scholarly journals Modeling and interpreting hydrological responses of sustainable urban drainage systems with explainable machine learning methods

2021 ◽  
Vol 25 (11) ◽  
pp. 5839-5858
Author(s):  
Yang Yang ◽  
Ting Fong May Chui
Keyword(s):  
Machine Learning ◽  
Management Practices ◽  
Hydrological Processes ◽  
Urban Drainage ◽  
Observation Data ◽  
Physical Processes ◽  
Large Uncertainty ◽  
Drainage Systems ◽  
Hydrological Responses ◽  
Urban Drainage Systems

Abstract. Sustainable urban drainage systems (SuDS) are decentralized stormwater management practices that mimic natural drainage processes. The hydrological processes of SuDS are often modeled using process-based models. However, it can require considerable effort to set up these models. This study thus proposes a machine learning (ML) method to directly learn the statistical correlations between the hydrological responses of SuDS and the forcing variables at sub-hourly timescales from observation data. The proposed methods are applied to two SuDS catchments with different sizes, SuDS practice types, and data availabilities in the USA for discharge prediction. The resulting models have high prediction accuracies (Nash–Sutcliffe efficiency, NSE, >0.70). ML explanation methods are then employed to derive the basis of each ML prediction, based on which the hydrological processes being modeled are then inferred. The physical realism of the inferred hydrological processes is then compared to that would be expected based on the domain-specific knowledge of the system being modeled. The inferred processes of some models, however, are found to be physically implausible. For instance, negative contributions of rainfall to runoff have been identified in some models. This study further empirically shows that an ML model's ability to provide accurate predictions can be uncorrelated with its ability to offer plausible explanations to the physical processes being modeled. Finally, this study provides a high-level overview of the practices of inferring physical processes from the ML modeling results and shows both conceptually and empirically that large uncertainty exists in every step of the inference processes. In summary, this study shows that ML methods are a useful tool for predicting the hydrological responses of SuDS catchments, and the hydrological processes inferred from modeling results should be interpreted cautiously due to the existence of large uncertainty in the inference processes.

scholarly journals Modeling and interpreting hydrological responses of sustainable urban drainage systems with explainable machine learning methods

2020 ◽  
Author(s):  
Yang Yang ◽  
Ting Fong May Chui
Keyword(s):  
Machine Learning ◽  
Management Practices ◽  
Hydrological Processes ◽  
Rainfall Time Series ◽  
Hydrograph Separation ◽  
Learning Methods ◽  
Drainage Systems ◽  
Factors Affecting ◽  
Machine Learning Methods ◽  
Hydrological Responses

Abstract. Sustainable drainage systems (SuDS) are decentralized stormwater management practices that mimic the natural drainage processes. Their modeling is often challenged by insufficient data and unknown factors affecting the hydrological processes. This study uses machine learning methods to model directly the correlation between hydrological responses and rainfalls at fine temporal scales in two catchments of different sizes. A feature engineering method is developed to extract useful information from rainfall time series and is used in combination with a nested cross-validation procedure to derive high-quality models and to estimate their generalization errors. The SHAP method is adopted to explain the basis of each prediction, which is then used for estimating catchment response time and hydrograph separation. The explanations of the predictions provide valuable insights into the models’ behavior and the involved hydrological processes. Thus, interpreting machine learning models is found as a useful way to study catchment hydrology.

scholarly journals Machine Learning and Urban Drainage Systems: State-of-the-Art Review

Water ◽  
2021 ◽  
Vol 13 (24) ◽  
pp. 3545
Author(s):  
Soon-Ho Kwon ◽  
Joong-Hoon Kim
Keyword(s):  
Machine Learning ◽  
State Of The Art ◽  
Urban Drainage ◽  
Complex Data ◽  
New Paradigm ◽  
Neural Network Models ◽  
Drainage Systems ◽  
Time Operation ◽  
Urban Drainage Systems ◽  
Scientific Engineering

In the last decade, machine learning (ML) technology has been transforming daily lives, industries, and various scientific/engineering disciplines. In particular, ML technology has resulted in significant progress in neural network models; these enable the automatic computation of problem-relevant features and rapid capture of highly complex data distributions. We believe that ML approaches can address several significant new and/or old challenges in urban drainage systems (UDSs). This review paper provides a state-of-the-art review of ML-based UDS modeling/application based on three categories: (1) operation (real-time operation control), (2) management (flood-inundation prediction) and (3) maintenance (pipe defect detection). The review reveals that ML is utilized extensively in UDSs to advance model performance and efficiency, extract complex data distribution patterns, and obtain scientific/engineering insights. Additionally, some potential issues and future directions are recommended for three research topics defined in this study to extend UDS modeling/applications based on ML technology. Furthermore, it is suggested that ML technology can promote developments in UDSs. The new paradigm of ML-based UDS modeling/applications summarized here is in its early stages and should be considered in future studies.

scholarly journals Root causes of failures in sustainable urban drainage systems (SUDS): an exploratory study in 11 municipalities in the Netherlands

