Research on nowcasting prediction technology for flooding scenarios based on data-driven and real-time monitoring.

With the impact of global climate change and the urbanization process, the risk of urban flooding has increased rapidly, especially in developing countries. Real-time monitoring and prediction of flooding extent and drainage system are the foundation of effective urban flood emergency management. Therefore, this paper presents a rapid nowcasting prediction method of urban flooding based on data-driven and real-time monitoring. The proposed method firstly adopts a small number of monitoring points to deduce the urban global real-time water level based on a machine learning algorithm. Then, a data-driven method is developed to achieve dynamic urban flooding nowcasting prediction with real-time monitoring data and high-accuracy precipitation prediction. The results show that the average MAE and RMSE of the urban flooding and conduit system in the deduction method for water level are 0.101 and 0.144, 0.124 and 0.162, respectively, while the flooding depth deduction is more stable compared to the conduit system by probabilistic statistical analysis. Moreover, the urban flooding nowcasting method can accurately predict the flooding depth, and the R2 are as high as 0.973 and 0.962 of testing. The urban flooding nowcasting prediction method provides technical support for emergency flood risk management.

Green infrastructure layout based on a dynamic operation feature of drainage systems.

Increasingly frequent urban floods strain the traditional grey infrastructure, overwhelming the capacity of drainage networks and causing challenges in managing stormwater. The heavy precipitation leads to flooding and damage to drainage systems. Consequently, efficient mitigation strategies for flooding have been researched deeply. Green infrastructure (GI) has proved to be effective in responding the increasing risk of flood and alleviate pressure on drainage systems. However, as the primary infrastructure of stormwater management, there is still a lack of attention to the dynamic operation feature of urban sewer systems during precipitation events. To fill this gap, we proposed a novel approach that integrates hydraulic characteristics and the topological structure of a sewer network system. This approach aims to identify influential nodes, which contribute to the connectivity of the sewer network amidst dynamic changes in inflow during precipitation events. Furthermore, we adopted rain barrels to serve as exemplars of GI, and 14 GI layout schemes are produced based on the different ranks of influential nodes. Implementing GI measures on both poorly performing and well-performing nodes can yield distinct benefits in mitigating node flooding. This approach provides a new perspective for stormwater management, establishing effective synergy between GI and the drainage system.

Flow rate influence on sediment depth estimation in sewers using temperature sensors.

Enhancing sediment accumulation monitoring techniques in sewers will enable a better understanding of the build-up processes to develop improved cleaning strategies. Thermal sensors provide a solution to sediment depth estimation by passively monitoring temperature fluctuations in the wastewater and sediment beds, which allows evaluation of the heat-transfer processes in sewer pipes. This study analyses the influence of the flow conditions on heat-transfer processes at the water-sediment interface during dry weather flow conditions. For this purpose, an experimental campaign was performed by establishing different flow, temperature patterns, and sediment depth conditions in an annular flume, which ensured steady flow and room-temperature conditions. Numerical simulations were also performed to assess the impact of flow conditions on the relationships between sediment depth and harmonic parameters derived from wastewater and sediment-bed temperature patterns. Results show that heat transfer between water and sediment occurred instantaneously for velocities greater than 0.1 m/s, and that sediment depth estimations using temperature-based systems were barely sensitive to velocities between 0.1 and 0.4 m/s. A depth estimation accuracy of ±7 mm was achieved. This confirms the ability of using temperature sensors to monitor sediment build-up in sewers under dry weather conditions, without the need for flow monitoring.

Connecting blue-green infrastructure elements to reduce combined sewer overflows.

