Shall we always use hydraulic models? A graph neural network metamodel for water system calibration and uncertainty assessment.

Representing reality in a numerical model is complex. Conventionally, hydraulic models of water distribution networks are a tool for replicating water supply system behaviour through simulation by means of approximation of physical equations. A calibration process is mandatory to achieve plausible simulation results. However, calibration is affected by a set of intrinsic uncertainty sources, mainly related to the lack of system knowledge. This paper proposes a breakthrough approach for calibrating hydraulic models through a graph machine learning approach. The main idea is to create a graph neural network metamodel to estimate the network behaviour based on a limited number of monitoring sensors. Once the flows and pressures of the entire network have been estimated, a calibration is carried out to obtain the set of hydraulic parameters that best approximates the metamodel. Through this process, it is possible to estimate the uncertainty that is transferred from the few available measurements to the final hydraulic model. The paper sparks a discussion to assess under what circumstances a graph-based metamodel might be a solution for water network analysis.

Dual graph characteristics of water distribution networks-how optimal are design solutions?

Urban water infrastructures are an essential part of urban areas. For their construction and maintenance, major investments are required to ensure an efficient and reliable function. Vital parts of the urban water infrastructures are water distribution networks (WDNs), which transport water from the production (sources) to the spatially distributed consumers (sinks). To minimize the costs and at the same time maximize the resilience of such a system, multi-objective optimization procedures (e.g., meta-heuristic searches) are performed. Assessing the hydraulic behavior of WDNs in such an optimization procedure is no trivial task and is computationally demanding. Further, deciding how close to optimal design solutions the current solutions are, is difficult to assess and often results in an unnecessary extent of experiment. To tackle these challenges, an answer to the questions is sought: when is an optimization stage achieved from which no further improvements can be expected, and how can that be assessed? It was found that graph characteristics based on complex network theory (number of dual graph elements) converge towards a certain threshold with increasing number of generations. Furthermore, a novel method based on network topology and the demand distribution in WDNs, specifically based on changes in 'demand edge betweenness centrality', for identifying that threshold is developed and successfully tested. With the proposed novel approach, it is feasible, prior to the optimization, to determine characteristics that optimal design solutions should fulfill, and thereafter, test them during the optimization process. Therewith, numerous simulation runs of meta-heuristic search engines can be avoided.

Graph method for critical pipe analysis of branched and looped drainage networks.

Enhancing resilience of drainage networks is a crucial practice to protect both humans and nature. One way to enhance resilience is to identify critical parts of drainage networks for targeted management and maintenance strategies. While hydrodynamic modelling approaches for identification are computationally intensive, in this study, a novel method based on complex network analysis is used to determine the most critical pipes in a benchmark and a real network of an Alpine municipality. For evaluation, the results of the proposed graph method are compared with hydrodynamic simulations in terms of accuracy and computational time. Results show that the proposed method is very accurate (R2 = 0.98) for branched benchmark network while the accuracy reduces slightly for the more complex real network (R2 = 0.96). Furthermore, the accuracy of the proposed method decreases with increasing loop degree and when the system is pressured with higher return period rainfall. Although the outcomes of the proposed method show slight differences to hydrodynamic modelling, it is still very useful because the computational time and data required are much less than a hydrodynamic model.

Smart water campus - a testbed for smart water applications.

The Internet of Things concept includes low-cost sensors in combination with innovative wireless communication technology, supporting a large-scale implementation of measurement equipment in the field of urban water infrastructure (UWI). At present, the potentials of such smart solutions are often unclear, making it difficult for decision-makers to justify investments. To address this shortcoming, the Smart Campus is represented as an innovative testbed for smart and data-driven applications in the field of network-based UWI. During the last few years, the campus area of the University of Innsbruck has been comprehensively equipped with a variety of low-cost sensors for monitoring and controlling the UWI in high resolution (1-15 min). The experiences showed that the quality of service is influenced by the choice of communication technology and the installation location, thereby affecting the desired applications. Additionally, water distribution and urban drainage network including nature-based solutions have been integrated into an overall monitored system extended by measures to involve the urban population. This integrative approach allows the usage of synergies for the implementation and supports cross-system improvements (e.g., smart rainwater harvesting). However, an integration of different participants also implies new requirements for the project team (e.g., including social science).

Battle of centralized and decentralized urban stormwater networks: From redundancy perspective.

