspaper images. This research provides a robust, low-cost, and rapid alternative for authorities to accurately predict roadway flooding risks, bridging the gap between topographic simplicity and hydrodynamic complexity to enhance flood mitigation strategies.
Keywords: Flow path flood extension, Hierarchical Filling-and-Spilling Algorithm (HFSA), Height Above Nearest Drainage (HAND), Pluvial Flooding, Safer_RAIN, Kinematic_SaferRAIN, SaferPlaces
How to cite: Kyaw, K. K., Mediero, L., Luzzi, V., Bagli, S., and Castellarin, A.: Kinematic Hierarchical Filling-and-Spilling Algorithm for Advanced DEM-based Modelling of Pluvial Flooding in Urban Areas, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13010, https://doi.org/10.5194/egusphere-egu26-13010, 2026.
Urban drainage networks play a critical role in mitigating flood risk in rapidly urbanizing environments, yet their dynamic behavior under rainfall forcing remains difficult to describe using simple mathematical models. Most existing studies rely on detailed hydrodynamic simulators, such as EPA-SWMM, which are well suited for design and scenario analysis but are less suitable to system-level analysis and control-oriented modeling due to their complexity.
This study proposes a control-oriented modeling framework for urban drainage systems based on system identification techniques applied to SWMM-generated input-output data. Rainfall-runoff and hydraulic responses are simulated for an urban drainage network under multiple storm events. Time series of inflow and water depth are extracted at selected junctions identified as critical using flooding and surcharge indicators.
A physics-based mass balance formulation is adopted, and local storage-outflow dynamics are linearized around representative operating conditions. Using the resulting input-output data, first-order dynamic models are identified for individual drainage nodes. The derived transfer functions capture the dominant inflow-depth dynamics, where the static gain represents the steady-state sensitivity of water depth to inflow variations and the time constant reflects the effective storage behavior of the drainage node.
Model performance is evaluated by testing the identified models under different rainfall events. The proposed approach provides compact and interpretable models that bridge detailed hydraulic simulation and control-oriented analysis, supporting digital water and digital twin applications for urban drainage systems.
How to cite: Pandey, R. and Das Bhowmik, R.: Control-Oriented System Identification of Urban Drainage Dynamics for Flood Mitigation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10149, https://doi.org/10.5194/egusphere-egu26-10149, 2026.
Urban areas are increasingly facing the limitations of existing drainage systems due to the growing frequency and intensity of extreme rainfall caused by climate change. As underground space becomes more developed and urban infrastructure grows more complex, effective flood mitigation requires a planning tool capable of systematically evaluating large-scale drainage facilities such as deep stormwater tunnels. This study proposes a SaaS-based integrated analysis and design system that automates the entire process of rainfall–runoff simulation, tunnel operation optimization, and surface inundation assessment.
The system consists of three major components:
(1) a SWMM-based hydrologic and hydraulic simulation engine,
(2) an automated calculation module for determining the storage capacity, inlet configuration, pumping requirements, and diversion channel sizing of deep stormwater tunnels, and
(3) a cloud-based data environment for storing, managing, and visualizing large hydrologic and topographic datasets.
Input data—including terrain, catchment characteristics, sewer networks, and design storm information—are automatically processed in the cloud database. The computation server conducts repeated SWMM simulations for various rainfall scenarios, generating results related to pipe surcharge, water-level variation, and overall drainage performance. A 2D inundation model is integrated to assess surface flooding before and after the construction of a deep stormwater tunnel, enabling spatial comparison of flood reduction effects.
The system outputs include required tunnel storage volume, surcharge-prone locations, inundation depth maps, and comparative scenario analyses. Users can easily generate, modify, and store multiple design alternatives and share them within a project team. This integrated modeling environment significantly improves the efficiency of complex drainage analyses and enhances the reliability of decision-making for urban flood mitigation infrastructure.
Overall, the proposed system serves as a next-generation digital tool that supports intuitive and comprehensive evaluation of urban drainage conditions. It is expected to markedly improve the practical applicability and planning efficiency of deep stormwater tunnel projects in future urban flood management efforts.