By infiltrating and retaining stormwater, Blue-Green Infrastructure (BGI) can help to reduce Combined Sewer Overflows (CSOs), one of the main causes of urban water pollution. Several studies have evaluated the ability of individual BGI types to reduce CSOs; however, the effect of combining these elements, likely to occur in reality, has not yet been thoroughly evaluated. Moreover, the CSO volume reduction potential of relevant components of the urban drainage system, such as detention ponds, has not been quantified using hydrological models. This study presents a systematic way to assess the potential of BGI combinations to mitigate CSO discharge in a catchment near Zurich (Switzerland). Sixty BGI combinations, including four BGI elements (bioretention cells, permeable pavement, green roofs, and detention ponds) and four different implementation rates (25%, 50%, 75%, and 100% of the available sewer catchment area) are evaluated for four runoff routing schemes. Results reveal that BGI combinations can provide substantial CSO volume reductions; however, combinations including detention ponds can potentially increase CSO frequency, due to runoff prolongation. When runoff from upstream areas is routed to the BGI, the CSO discharge reductions from combinations of BGI elements differ from the cumulative CSO discharge reductions achieved by individual BGI types, indicating that the sum of effects from individual BGI types cannot accurately predict CSO discharge in combined BGI scenarios. Moreover, larger BGI implementation areas are not consistently more cost-effective than small implementation areas, since the additional CSO volume reduction does not outweigh the additional costs. The best-performing BGI combination depends on the desired objective, being CSO volume reduction, CSO frequency reduction or cost-effectiveness. This study emphasizes the importance of BGI combinations and detention ponds in CSO mitigation plans, highlighting their critical factors-BGI types, implementation area, and runoff routing- and offering a novel and systematic approach to develop tailored BGI strategies for urban catchments facing CSO challenges.

Combined sewer overflow mitigation through SUDS - A review on modelling practices, scenario elaboration, and related performances.

Hydrologic-hydraulic modelling of urban catchment is an asset for land managers to simulate Sustainable Urban Drainage Systems (SUDS) implementation to fulfil combined sewer overflow (CSO) regulations. This review aims to assess the current practices in modelling SUDS scenarios at large scale for CSO mitigation encompassing every stage of the modelling process from the choice of the equation to the validation of the initial state of the urban system, right through to the elaboration, modelling, and selection of SUDS scenarios to evaluate their performance on CSO. Through a quantitative and qualitative analysis of 50 published studies, we found a diversity of choices when modelling the status quo of the urban system. Authors generally do not explain the modelling processes of slow components (deep infiltration, groundwater infiltration) and interconnexion between SUDS and the sewer system. In addition, only a few authors explain how CSO structures are modelled. Furthermore, the modelling of SUDS implementation at catchment scale is highlighted in the 50 studies retrieved with three different approaches going from simplified to detailed. SUDS modelling choices seem to be consistent with the objectives: studies focusing on dealing with several objectives at the time typically opt for a complex system configuration that includes the surface processes, network, CSO, SUDS, and often the soil and/or groundwater components. Conversely, authors who have selected a basic configuration generally aim to address a single, straightforward question (e.g., which type of SUDS). However, elaboration and selection of scenarios for CSO mitigation is mainly based on local constraints, which does not allow hydrological performance to be directly optimised. In conclusion, to improve current practices in modelling SUDS scenarios at large scale for CSO mitigation, authors suggest to: (i) improve clear practices of CSO modelling, calibration and validation at the urban catchment scale, (ii) develop methods to optimize the performance of scenarios for CSO mitigation using hydrological drivers, and (iii) improve parsimonious and user-friendly models to simulate SUDS scenarios in a context of data scarcity.

Integrated monitoring and modeling to disentangle the complex spatio-temporal dynamics of urbanized streams under drought stress.

We have a poor understanding of how urban drainage and other engineered components interact with more natural hydrological processes in green and blue spaces to generate stream flow. This limits the scientific evidence base for predicting and mitigating the effects of future development of the built environment and climate change on urban water resources and their ecosystem services. Here, we synthesize > 20 years of environmental monitoring data to better understand the hydrological function of the 109-km2 Wuhle catchment, an important tributary of the river Spree in Berlin, Germany. More than half (56%) of the catchment is urbanized, leading to substantial flow path alterations. Young water from storm runoff and rapid subsurface flow provided around 20% of stream flow. However, most of it was generated by older groundwater (several years old), mainly recharged through the rural headwaters and non-urban green spaces. Recent drought years since 2018 showed that this base flow component has reduced in response to decreased recharge, causing deterioration in water quality and sections of the stream network to dry out. Attempts to integrate the understanding of engineered and natural processes in a traditional rainfall-runoff model were only partly successful due to uncertainties over the catchment area, effects of sustainable urban drainage, adjacent groundwater pumping, and limited conceptualization of groundwater storage dynamics. The study highlights the need for more extensive and coordinated monitoring and data collection in complex urban catchments and the use of these data in more advanced models of urban hydrology to enhance management.