Recent research underpinned the effectiveness of topological decentralization for urban stormwater networks (USNs) during the planning stage in terms of both capital savings and resilience enhancement. However, how centralized and decentralized USNs' structures with various degrees of redundancy (i.e., redundant water flow pathways) project resilience under functional and structural failure remains an unresolved issue. In this work, we present a systemic and generic framework to investigate the impact of adding redundant flow paths on resilience based on three strategies for optimal centralized versus decentralized USNs. Furthermore, a tailored graph-theory based measure (i.e., eigenvector centrality) is proposed to introduce redundant paths to the critical locations of USNs. The proposed framework is then applied to a real large-scale case study. The results confirm the critical role of layout decentralization under both functional (e.g., extreme precipitation events), and structural failure (e.g., pipe collapse). Moreover, the findings indicate that the implementation of redundant paths could increase resilience performance by up to 8% under functional failure without changing the network's major structural characteristics (i.e., sewer diameters, lengths, and storage capacity), only by leveraging the effective flow redistribution. The scheme proposed in this study can be a fruitful initiative for further improving the USNs' resilience during both planning and rehabilitation stages.

Using complex network analysis for water quality assessment in large water distribution systems.

Assessing and modelling the water quality in a water distribution system (WDS) are highly important to ensure a reliable supply with sufficient water quality. Owing to the high computational burden of such an analysis, frequently, simplifications are required or surrogate models are used (e.g., reducing the level of detail of the network model), neglecting significant aspects. For large (currently all-pipe) models and/or recurrent simulations (e.g., integrated studies, sensitivity analysis, deep uncertainty analysis, design, and optimization), the computational burden further increases. In this study, a novel complex network analysis-based approach for high-computational efficiency water quality assessment in a WDS is developed and comprehensively tested (R² values in comparison with state-of-the-art nodal water qualities in median of 0.95 are achieved). The proposed model is successfully utilized in a design study to identify the design solutions exceeding water quality thresholds with a correct identification rate between 96% and 100%. The computational efficiency is determined to be a factor 4.2e-06 less than that of state-of-the-art models. Therefore, the proposed model significantly improves the water quality assessment for such tasks in large WDSs.

Efficient integration of IoT-based micro storages to improve urban drainage performance through advanced control strategies.

The smart rain barrel (SRB) consists of a conventional RB with storage volumes between 200 and 500 L, which is extended by a remotely (and centrally) controllable discharge valve. The SRB is capable of releasing stormwater prior to precipitation events by using high-resolution weather forecasts to increase detention capacity. However, as shown in a previous work, a large-scale implementation combined with a simultaneous opening of discharge valves clearly reduced the effectiveness. The aim of this work was to systematically investigate different control strategies for wet weather by evaluating their impact on sewer performance. For the case study, an alpine municipality was hypothetically retrofitted with SRBs (total additional storage volume of 181 m3). The results showed that combined sewer overflow (CSO) volume and subsequently pollution mass can be reduced by between 7 and 67% depending on rain characteristics (e.g., rain pattern, amount of precipitation) and an applied control strategy. Effectiveness of the SRBs increases with lower CSO volume, whereas more advanced control strategies based on sewer conditions can clearly improve the system's performance compared to simpler control strategies. For higher CSO volume, the SRBs can postpone the start of an CSO event, which is important for a first-flush phenomenon.

Using Complex Network Analysis for Optimization of Water Distribution Networks.

The optimization of water networks supports the decision-making process by identifying the optimal trade-off between costs and performance (e.g., resilience and leakage). A major challenge in the domain of water distribution systems (WDSs) is the network (re)design. While the complex nature of WDS has already been explored with complex network analysis (CNA), literature is still lacking a CNA of optimal water networks. Based on a systematic CNA of Pareto-optimal solutions of different WDSs, several graph characteristics are identified, and a newly developed CNA design approach for WDSs is proposed. The results show that obtained designs are comparable with results found by evolutionary optimization, but the CNA approach is applicable for large networks (e.g., 150,000 pipes) with a substantially reduced computational effort (runtime reduction up to 5 orders of magnitude).

Future trajectories of urban drainage systems: A simple exploratory modeling approach for assessing socio-technical transitions.

In this work, we present a novel approach to explore future trajectories in urban drainage systems, emphasizing the adoption and implementation of sustainable 'nature-based' stormwater management strategies. The focus is on the development and long-term assessment of socio-technical pathways to create a multifunctional stormwater system at the city scale. The innovation is to identify and represent the socio-technical pathways by means of adoption curves for such transition processes. We combine urban planning policies and state-of-the-art urban engineering approaches with societal aspects and analyze them with traditional biophysical models (hydrologic-hydraulic sewer modeling). In doing so, different pathways from a current to a future system state are investigated under a variety of political, population and climate scenarios. Results allow for strategy screening by addressing the spatial and temporal implementation of decentralized stormwater control measures, to enable a successful transition to a sustainable future city. The model is applied to an ongoing transition of Kiruna, a city in Sweden, considering 36 different future trajectories over a transition period of 23 years. Results show that the trajectory of raingarden implementation under a sustainability policy can alleviate the adverse effects of urbanization (growth scenario). While this trajectory resulted in, for example, nearly the same sewer surcharge performance as that characterized by declining urbanization (stagnation) and a business-as-usual policy (with expected raingarden uptake rates approximately one-third lower), significantly better ecological performances (e.g. runoff treatment ratios up to 50%) are achieved.