Acknowledgements
This work was supported by Korea Environment Industry & Technology Institute(KEITI) through Technology development project to optimize planning, operation, and maintenance of urban flood control facilities, funded by Korea Ministry of Climate, Energy, Environment(MCEE)(RS-2024-00398012)
How to cite: Lee, Y. E., Kim, M., and Park, J. P.: SaaS-Based Integrated Analysis System for Deep Stormwater Tunnels to Reduce Urban Flooding, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1383, https://doi.org/10.5194/egusphere-egu26-1383, 2026.
During the long-term operation of urban drainage networks (UDNs), the accumulation of sediments and debris can cause partial or complete blockages. Blockages within the pipe network may lead to flooding at manholes, and such overflows can disrupt traffic and increase pollution emissions in urban areas. Identifying the locations of partial blockages and implementing a regular pipe monitoring and cleaning plan support proactive maintenance and operation of UDNs. This approach helps increase network resilience, prevents complete pipe choking, and reduces flood inundation during high-intensity rainfall over long-term operation. With Internet of Things IoT-based smart urban water systems, sensor data can be used for the UDN anomalies detection, such as blockages, through water-level monitoring and time-series analysis. To improve the accuracy of blockage detection, determining the optimal sensor locations while considering implementation costs is a fundamental part of IoT-based UDN anomaly detection. Therefore, this work focuses on the optimal placement of sensors in UDNs for anomaly detection, supported by graph signal analysis. Water elevation variation data are modeled as signals on a graph representation of the UDN, providing a robust framework for effective anomaly detection.
In this research, at first, graph clustering is applied to divide the UDN into monitoring zones corresponding to the number of sensors. Subsequently, the optimal sensor locations within the monitoring zones are determined. For that, a genetic algorithm (GA) is used to determine the optimal location of each sensor within its corresponding cluster (monitoring zone). Therefore, the sensor network is modeled as a graph in which vertices correspond to sensor locations at manholes, and edges represent the minimum shortest paths connecting these locations. The objective function for optimal sensor placement is based on the graph Fourier analysis of that sensor network subgraph. Finally, water elevation variation data are assigned to the graph nodes as node signals. Using the graph Fourier transform (GFT), the graph Fourier coefficients of these signals are computed. The proportion of high-frequency components, defined as the energy contained in the largest 50% of Laplacian eigenvalues relative to the total signal energy, is used as a metric for anomaly detection efficiency. Nodes exhibiting high-energy components at these large eigenvalues are more suitable for blockage detection, as such high-frequency variations indicate localized disturbances. These variations have greater potential for accurately identifying pipe blockages and reducing misinterpretation in the sensor network under multiple blockage scenarios. The proposed method is implemented in a real-world UDN in an alpine region, and the performance of the sensor placement strategy is validated through sensitivity analysis under modelling multiple pipe blockage scenarios and varying numbers of sensors.
Funding: The project “RESTORE” is funded by the Austrian Science Fund (FWF) P 36737-N.
How to cite: Rajabi, M., Hajibabaei, M., and Sitzenfrei, R.: Optimal Sensor Placement for Pipe Blockage Detection in Urban Drainage Networks Using Graph Signal Processing, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19050, https://doi.org/10.5194/egusphere-egu26-19050, 2026.
Urban water distribution networks (WDNs) are intended to reliably supply safe drinking water to users, while ensuring an economically and environmentally sustainable management of the supply infrastructure. However, WDNs are prone to water losses, mostly due to ageing components and insufficient maintenance. In the last decades, substantial research efforts have focused on developing methods for water loss reduction, targeting an optimal management of the WDNs infrastructure and quantifiable economic and water savings. Proposed approaches include programmed replacement of pipes, pressure management by means of optimally deployed pressure reduction valves, and prompt leak detection and subsequent localisation by means of sensor data and automated algorithms.
Beyond their direct impact on water supply efficiency, leakages in WDNs also contribute to an underestimated and, so far, understudied problem, namely the hydrogeological instability in urban environments, which can manifest as ground subsidence, surface deformation, or the formation of sinkholes. Such processes can cause substantial damage to infrastructure, disrupt traffic circulation, damage vehicles, compromise underground utilities, weaken the structural integrity of buildings, and, in the most severe cases, pose a threat to public safety. Incorporating planning and management strategies aimed at reducing hydrogeological risks associated with pipe leaks in WDNs is thus key to fostering the resilience of WDNs within the urban environment, besides water supply efficiency and reliability. This study presents an optimal sensor placement framework for mitigating the risk of Hydrogeological Disruption from Leakage (HDL).