An improved global resilience assessment method for urban drainage systems: A case study of Haidian Island, south China.

Resilience assessment for urban drainage systems is a fundamental aspect of building resilient cities. Recently, some scholars have proposed the Global Resilience Analysis (GRA) method, which assesses resilience based on the functional performance of different system failure scenarios. Compared to traditional system dynamics methods, the GRA method considers the impact of internal structural failure on resilience but requires a large amount of computation. This research proposed an improved GRA method to enhance computational efficiency and practicality by reducing the number of system scenario simulations. Firstly, a hydrodynamic model of the drainage network of Haidian Island has been constructed using the Storm Water Management Model (SWMM) and Python. Secondly, the GRA method was improved using cluster analysis and convergence analysis to reduce the simulation scenarios. Thirdly, a resilience assessment index was established through system function functions, and two types of resilience enhancement measures, centralized and distributed, were proposed. The results show: (i) resilience assessment increases the computational efficiency by 25% compared to the traditional GRA method; (ii) the resilience index of the existing drainage network within Haidian Island is less than the design value (0.7) in all failure scenarios, indicating a lower level of recovery capability; (iii) compared to the centralized strategy, which is only effective when the system failure level is less than 9%, the distributed strategy enhances the resilience of the urban drainage system at a higher failure level (77%).

Real-time control of urban drainage systems using neuro-evolution.

With climate change and urbanization, existing urban drainage systems are being stressed beyond their design capacity in many parts of the world. Real-time control (RTC) can improve the performance of these systems and reduce the need for system upgrades. However, developing optimal control policies for RTC is a challenging research area due to computational demands, high uncertainties and system dynamics. This study presents a new RTC method using neuro-evolution for controlling combined sewer overflow (CSO) in urban drainage systems. Neuro-evolution is an approach to neural network research by evolutionary algorithms. Neuro-evolution realizes RTC by training the control policy in advance, thus avoiding the online optimization process in the application period. The simulation results of the benchmark Astlingen network indicate that the trained control policy outperforms the equal filling degree strategy in terms of CSO volume reduction and robustness in the face of tank level uncertainty. The performance analysis of the typical CSO events shows that the control policy mainly makes positive contributions during 'small' CSO events rather than 'large' ones. In particular, the effectiveness of the control policy in 'small' CSO events is more prominent in the initial phase of the events compared with the final phase. This work stands to support a foundation for future studies in the control of urban water systems based on neuro-evolution.

Exploring transitions of sewer wastewater infrastructure towards decentralisation using the modular model TURN-Sewers.

We present a new modular model called TURN-Sewers for exploring different adaptations of centralised wastewater infrastructure towards more decentralised wastewater systems under different urban development scenarios. The modular model is flexible and computationally efficient in exploring transitions at the city scale, allowing for the comparison of different policies and management strategies for sanitary wastewater infrastructure. TURN-Sewers includes independent modules that simulate the generation, dimensioning, deterioration, management, and calculation of performance indicators for different wastewater systems. This model can use readily available spatial information to support infrastructure planners and other stakeholders in exploring different transition pathways from centralised to decentralised wastewater infrastructure. An illustrative example demonstrates how TURN-Sewers can generate multiple future alternatives, define different infrastructure management strategies regarding system expansion, rehabilitation and transition, and assess the economic, hydraulic and structural impacts.

Exploring optimal deep tunnel sewer systems to enhance urban pluvial flood resilience in the gangnam region, South Korea.

