Prescriptive analytics for low and high water management in German waterways
-
Marianne Brum
,
Dennis Meißner
Abstract
The real-time management of multi-purpose storage reservoirs aims at an efficient operation of existing hydraulic infrastructure. This management process can be structured as a prescriptive analytics setup that considers both current and predicted system states to recommend actions and outline potential implications. In application to a reservoir and river system, it combines hydrological modelling components for the system schematization, observations and data assimilation for the identification of the current system state, meteorological and hydrological predictions as well as optimization-based techniques to support decision-making regarding reservoir operations. In this paper, we present the application of such a framework to the short-term management of the Eder and Diemel storage reservoirs. These reservoirs are operated by the German Federal Waterways and Shipping Administration (WSV) with the primary goal to support navigation in the River Weser during low flow periods. In addition, partially conflicting objectives such as flood protection, energy generation and recreation are considered. The implementation includes an explicit consideration of forecast uncertainty and its impact on the decision-making by using probabilistic forecasts in combination with a multi-stage stochastic optimization approach. We demonstrate the applicability of the approach based on low and high water use cases. Special attention is paid on the benefits of the probabilistic forecast in combination with the multi-stage stochastic optimization versus a deterministic setup. It provides an explicit translation of the forecast uncertainty in the decision variables, in this case the reservoir releases helping the operators to better anticipate the range of future release decisions. Furthermore, the stochastic approach is expected to provide more stable decisions in an operational setting, based on more stable forecasts by considering various possible realizations of the future instead of picking a single one, which gets random after 4–5 days.
Zusammenfassung
Die Echtzeitbewirtschaftung von Mehrzweckspeichern zielt auf einen effizienten Betrieb vorhandener hydraulischer Infrastruktur ab. Dieser Prozess kann als präskriptives Optimierungsmodell strukturiert werden, das sowohl aktuelle als auch prognostizierte Systemzustände berücksichtigt, um Handlungsempfehlungen zu geben und potenzielle Auswirkungen zu skizzieren. In Anwendung auf ein Talsperren- und Flusssystem kombiniert es hydrologische Modellierungskomponenten für die System-Schematisierung, Beobachtungen und Datenassimilation zur Identifizierung des aktuellen Systemzustands, meteorologische und hydrologische Vorhersagen sowie optimierungsbasierte Techniken zur Unterstützung der Entscheidungsfindung bezüglich des Talsperrenbetriebs. In diesem Paper präsentieren wir die Anwendung eines solchen Rahmens auf das Kurzzeitmanagement der Eder- und Diemeltalsperre. Diese Speicher werden von der deutschen Wasserstraßen- und Schifffahrtsverwaltung des Bundes (WSV) betrieben mit dem Hauptziel, die Schifffahrt auf der Weser während Niedrigwasserperioden zu unterstützen. Darüber hinaus sind teilweise konkurrierende Ziele wie Hochwasserschutz, Energieerzeugung und Erholung zu berücksichtigen. Die Umsetzung beinhaltet eine explizite Berücksichtigung der Prognoseunsicherheit und ihrer Auswirkungen auf die Entscheidungsfindung durch die Verwendung probabilistischer Prognosen in Kombination mit einem mehrstufigen stochastischen Optimierungsansatz. Wir zeigen die Anwendbarkeit des Ansatzes anhand von Niedrig- und Hochwassersituationen auf. Besonderes Augenmerk wird auf die Vorteile der probabilistischen Prognose in Kombination mit der mehrstufigen stochastischen Optimierung im Vergleich zu einem deterministischen Setup gelegt. Es bietet eine explizite Übersetzung der Prognoseunsicherheit in die Entscheidungsvariablen, in diesem Fall die Talsperrenabgabe, um den Betreibern zu helfen, den Bereich zukünftiger Abgabeentscheidungen besser abzuschätzen. Darüber hinaus wird erwartet, dass der stochastische Ansatz in einem operationellen Umfeld robustere Entscheidungen bietet, basierend auf stabileren Vorhersagen durch die Berücksichtigung verschiedener möglicher Realisierungen der Zukunft anstelle der Auswahl einer einzigen, die nach 4-5 Tagen zufällig wird.