The framework sequentially combines a spatial risk zonation approach with an optimal pressure sensor placement for accurate leak localisation, where sensor placement is driven by the objective of maximising leak localisation accuracy in the most vulnerable and exposed areas of the city. The optimisation process makes use of an evolutionary algorithm (GA from the package Pymoo) where candidate pressure sensor configurations are evaluated across different leak scenarios. The objective function combines the minimum hydraulic paths between actual and predicted leaks with weights representative of different risk levels, while the leak scenarios are produced with the hydraulic model included in the WNTR library under the assumption that a single active leak occurs along pipes. Localisation is based on a sensitivity matrix that maps the pressure response of the network nodes to specific leak scenarios, characterised by their location and magnitude. To avoid the introduction of binarisation thresholds, which would burden the optimisation process, a threshold-independent cosine similarity measure is adopted to evaluate the directional consistency between the pressure residual vector and the sensitivity vectors.
Our framework is tested on the L-Town benchmark WDN, a realistic WDN inspired by a real-world infrastructure, for which risk zonation is assessed considering exposure (qualitative exposed value of real estate assets, population density, and road network importance) and hazard (operational and pipe intrinsic factors). The numerical results demonstrate that the approach is effective in choosing the set of sensors that reduces the distance between the predicted leak localisation and the actual leak point, rewarding urban areas characterised by a higher potential HDL risk.
How to cite: Medio, G., Cozzolino, L., Varra, G., Della Morte, R., and Cominola, A.: Mitigation of Hydrogeological Risk Caused by Leakage in Urban Water Distribution Networks: An Optimal Sensor Placement Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17061, https://doi.org/10.5194/egusphere-egu26-17061, 2026.
Despite accelerating digitalization and the wide availability of artificial intelligence tools, many water distribution systems (WDS) still lack system-scale customer monitoring, advanced metering infrastructures, or up-to-date calibrated hydraulic models. For the majority of such systems, leak detection remains predominantly data-driven, relying on time series from a sparse set of pressure sensors and flow meters, the placement of which is typically determined by expert knowledge. Sensor data may be augmented by simplified network graphs or uncalibrated hydraulic models, yet insights remain limited and the problem highly imbalanced, as the small number of (engineered) leak events restricts both model training and evaluation.
Under these challenging conditions, combinations of time series analysis and machine learning models have shown strong potential for automated, data-driven leak detection. However, the amount, quality, and structure of data required for robust model development and evaluation can hinder practical implementation. A limited number of devices, operational constraints, and environmental risks often lead to temporary installations and repeated sensor relocation, further reducing the availability of consistent training data.
This work investigates the potential of hybrid modeling approaches, transfer learning strategies, and the encoding of distribution system structure for data-driven leak detection under these constraints. Data quality and temporal granularity are examined at the sensor level. By mapping edge computing concepts to structural units inherent to WDS (e.g., district metered areas), the performance of anomaly detection algorithms is evaluated across different sensor combinations and spatial scales.
Model development and evaluation are based on high-resolution hydraulic (pressure and flow) time series and operational data collected over three years in an operational WDS. Beyond assessing the proposed methodologies, this study enables an in-depth discussion of the limitations of data-driven leak detection under conditions of incomplete instrumentation and expert sensor placement.
Acknowledgements: The data for this work was generated during research project “ADAM - Advanced data-driven modeling of water distribution systems” funded by Vienna Water.
How to cite: Pointl, M. and Fuchs-Hanusch, D.: Leak Detection under Sparse Instrumentation: Implications for Data-Driven Methods in Water Distribution Systems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17580, https://doi.org/10.5194/egusphere-egu26-17580, 2026.
The demand for drinking water has noticeably increased due to ever-increasing population growth and climate change. In addition, a significant amount of water is lost from water distribution networks (WDNs) through leakages because of pipe aging, corrosion, structural flaws, etc. Consequently, detecting leakage in WDNs can play an essential role in maintaining sustainable water-supply infrastructures.