Urban pluvial flooding is becoming a global concern, exacerbated by urbanization and climate change, especially in rapidly developing areas where existing sewer systems lag behind growth. In order to minimize a system's functional failures during extreme rainfalls, localized engineering solutions are required for urban areas chronically suffering from pluvial floods. This study critically evaluates the Deep Tunnel Sewer System (DTSS) as a robust grey infrastructure solution for enhancing urban flood resilience, with a case study in the Gangnam region of Seoul, South Korea. To do so, we integrated a one-dimensional sewer model with a rapid flood spreading model to identify optimal routes and conduit diameters for the DTSS, focusing on four flood-related metrics: the total flood volume, the flood duration, the peak flooding rate, and the number of flooded nodes. Results indicate that, had the DTSS been in place, it could have reduced historical flood volumes over the last decade by 50.1-99.3%, depending on the DTSS route. Regarding the conduit diameter, an 8 m diameter was found to be optimal for minimizing all flood-related metrics. Our research also developed the Intensity-Duration-Frequency (IDF) surfaces in three dimensions, providing a correlation between simulated flood-related metrics and design rainfall characteristics to distinguish the effect of DTSS on flood risk reduction. Our findings demonstrate how highly engineered solutions can enhance urban flood resilience, but they may still face challenges during extreme heavy rainfalls with a 80-year frequency or above. This study contributes to rational decision-making and emergency management in the face of increasing urban pluvial flood risks.

Automated screening of control potential with spatially explicit results to support dialogue about sewer overflow reduction and beyond.

For real-time control to become a standard measure for upgrading urban drainage systems, control potential screenings need to be easily integrated into the early planning processes that already take place. However, current screening methods are either not aligned with the present planning process, unrelatable for water managers or too time-consuming. We therefore developed an automated screening methodology through a co-design process with six Danish utilities. The process started out from a literature review, included interviews and workshops, and resulted in the control potential screening tool COPOTO. In the co-design process, utilities generally responded that indicators based solely on an assessment of static system attributes are insufficient. Thus, COPOTO instead post-processes the results of urban drainage simulation models that are commonly available. The decision context considered in initial planning phases was found to include environmental, economic, social and technical objectives that were highly case-dependent. When presenting CSO reduction potentials, the utilities therefore generally preferred interactive, spatially explicit visualisations that link the CSO reduction at a particular location to the storages and actuators that need to be activated. This enables water managers to discuss, for example, operational constraints of a considered control location. COPOTO provides such assessments with very limited manual and computational effort and thus facilitates the integration of real-time control into standard planning workflows of utilities.

Removal and fate of microplastics in permeable pavements: An experimental layer-by-layer analysis.

The increasing prevalence of microplastics (MP) in urban environments has raised concerns over their negative effects on ecosystems and human health. Stormwater runoff, and road dust and sediment, act as major vectors of these pollutants into natural water bodies. Sustainable urban drainage systems, such as permeable pavements, are considered as potential tools to retain particulate pollutants. This research evaluates at laboratory scale the efficiency of permeable interlocking concrete pavements (PICP) and porous concrete pavements (PCP) for controlling microplastics, including tire wear particles (TWP) which constitute a large fraction of microplastics in urban environments, simulating surface pollution accumulation and Mediterranean rainfall conditions. Microplastic levels in road dust and sediments and stormwater runoff inputs were 4762 ± 974 MP/kg (dry weight) and 23.90 ± 17.40 MP/L. In infiltrated effluents, microplastic levels ranged from 2.20 ± 0.61 to 5.17 ± 1.05 MP/L; while tire wear particle levels ranged between 0.28 ± 0.28 and 3.30 ± 0.89 TWP/L. Distribution of microplastics within the layers of PICP and PCP were also studied and quantified. Microplastics tend to accumulate on the pavements surface and in geotextile layers, allowing microplastic retention efficiencies from 89 % to 99.6 %. Small sized (< 0.1 mm) fragment shaped microplastics are the most common in effluent samples. The results indicate that permeable pavements are a powerful tool to capture microplastics and tire wear particles, especially by surface and geotextile layers. The study aims to shed light on the complex mobilisation mechanisms of microplastics, providing valuable insights for addressing the growing environmental concern of microplastic pollution in urban areas.