On the sensitivity of geospatial low impact development locations to the centralized sewer network.

In the future, infrastructure systems will have to become smarter, more sustainable, and more resilient requiring new methods of urban infrastructure design. In the field of urban drainage, green infrastructure is a promising design concept with proven benefits to runoff reduction, stormwater retention, pollution removal, and/or the creation of attractive living spaces. Such 'near-nature' concepts are usually distributed over the catchment area in small scale units. In many cases, these above-ground structures interact with the existing underground pipe infrastructure, resulting in hybrid solutions. In this work, we investigate the effect of different placement strategies for low impact development (LID) structures on hydraulic network performance of existing drainage networks. Based on a sensitivity analysis, geo-referenced maps are created which identify the most effective LID positions within the city framework (e.g. to improve network resilience). The methodology is applied to a case study to test the effectiveness of the approach and compare different placement strategies. The results show that with a simple targeted LID placement strategy, the flood performance is improved by an additional 34% as compared to a random placement strategy. The developed map is easy to communicate and can be rapidly applied by decision makers when deciding on stormwater policies.

Morphogenesis of Urban Water Distribution Networks: A Spatiotemporal Planning Approach for Cost-Efficient and Reliable Supply.

Cities and their infrastructure networks are always in motion and permanently changing in structure and function. This paper presents a methodology for automatically creating future water distribution networks (WDNs) that are stressed step-by-step by disconnection and connection of WDN parts. The associated effects of demand shifting and flow rearrangements are simulated and assessed with hydraulic performances. With the methodology, it is possible to test various planning and adaptation options of the future WDN, where the unknown (future) network is approximated via the co-located and known (future) road network, and hence different topological characteristics (branched vs. strongly looped layout) can be investigated. The reliability of the planning options is evaluated with the flow entropy, a measure based on Shannon's informational entropy. Uncertainties regarding future water consumption and water loss management are included in a scenario analysis. To avoid insufficient water supply to customers during the transition process from an initial to a final WDN state, an adaptation concept is proposed where critical WDN components are replaced over time. Finally, the method is applied to the drastic urban transition of Kiruna, Sweden. Results show that without adaptation measures severe performance drops will occur after the WDN state 2023, mainly caused by the disconnection of WDN parts. However, with low adaptation efforts that consider 2-3% pipe replacement, sufficient pressure performances are achieved. Furthermore, by using an entropy-cost comparison, the best planning options are determined.

Info-Gap robustness pathway method for transitioning of urban drainage systems under deep uncertainties.

In the urban water cycle, there are different ways of handling stormwater runoff. Traditional systems mainly rely on underground piped, sometimes named 'gray' infrastructure. New and so-called 'green/blue' ambitions aim for treating and conveying the runoff at the surface. Such concepts are mainly based on ground infiltration and temporal storage. In this work a methodology to create and compare different planning alternatives for stormwater handling on their pathways to a desired system state is presented. Investigations are made to assess the system performance and robustness when facing the deeply uncertain spatial and temporal developments in the future urban fabric, including impacts caused by climate change, urbanization and other disruptive events, like shifts in the network layout and interactions of 'gray' and 'green/blue' structures. With the Info-Gap robustness pathway method, three planning alternatives are evaluated to identify critical performance levels at different stages over time. This novel methodology is applied to a real case study problem where a city relocation process takes place during the upcoming decades. In this case study it is shown that hybrid systems including green infrastructures are more robust with respect to future uncertainties, compared to traditional network design.

Investigating the interactions of decentralized and centralized wastewater heat recovery systems.

In the urban water cycle there are different sources for extracting energy. In addition to potential and chemical energy in the wastewater, thermal energy can also be recovered. Heat can be recovered from the wastewater with heat exchangers that are located decentralized and/or centralized at several locations throughout the system. It can be recovered directly at the source (e.g. in the showers and bathrooms), at building block level (e.g. warm water tanks collecting all grey water), in sewers or at the wastewater treatment plant. However, an uncoordinated installation of systems on such different levels can lead to competing technologies. To investigate these interactions, a modelling environment is set up, tested and calibrated based on continuous sewer temperature and flow measurements. With that approach different heat recovery scenarios on a household level (decentralized) and of in-sewer heat recovery (centralized) are investigated. A maximum performance drop of 40% for a centralized energy recovery system was estimated when all bathrooms are equipped with decentralized recovery systems. Therefore, the proposed modelling approach is suitable for testing different future conditions and to identify robust strategies for heat recovery systems from wastewater.