Recently, hybrid approaches combining different methods, such as statistical, hydraulic, and machine learning (ML) methods, have attracted significant attention among the scientific community. In light of the above, this study presents a hybrid method for leakage detection that relies on two steps: i) a robust ML-based procedure to identify the leakage area, and ii) an optimization algorithm to further narrow down the suspect leakage area detected in the first step.
In the first step, the WDN is divided into leakage areas using the k-means clustering algorithm. Then, a Support Vector Machine (SVM) algorithm is used to identify the suspect leakage area. Pressure differences (i.e., pressure differences between 24h leakage scenarios and the daily average pressure of the leakage-free scenario as baseline condition) are considered as input, and leakage areas are assumed as targets for the SVM algorithm. Moreover, noise related to demand pattern uncertainty and pressure sensor inaccuracy is added to the model. In the second step, a two-step optimization algorithm is applied. It relies on: a minimization process based on a derivative-free optimizer that reduces the difference between simulated and measured data at the pressure/flow sensors placed in the WDN, and a filtering-clustering-ranking algorithm that eliminates nodes where the leaked volume is assumed to be negligible by giving a priority list of nodes for further inspection.
The proposed method was tested on L-Town, a large-scale WDN used as benchmark for the Battle of the Leakage Detection and Isolation Methods (BattLeDIM). The preliminary results indicate that the proposed approach can effectively identify leakage areas, especially in large-scale WDNs, potentially offering a practical tool to water utilities managing complex distribution networks.
How to cite: Haghdoost, M., D’Aniello, A., Pianese, D., and Cimorelli, L.: A novel hybrid approach for leakage area detection in water distribution networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19741, https://doi.org/10.5194/egusphere-egu26-19741, 2026.
Reliable and explainable leakage localization in water distribution networks (WDNs) is critical for minimizing non-revenue water, reducing inspection time, and improving the operational resilience of WDNs. Explainability is particularly important because leakage localization results directly inform high-cost and high-risk operational decisions, such as field inspections and pipe excavations, and must therefore be transparent and trustworthy to system operators. Machine Learning (ML) approaches have recently shown strong potential for leakage localization, which commonly formulated as a multi-class classification problem. In this setting, a WDN is first partitioned into different zones, and the ML model is then trained to predict the most likely zone containing the leakage. In previous work, four high-performing tree-based classifiers, including Random Forest, Gradient Boosting, XGBoost, and LightGBM, and three neural network models of increasing architectural depth (shallow NN, deep NN, and extra-deep NN) were trained and comparatively evaluated. While all implemented ML models performed well individually when the number of classes was small, their performance degraded to varying degrees as the number of classes increased. However, the differential performance across algorithms suggests potential for ensemble methods to highlight their complementary strengths. This work proposes an explainable ensemble ML framework for multi-class leak zone classification using pressure measurements that systematically combines the outputs from different ML models, thus strengthening the robustness beyond individual approaches. Building on prior evaluations of individual models, different classifiers are evaluated by integrating the outputs of multiple models using three complementary ensemble strategies: majority voting, which combines discrete leak localisation decisions; weighted averaging, which assigns reliability-based weights to individual model predictions; and stacking, where a meta-model is trained to learn how to optimally combine the outputs of several base classifiers.
Model performance is evaluated through multiple metrics including classification accuracy, precision, recall, and F1-scores, complemented by confusion matrix analysis and computational efficiency measurements. Additionally, a new metric, called Maximum Pipe Length Search (MPLS) is applied, which provides a physically interpretable measure of inspection effort on-side. MPLS quantifies the cumulative pipe length that would be inspected if operators followed the model-generated ranking of likely leakage zones until reaching the correct one. This metric bridges model predictions with actionable field strategies, offering a practical lens for utilities to compare model outputs in operational terms. This research investigates whether ensemble approaches can provide key advantages in the context of leakage localisation, including increased robustness to noisy sensor measurements, mitigation of the limitations of individual models, and improved generalisability across varying hydraulic and operational conditions.
Funding
This publication was produced as part of the "FOUND" project. This project is funded by the Federal Ministry of Agriculture and Forestry, Climate and Environmental Protection, Regions and Water Management (BMLUK) (Austria) (Project C300198).