Site selection for nature-based solutions for stormwater management in urban areas: An approach combining GIS and multi-criteria analysis.

In recent years, particularly following the definition of the UN Sustainable Development Goals (SDGs) for 2030, Nature-Based Solutions (NBS) have gained considerable attention, capturing the interest of both the scientific community and policymakers committed to addressing urban environmental issues. However, the need for studies to guide decision-makers in identifying suitable locations for NBS implementation within urban stormwater management is evident. To address this gap, the present study employs a methodological approach grounded in multi-criteria analysis integrated with Geographic Information Systems (GIS) to identify areas with potential for NBS implementation. In this process, ten NBS were proposed and tested in the drainage area of a shallow tropical urban lake in Londrina, southern Brazil. Additionally, the study investigates areas hosting lower-income populations, a relevant aspect for public managers given the diverse economic subsidies required to implement NBS. Furthermore, the study incorporates a preliminary analysis that evaluates the potential ecosystem benefits to determine the most suitable NBS for a specific site. The result shows that all the ten analyzed NBS were deemed suitable for the study area. Rain barrels had the highest percentage coverage in the study area (37.1%), followed by tree pits (27.9%), and rain gardens (25.4%). Despite having the highest distribution in the basin area, rain barrels exhibited only moderate ecosystem benefits, prompting the prioritization of other NBS with more significant ecological advantages in the final integrated map. In summary, the methodology proposed showed to be a robust approach to selecting optimal solutions in densely populated urban areas.

When does infrastructure hybridisation outperform centralised infrastructure paradigms? - Exploring economic and hydraulic impacts of decentralised urban wastewater system expansion.

We explore the dynamics of centralised and decentralised wastewater infrastructure across various scenarios and introduce novel insights into their performance regarding structural vulnerability, hydraulic capacity, and costs. This study determines circumstances under which infrastructure hybridisation outperforms traditional centralised infrastructure paradigms. We combined system analysis to map out the modelling problem with the model-based exploration of the transition space using the novel TURN-Sewers model. System diagramming was used to identify the parameters or combinations of parameters that significantly influence the performance indicators being assessed. This allowed the creation of relevant simulation scenarios to identify circumstances where a decentralised sewer system could outperform a centralised one. TURN-Sewers was applied to model the infrastructure maintenance and generation of new infrastructure over 20 years for a municipality on the Swiss Plateau, considering a population growth rate of 0.03 a-1. Results show that decentralisation in expansion areas with higher densification can outperform the hydraulic performance and structural vulnerability of expanding centralised sanitary wastewater infrastructure. Decentralised systems can also offer economic advantages when capital expenditure costs for small-scale wastewater treatment plants are significantly reduced compared to current costs, particularly at higher discount rates, e.g. reaping effects of economies of scale. The findings of this study emphasise the potential of transition pathways towards decentralisation in urban water infrastructures and the value of models that allow the exploration of this transition space.

A drainage network-based impact matrix to support targeted blue-green-grey stormwater management solutions.

Urban floods will continue to be an alarming issue worldwide due to climate change and urban expansion. The costly and less environmentally friendly grey infrastructure is not always the most adequate solution to resolve urban pluvial flooding issues. The combination of grey and blue-green infrastructures, also called hybrid infrastructure, has been considered a promising solution for urban stormwater management. Existing approaches for identifying suitable hybrid solutions frequently rely on global multi-objective optimization algorithms. We developed a pre-screening method that decomposes a drainage network into clusters of pipes connected to sub-catchments, based on pipe hydraulic characteristic that allows for the impact of infrastructure combinations (blue-green and grey) to be mapped. Four impact matrices are proposed to map the total, local, upstream, and downstream flood reduction of all possible blue-green, grey, and hybrid solutions. Using an urban catchment in Guangzhou (China) as a case study, results showed that such an exercise could identify prime candidate locations for blue-green and grey infrastructure while filtering out ineffective locations for flood reduction. Furthermore, the impact matrices enabled the identification of flood zones where blue-green infrastructure could handle flood mitigation without the need of local grey infrastructure upgrades. As such, they are not only useful for quick screening of suitable interventions for each flooded zone, but can also potentially serve as a priori knowledge before diving into the data and computationally expensive process of finding the most effective flood mitigation solutions.

