An automated toolchain for the data-driven and dynamical modeling of combined sewer systems

Troutman, Sara C.; Schambach, Nathaniel; Love, Nancy G.; Kerkez, Branko

doi:10.1016/j.watres.2017.08.065

Cited by 26 publications

(12 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While data-driven methods are experiencing a boom in the field of image processing, natural language processing, speech recognition, stock portfolio management, and other fields, so far only very few applications are known in the field of urban hydrology. Troutman et al (2017) combine Gaussian processes (dry-weather flows) and dynamical System Identification (wet weather discharge) aiming to simulate rainfall-run-off dynamics in a combined sewer network purely based on sensor data. Although the detection of novelty in monitored data had not been the primary objective, this approach could be applied to do so.…”

Section: Motivationmentioning

confidence: 99%

Metrology in Urban Drainage and Stormwater Management: Plug and Pray

2021

View full text Add to dashboard Cite

This book presents the advancements made in applied metrology in the field of Urban Drainage and Storm water Management over the past two decades in scientific research as well as in practical applications. Given the broadness of this subject (measuring principles, uncertainty in data, data validation, data storage and communication, design, maintenance and management of monitoring networks, technical details of sensor technology), the focus is on water quantity and a sound metrological basis. The book offers common ground for academics and practitioners when setting up monitoring projects in urban drainage and storm water management. This will enable an easier exchange of results so as to allow for a faster scientific progress in the field. A second, but equally important goal, is to allow practitioners access to scientific developments and gained experience when it comes to monitoring urban drainage and storm water systems. In-depth descriptions of international case studies covering all aspects discussed in the book are presented, along with self-training exercises and codes available for readers on a companion website. Numerous detailed examples are given in the book, with corresponding open-source codes and training files available to download here. ISBN: 9781789060102 (Paperback) ISBN: 9781789060119 (eBook)

show abstract

Section: Motivationmentioning

confidence: 99%

Metrology in Urban Drainage and Stormwater Management: Plug and Pray

2021

View full text Add to dashboard Cite

show abstract

“…However, linearized and conceptual internal models do not allow flow-dependent time delays, conceptualize the physically measurable levels and flows, furthermore restrict the flow deviation from steady-state solutions. Data-driven modeling has been reported in [Balla et al, 2020a] and in [Troutman et al, 2017], where grey-box and black-box identification have been used, respectively.…”

Section: Introductionmentioning

confidence: 99%

A Nonlinear Predictive Control Approach for Urban Drainage Networks Using Data-Driven Models and Moving Horizon Estimation

Balla

Schou

Bendtsen

et al. 2022

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

Real-time control of urban drainage networks is a complex task where transport flows are non-pressurized and therefore impose flow-dependent time delays in the system. Unfortunately, the installation of flow sensors is economically out of reach at most utilities, although knowing volumes and flows are essential to optimize system operation. In this article, we formulate joint parameter and state estimation based on level sensors deployed inside manholes and basins in the network. We describe the flow dynamics on the main pipelines by the level variations inside manholes, characterized by a system of coupled partial differential equations. These dynamics are approximated with kinematic waves where the network model is established with the water levels being the system states. Moving horizon estimation is developed where the states and parameters are obtained via the levels and estimated flow data, utilizing the topological layout of the network. The obtained model complexity is kept within practically achievable limits, suitable for nonlinear predictive control. The effectiveness of the control and estimation method is demonstrated on a high-fidelity model of a drainage network, acting as virtual reality. We use real rain and wastewater flow data and test the controller against the uncertainty in the disturbance forecasts.

show abstract

“…Physically-based models characterize the entire sewer system including catchment areas, dry weather flow patterns, transport within sewer systems, and other physical model components. Corresponding model parameters are often hard to determine and high uncertainties are inevitable [47].…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Combined Sewer System Operation Strategies Based on Highly Resolved Online Data

Bachmann-Machnik

Brüning

Bakhshipour

et al. 2021

Water

View full text Add to dashboard Cite

Operational and structural interventions in the field of stormwater management are usually planned based on long-term simulations using rainfall-runoff models. The simulation results are often highly uncertain due to imperfections of the model structure and inevitable uncertainties of input data. The trend towards monitoring of combined sewer overflows (CSO) structures produces more and more data which can be used to replace parts of the models and reduce uncertainty. In this study we use highly resolved online flow and quality monitoring data to optimize static outflow settings of CSO tanks. In a second step, the additional benefit of real time control (RTC) strategies is assessed. In both cases the aim is the reduction of CSO emissions. The methodology is developed on a conceptual drainage system with two CSO tanks and then applied to a case study area in Southern Germany with six tanks. A measured time series of six months is sufficient for reliable optimization results in the conceptual catchment as well as in the case study area system. In the investigated system the choice of the optimization objective (minimum overflow volume or total suspended solids (TSS) load) had no significant influence on the result. The presented method is particularly suitable for areas in which reliable monitoring data are available, but hydrological parameters of the catchment areas are uncertain. One strength of the proposed approach lies in the accurate representation of the distribution of emissions between the individual CSO structures over an entire system. This way emissions can be fitted to the sensitivity of the receiving water body at the specific outlets.

show abstract

An automated toolchain for the data-driven and dynamical modeling of combined sewer systems

Cited by 26 publications

References 35 publications

Metrology in Urban Drainage and Stormwater Management: Plug and Pray

Metrology in Urban Drainage and Stormwater Management: Plug and Pray

A Nonlinear Predictive Control Approach for Urban Drainage Networks Using Data-Driven Models and Moving Horizon Estimation

Evaluation of Combined Sewer System Operation Strategies Based on Highly Resolved Online Data

Contact Info

Product

Resources

About