2021 ◽  
Author(s):  
Vita Vollaers ◽  
Eva Nieuwenhuis ◽  
Frans van de Ven ◽  
Jeroen Langeveld
Keyword(s):  
Management Practices ◽  
Water System ◽  
Maintenance Phase ◽  
Pilot Scale ◽  
Urban Systems ◽  
Urban Drainage ◽  
Urban Water ◽  
Structured Interviews ◽  
Drainage Systems ◽  
Urban Drainage Systems

Abstract Despite being widely implemented, sustainable urban drainage systems (SUDS) do not always function flawlessly. While SUDS have been tested extensively and seem to perform well on a laboratory or pilot scale, practitioners' experience is different: failures in SUDS occur regularly in practice, resulting in malfunctioning systems, water nuisance and high costs. To anticipate their malfunctioning, and thus to improve their performance, a better understanding of failures occurring in SUDS and their underlying causes is needed. Based on an explorative case-study approach, consisting of site visits and semi-structured interviews with urban water professionals, this study presents an inventory of technical failures in SUDS and an analysis of their root causes. In total, 70 cases in 11 Dutch municipalities have been documented. The results show that the interfaces between SUDS and other urban systems are prominent failure locations. In addition, we found that failures originate from the entire development process of SUDS, i.e., from the design, construction and user/maintenance phase. With respect to the causes underlying these failures, our results show that these are mainly socio-institutional in nature. These are valuable insights for both practitioners and scholars, contributing to a renewed socio-technical urban water system with more sustainable water management practices.

Review of operation of urban drainage systems in cold weather: water quality considerations

2003 ◽  
Vol 48 (9) ◽  
pp. 11-20 ◽  
Author(s):  
J. Marsalek ◽  
G. Oberts ◽  
K. Exall ◽  
M. Viklander
Keyword(s):  
Best Management Practices ◽  
Management Practices ◽  
Melting Process ◽  
Cold Climate ◽  
Cold Weather ◽  
Urban Drainage ◽  
Drainage Systems ◽  
Urban Drainage Systems ◽  
Snow Water ◽  
Snow And Ice

Cold climate imposes special requirements on urban drainage systems, arising from extended storage of precipitation and pollutants in the catchment snowpack, processes occurring in the snowpack, and changes in catchment surface and transport network by snow and ice. Consequently, the resulting catchment response and runoff quantity differ from those experienced in snow- and ice-free seasons. Sources of pollutants entering urban snowpacks include airborne fallout, pavement and roadside deposits, and applications of de-icing and anti-skid agents. In the snowpack, snow, water and chemicals are subject to various processes, which affect their movement through the pack and eventual release during the melting process. Soluble constituents are flushed from the snowpack early during the melt; hydrophobic substances generally stay in the pack until the very end of melt and coarse solids with adsorbed pollutants stay on the ground after the melt is finished. The impacts of snowmelt on receiving waters have been measured mostly by the snowmelt chemical composition and inferences about its environmental significance. Recently, snowmelt has been tested by standard bioassays and often found toxic. Toxicity was attributed mostly to chloride and trace metals, and contributed to reduced diversity of benthic and plant communities. Thus, snowmelt and winter runoff discharged from urban drainage threaten aquatic ecosystems in many locations and require further studies with respect to advancing their understanding and development of best management practices.

Stormwater infiltration trenches: a conceptual modelling approach

2009 ◽  
Vol 60 (1) ◽  
pp. 185-199 ◽  
Author(s):  
Gabriele Freni ◽  
Giorgio Mannina ◽  
Gaspare Viviani
Keyword(s):  
Best Management Practices ◽  
Management Practices ◽  
Urban Drainage ◽  
Conceptual Modelling ◽  
Drainage Systems ◽  
Performance Reduction ◽  
Urban Drainage Systems ◽  
The Impact ◽  
Modelling Approach ◽  
Modelling Approaches

In recent years, limitations linked to traditional urban drainage schemes have been pointed out and new approaches are developing introducing more natural methods for retaining and/or disposing of stormwater. These mitigation measures are generally called Best Management Practices or Sustainable Urban Drainage System and they include practices such as infiltration and storage tanks in order to reduce the peak flow and retain part of the polluting components. The introduction of such practices in urban drainage systems entails an upgrade of existing modelling frameworks in order to evaluate their efficiency in mitigating the impact of urban drainage systems on receiving water bodies. While storage tank modelling approaches are quite well documented in literature, some gaps are still present about infiltration facilities mainly dependent on the complexity of the involved physical processes. In this study, a simplified conceptual modelling approach for the simulation of the infiltration trenches is presented. The model enables to assess the performance of infiltration trenches. The main goal is to develop a model that can be employed for the assessment of the mitigation efficiency of infiltration trenches in an integrated urban drainage context. Particular care was given to the simulation of infiltration structures considering the performance reduction due to clogging phenomena. The proposed model has been compared with other simplified modelling approaches and with a physically based model adopted as benchmark. The model performed better compared to other approaches considering both unclogged facilities and the effect of clogging. On the basis of a long-term simulation of six years of rain data, the performance and the effectiveness of an infiltration trench measure are assessed. The study confirmed the important role played by the clogging phenomenon on such infiltration structures.

Sign in / Sign up

Export Citation Format

Share Document