Improving the interpretability of deep reinforcement learning in urban drainage system operation.

Deep reinforcement learning (DRL) has been increasingly used as an adaptive and efficient solution for real-time control (RTC) of the urban drainage system (UDS). Despite the promising potential of DRL, it is a black-box model whose control logic and control consequences are difficult to be understood and evaluated. This leads to issues of interpretability and poses risks in practical applications. This study develops an evaluation framework to analyze and improve the interpretability of DRL-based UDS operation. The framework includes three analysis methods: Sobol sensitivity analysis, tree-based surrogate modelling, and conditional probability analysis. It is validated using two different DRL approaches, i.e., deep Q-learning network (DQN) and proximal policy optimization (PPO), which are trained to reduce combined sewer overflow (CSO) discharges and flooding in a real-world UDS. According to the results, the two DRLs have been shown to perform better than a rule-based control system that is currently being used. Sobol sensitivity analysis indicates that DQN is particularly sensitive to the flow of links and rainfall, while PPO is sensitive to all the states. Tree-based surrogate models effectively reveal the control logic behind the DRLs and indicate that PPO is more comprehensible but DQN is more forward-looking. Conditional probability analysis demonstrates the potential control consequences of the DRLs and identifies three situations where the DRLs are ineffective: a) the storage of UDS is fully utilized; b) peak flows have already passed through actuators; c) a substantial amount of water enters one location simultaneously. The proposed evaluation framework enhances the interpretability of DRL in UDS operations, fostering trust and confidence from operators, stakeholders, and regulators.

Identification of high oxygen-consuming substances in stormwater drainage systems illicitly connected with sewage system.

The dissolved oxygen content in water is an important indicator for assessing the quality of the water environment, and maintaining a certain amount of dissolved oxygen is essential for the healthy development of the ecological environment. When a water body is anoxic, the activity of anaerobic microorganisms increases and organic matter is decomposed to produce a large number of blackening and odorizing substances, resulting in black and odorous water bodies, which is a very common and typical phenomenon in China. Presently, there is still a relatively universal occurrence of illicitly connected stormwater and sewage pipes in the urban drainage pipe network in China, which makes oxygen-consuming substances be directly discharged into rivers through stormwater pipes and consume the dissolved oxygen in the water bodies, resulting in an oxygen deficiency of the water. This induces seasonal or year-round black and stink phenomena in urban rivers. Hence, identifying high oxygen-consuming substances, which lays the foundation for the subsequent removal of oxygen-consuming substances, is essential. Through a series of comparisons of water quality indicators and analysis of organic characteristics, it was found that the oxygen consumption capacity of domestic sewage was higher than that of industrial wastewater in the selected area of this study, and the oxygen-consuming substances of domestic sewage were small molecular amino acids. By comparing 20 conventional free amino acids, it was found that seven of them consumed oxygen easily, and compared with chemical oxygen consumption, biological oxygen consumption was in a leading position.

A framework for real-time operation of urban detention reservoirs: Application of the cellular automata and rainfall nowcasting.