Simplifying impact of urban development on sewer systems.

Linking urban development and urban drainage models is a more and more popular approach when impacts of pavement of urban areas on sewer system performance are evaluated. As such an approach is a difficult task, this is not a feasible procedure for everyday engineering practice. We propose an alternative method, based on a developed simple near-quadratic relationship, which directly translates change (increase or decrease) of paved area into a change in the return period (RP) of the design rainfall event or design rainfall intensity. This formula is simple to use and compatible with existing design guidelines. A further advantage is that the calculated design RP can also be used to communicate the impact of a change in impervious areas to stakeholders or the public community. The method is developed using a set of 250 virtual and two real-world case studies and hydrodynamic simulations. It is validated on a small catchment for which we compare system performance and redesigned pipe diameters. Of course such a simplification contains different uncertainties. But these uncertainties have to be seen in the context of overall uncertainties when trying to predict city development into the future. Hence it still is a significant advantage compared to today's engineering practice.

Assessing the impact of transitions from centralised to decentralised water solutions on existing infrastructures--integrated city-scale analysis with VIBe.

Traditional urban water management relies on central organised infrastructure, the most important being the drainage network and the water distribution network. To meet upcoming challenges such as climate change, the rapid growth and shrinking of cities and water scarcity, water infrastructure needs to be more flexible, adaptable and sustainable (e.g., sustainable urban drainage systems, SUDS; water sensitive urban design, WSUD; low impact development, LID; best management practice, BMP). The common feature of all solutions is the push from a central solution to a decentralised solution in urban water management. This approach opens up a variety of technical and socio-economic issues, but until now, a comprehensive assessment of the impact has not been made. This absence is most likely attributable to the lack of case studies, and the availability of adequate models is usually limited because of the time- and cost-intensive preparation phase. Thus, the results of the analysis are based on a few cases and can hardly be transferred to other boundary conditions. VIBe (Virtual Infrastructure Benchmarking) is a tool for the stochastic generation of urban water systems at the city scale for case study research. With the generated data sets, an integrated city-scale analysis can be performed. With this approach, we are able to draw conclusions regarding the technical effect of the transition from existing central to decentralised urban water systems. In addition, it is shown how virtual data sets can assist with the model building process. A simple model to predict the shear stress performance due to changes in dry weather flow production is developed and tested.

A planning algorithm for quantifying decentralised water management opportunities in urban environments.

With global change bringing about greater challenges for the resilient planning and management of urban water infrastructure, research has been invested in the development of a strategic planning tool, DAnCE4Water. The tool models how urban and societal changes impact the development of centralised and decentralised (distributed) water infrastructure. An algorithm for rigorous assessment of suitable decentralised stormwater management options in the model is presented and tested on a local Melbourne catchment. Following detailed spatial representation algorithms (defined by planning rules), the model assesses numerous stormwater options to meet water quality targets at a variety of spatial scales. A multi-criteria assessment algorithm is used to find top-ranking solutions (which meet a specific treatment performance for a user-defined percentage of catchment imperviousness). A toolbox of five stormwater technologies (infiltration systems, surface wetlands, bioretention systems, ponds and swales) is featured. Parameters that set the algorithm's flexibility to develop possible management options are assessed and evaluated. Results are expressed in terms of 'utilisation', which characterises the frequency of use of different technologies across the top-ranking options (bioretention being the most versatile). Initial results highlight the importance of selecting a suitable spatial resolution and providing the model with enough flexibility for coming up with different technology combinations. The generic nature of the model enables its application to other urban areas (e.g. different catchments, local municipal regions or entire cities).

Automatic generation of water distribution systems based on GIS data.

In the field of water distribution system (WDS) analysis, case study research is needed for testing or benchmarking optimisation strategies and newly developed software. However, data availability for the investigation of real cases is limited due to time and cost needed for data collection and model setup. We present a new algorithm that addresses this problem by generating WDSs from GIS using population density, housing density and elevation as input data. We show that the resulting WDSs are comparable to actual systems in terms of network properties and hydraulic performance. For example, comparing the pressure heads for an actual and a generated WDS results in pressure head differences of ±4 m or less for 75% of the supply area. Although elements like valves and pumps are not included, the new methodology can provide water distribution systems of varying levels of complexity (e.g., network layouts, connectivity, etc.) to allow testing design/optimisation algorithms on a large number of networks. The new approach can be used to estimate the construction costs of planned WDSs aimed at addressing population growth or at comparisons of different expansion strategies in growth corridors.