How to cite: Mastouri, I., Oberascher, M., Steins, E., Cominola, A., Rejeb, L., and Sitzenfrei, R.: An Ensemble Learning Approach for Leakage Localization in Water Distribution Networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4808, https://doi.org/10.5194/egusphere-egu26-4808, 2026.
The optimal design of Water Distribution Networks (WDNs) is an NP-hard problem governed by nonlinear hydraulics and exponentially increasing discrete design choices. Conventional optimisation methods frequently struggle with computational burden, scalability limits, and premature convergence. Recent graph theory (GT) and complex network analysis (CNA) approaches offer rapid diameter assignment but rely on surrogate friction weights and lack topographic and hydraulic integration. To address these limitations, we introduce a scalable probabilistic growth algorithm inspired by the energy-minimising evolution of natural river networks. The method evaluates candidate connections using a composite metric of flow, distance, and elevation, while incorporating full hydraulic feedback by running EPANET at every iteration. The algorithm was tested on benchmark networks of increasing size, including the GoYung network, a large network with 3,558 nodes, and the 150,630-pipe VertRome network, which is beyond the computational reach of traditional evolutionary algorithms. The proposed approach achieved optimal solutions for GoYung and high-quality designs for the larger networks with significantly reduced computation times. Overall, this probabilistic framework provides an efficient, hydraulically informed, and highly scalable methodology for large-scale WDN optimisation.
How to cite: Sahu, S. K., Borse, D., and Biswal, B.: Beyond Evolutionary Algorithms: A Scalable River-Network Approach to Water Distribution Network Design, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1173, https://doi.org/10.5194/egusphere-egu26-1173, 2026.
Controlling pressure in urban water distribution systems is a key operational challenge, as excessive pipeline pressure can lead to leaks, pipe bursts, and significant economic losses for utilities. This issue is relevant in networks with intermittent water supply, where frequent cycles of rapid filling and emptying generate transient pressures that, if not properly managed, can lead to structural damage and reduced operational reliability. This study investigates the air expulsion process and transient pressure response in a rapidly filling pipeline, with emphasis on the effect of a downstream orifice for controlled air release. A two-dimensional CFD model was developed in ANSYS Fluent using the Volume of Fluid (VOF) multiphase approach and validated against experimental measurements. The effects of orifice size and water column length on transient pressures and the associated air–water interactions were systematically examined. Results show that transient pressure behavior is strongly governed by orifice size. Without an orifice, the air-cushioning effect is maintained, leading to lower pressure peaks and smoother transients. Larger orifices promote rapid air release, weakening the cushioning effect and producing water-hammer–dominated transients with significantly higher-pressure amplitudes. In contrast, smaller orifices partially release air while compressing the trapped air until water slams against the pipe end, creating a temporary water block. In this case, the air is not fully expelled, and both cushioning and water-hammer effects occur simultaneously. These results enable the identification of orifice size ranges that control the transition between air-cushioned filling and water-hammer-dominated response for practical air-release design. The simulations also capture temperature variations in the entrapped air, providing additional insight into the thermodynamic interactions during pipeline filling and air expulsion. Overall, the numerical framework captures the full range of transient behaviours associated with rapid filling and air expulsion and offers practical guidance for designing safer filling strategies and controlling pressure in urban water pipelines.
How to cite: Kumar, M. R. A., Hanmaiahgari, P. R., and Lambert, M.: Numerical investigation of air expulsion and transient pressurisation during rapid filling of water distribution pipelines, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2139, https://doi.org/10.5194/egusphere-egu26-2139, 2026.
Smart metering systems are one of the key components of the digital transformation of the water sector and represent a significant advancement over traditional meters. However, it has been widely demonstrated that only data gathered at a sufficiently fine resolution (1 min or even better 1 sec) allows properly performing end-use disaggregation and classification while due to technical reasons, such as battery life, smart metered water consumption data are generally registered and collected by Water Utilities at daily or hourly time steps. As an alternative to approaches relying on data provided by smart meters, methods for end-use water consumption characterization based on pressure data have progressively gained attention, due to the technical and economic advantages associated with the installation of pressure sensors compared to flow meters.