Detention reservoirs are employed in urban drainage systems to reduce peak flows downstream of reservoirs. In addition to the volume of detention reservoirs, their operational policies could significantly affect their performance. This paper presents a framework for the real-time coordinated operation of detention reservoirs using deep-learning-based rainfall nowcasting data. Considering the short concentration time of urban basins, the real-time operating policies of urban detention reservoirs should be developed quickly. In the proposed framework, a cellular automata (CA)-based optimization algorithm is linked with the storm water management model (SWMM) to optimize real-time operating policies of gates at the inlets and outlets of detention reservoirs. As CA-based optimization models are not population-based, their computational costs are much less than population-based metaheuristic optimization techniques such as genetic algorithms. To evaluate the applicability and efficiency of the framework, it is applied to the east drainage catchment (EDC) of Tehran metropolitan area in Iran. The results illustrate that the proposed framework could reduce the overflow volume by up to 60%. For complete flood control in the study area, in addition to the real-time operation of detention reservoirs, constructing five tunnels with a total length of 13200 m is recommended. To evaluate the performance of the CA-based optimization model, its results are compared with those obtained from the non-dominated sorting genetic algorithm III (NSGA-III). It is shown that the CA-based model provides similar results with only 5% of the run-time of NSGA-III. A sensitivity analysis is also performed to evaluate the effects of optimization models' parameters on their performance.

A resilience-based robustness evaluation framework for sustainable urban flood management under uncertainty.

Urban drainage systems (UDSs) may experience failure encountering uncertain future conditions. These uncertainties arise from internal and external threats such as sedimentation, blockage, and climate change. In this paper, a new resilience-based framework is proposed to assess the robustness of urban flood management strategies under some distinct future scenarios. The robustness values of flood management strategies are evaluated by considering reliability, resiliency, and socio-ecological resilience criteria. The socio-ecologic resilience criteria are proposed considering the seven principles of building resilience proposed by Biggs et al. (2012). The evidential reasoning (ER) approach and the regret theory are utilized to calculate the total robustness of the flood management strategies. In this framework, the non-dominated sorting genetic algorithms III (NSGA-III) optimization model and the storm water management model (SWMM) simulation model are linked and run to quantify the criteria. The novelty of this paper lies in presenting a new framework to increase the sustainability and resilience of cities against floods considering the deep uncertainties in the main economic, social, and hydrological factors. This methodology provides policies for redesigning and sustainable operation of urban infrastructures to deal with floods. To evaluate the applicability and efficiency of the framework, it is applied to the East drainage catchment of the Tehran metropolitan area in Iran. The results show that real-time operation of existing flood detention reservoirs, along with implementing five new relief tunnels with a construction cost of 37.1 million dollars, is the most robust non-dominated strategy for flood management in the study area. Comparing the results of the proposed framework with those of a traditional framework shows that it can increase the robustness value by about 40% with the same implementation cost.

Machine learning-based surrogate model assisting stochastic model predictive control of urban drainage systems.

Quantifying the uncertainty of stormwater inflow is critical for improving the resilience of urban drainage systems (UDSs). However, the high computational complexity and time consumption obstruct the implementation of uncertainty-addressing methods for real-time control of UDSs. To address this issue, this study developed a machine learning-based surrogate model (MLSM) that maintains high-fidelity descriptions of drainage dynamics and meanwhile diminishes the computational complexity. With stormwater inflow and controls as inputs and system overflow as the output, MLSM is able to fast evaluate system performance, and therefore stochastic optimization becomes feasible. Thus, a real-time control strategy was built by combining MLSM with the stochastic model predictive control. This strategy used stochastic stormwater inflow scenarios as input and aimed to minimize the expected overflow under all scenarios. An ensemble of stormwater inflow scenarios was generated by assuming the forecast errors follow normal distributions. To downsize the ensemble, representative scenarios with their probabilities were selected using the simultaneous backward reduction method. The proposed control strategy was applied to a combined UDS of China. Results are as follows. (1) MLSM fit well with the original high-fidelity urban drainage model, while the computational time was reduced by 99.1%. (2) The proposed strategy consistently outperformed the classical deterministic model predictive control in both magnitude and duration dimensions of system resilience, when the consumed time compatible is with the real-time operation. It is indicated that the proposed control strategy could be used to inform the real-time operation of complex UDSs and thus enhance system resilience to uncertainty.