This study presents an innovative method for estimating water consumption events based on pressure measurements at two distinct in-line sections of the domestic inlet pipe, from which the head-loss time series can be derived. As first phase, the method converts the head-loss time series into flowrate by exploiting the pressure-flowrate relationship, allowing the reconstruction of the water-consumption time series. As second phase, information about water consumption at the level of individual consumption events is obtained. The flowrate time series is firstly processed by an algorithm for signal stabilization and combined events segmentation. Consequently, all individual events are analysed based on their features (e.g. duration, volume, etc.) to provide further information on water uses.
To validate the methodology, a residential user consisting of a single-family house was considered and pressure monitoring at two sections of the inlet pipe was performed over a period of about one month and a half with 1-s resolution. In addition, daily volume supplied to the user over the same period was obtained through the mechanical flowmeter for method validation. The pressure signals were converted in head-loss time series by accounting for sensor-elevation difference estimated over a time window of nil flow in the inlet pipe. Head-loss time series was then converted in flowrate time series by exploiting the relation between head-loss and flowrate, for which the hydraulic resistance of the inlet-pipe segment was preliminarily assessed through field tests. The total water consumption estimated over the monitoring period deviated from the observed one (i.e. that obtained from water-meter readings) of about 2.3%, confirming the capability of the methodology of effectively providing flowrate time series starting from pressure data. Flowrate time series was then subject to filtering and segmentation, resulting in over 7,500 individual end-use events, 18% of which overlapped in time. The characteristics of the above events were then investigated in a duration-volume mesh. Overall, the methodology was proven to provide insights into end uses of water that can support water utilities in the characterization and modelling of residential water consumption by exclusively relying on pressure data.
How to cite: Alvisi, S., Marsili, V., and Mazzoni, F.: Decoding residential water use through high-resolution pressure sensing , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8361, https://doi.org/10.5194/egusphere-egu26-8361, 2026.
Urban water scarcity, intensified by climate change and seasonal variability, is increasingly driving cities toward the use of reclaimed water as a supplementary supply. Reclaimed water, derived from treated wastewater or greywater, can significantly reduce pressure on potable water resources, yet its large-scale integration into existing urban water infrastructure remains poorly optimized. In particular, most existing reclaimed water distribution networks (RWDNs) are designed either manually or using computationally intensive methods and rarely account for their functional interactions with potable water distribution networks (PWDNs). This is critical because reductions in potable water demand due to reclaimed water use can lead to adverse water-quality effects in PWDNs, especially in low-density urban areas.
This work presents a citywide, integrated, and computationally efficient framework for the optimal design of RWDNs that explicitly considers their interdependence with PWDNs and urban spatial structure. The framework is capable of generating and evaluating a wide spectrum of RWDN configurations, ranging from fully centralized systems to highly decentralized, multi-source networks. It relies on openly available spatial data, including street networks, land-use information, and digital elevation models, making it transferable to different cities.
The framework automatically generates initial RWDN layouts based on the correlation between street networks and water networks and refines them using information on land plots, topography, spatial demand patterns in the PWDN, and reclaimed water origin–destination relationships. Based on this integrated spatial analysis, a wide range of centralized and decentralized network configurations is produced, and optimal pipe diameters are determined using efficient graph-based optimization methods. Each candidate network is automatically evaluated with respect to its impact on the existing PWDN. This enables the identification of feasible designs that satisfy water reuse requirements while minimizing adverse water-quality effects in the potable water system. At the same time, it explores how many small installations can be meaningfully interconnected and optimally coordinated with the central supply network so that water availability and demand match in space and time.
The final centralized and decentralized RWDNs are compared in terms of cost, water savings, and overall system performance. Application of the framework to a large European city demonstrates that well-designed decentralized and hybrid reclaimed water systems can substantially reduce potable water demand while maintaining acceptable water quality in PWDNs, highlighting the importance of integrated planning for urban water reuse.
Funding: This research was funded by the Austrian Science Fund (FWF) [10.55776/P36737].
How to cite: Hajibabaei, M., Rajabi, M., and Sitzenfrei, R.: Integrated Design of Centralized and Decentralized Reclaimed Water Networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14461, https://doi.org/10.5194/egusphere-egu26-14461, 2026.