About the authors
Marianne Brum is a consultant with the Data and Forecast Services team at KISTERS in Germany. With a background in Civil Engineering and a MSc in Hydro-informatics, her main research interests lie in the intersection of water science and engineering, mathematics, and computer science. In particular, in the topics of data assimilation, uncertainty analysis, mathematical optimization and decision-support systems for reservoir management, and model integration.
Dennis Meißner is a scientific staff member at the Federal Institute of Hydrology located in Koblenz, Germany. With a background in civil and coastal hydraulic engineering his research interests are hydrological and hydrodynamic modelling as well as the development of operational forecasting systems and related services for short-term up to seasonal (probabilistic) predictions.
Dr. Bastian Klein is a scientific staff member at the Federal Institute of Hydrology located in Koblenz, Germany. With a background in hydrology his research interests are hydrological and statistical modelling as well as the development of operational forecasting systems and related services for short-term up to seasonal (probabilistic) predictions.
Jochen Hohenrainer is a scientific staff member at the Federal Institute of Hydrology located in Koblenz, Germany. With a background in hydrology his research interests are hydrological statistics and hydrological modelling as well as the development of models for water resources management.
Dr. Dirk Schwanenberg is the managing director of KISTERS HydroMet located in Aachen, Germany. With a background in civil engineering and water resources, his research interests cover the numerical modelling of water resources processes as well as the conceptional and technical integration of real-time forecasting and decision support systems.
-
Research ethics: Not applicable.
-
Author contributions: The authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: The authors state no conflict of interest.
-
Research funding: None declared.
-
Data availability: Data presented in the paper can be requested to the authors.
References
[1] R. Oppermann and K. Richter, Untersuchung zur Bewirtschaftung der Edertalsperre, Bundesanstalt für Gewässerkunde, Bericht BfG-1408 (in German), 2004.Search in Google Scholar
[2] V. M. Ponce, “Engineering hydrology, principles and practices,” 1989. Available at: http://ponce.sdsu.edu/textbookhydrologyp252.html Accessed: Jul. 30, 2022.Search in Google Scholar
[3] F. Akram, M. Rasul, M. M. Khan, and M. S. Amir, “Comparison of different hydrograph routing techniques in XPSTORM modelling software: a case study. World academy of science, engineering and technology, international journal of environmental, chemical, ecological,” Geol. Geophys. Eng., vol. 8, pp. 213–223, 2014.Search in Google Scholar
[4] Y. Liu, et al.., “Advancing data assimilation in operational hydrologic forecasting: progresses, challenges, and emerging opportunities,” Hydrol. Earth Syst. Sci., vol. 16, no. 10, pp. 3863–3887, 2012. https://doi.org/10.5194/hess-16-3863-2012.Search in Google Scholar
[5] R. H. Reichle, “Data assimilation methods in the Earth sciences,” Adv. Water Resour., vol. 31, no. 11, pp. 1411–1418, 2008. https://doi.org/10.1016/j.advwatres.2008.01.001.Search in Google Scholar
[6] J. Dupačová, N. Gröwe-Kuska, and W. Römisch, “Scenario reduction in stochastic programming,” Math. Program., vol. 95, no. 3, pp. 493–511, 2003.10.1007/s10107-002-0331-0Search in Google Scholar
[7] K. Sutiene, D. Makackas, and H. Pranevius, “Multistage k-means clustering for scenario tree construction,” Informatica, vol. 21, no. 1, pp. 123–138, 2010.10.15388/Informatica.2010.277Search in Google Scholar
[8] L. Raso, N. van de Giesen, P. Stive, D. Schwanenberg, and P.-J. van Overloop, “Tree structure generation from ensemble forecasts for real time control,” Hydrol. Process., vol. 27, no. 1, pp. 75–82, 2013.10.1002/hyp.9473Search in Google Scholar
[9] D. Schwanenberg, F. M. Fan, S. Naumann, J. I. Kuwajima, R. Alvarado Montero, and A. Assis dos Reis, “Short-term reservoir optimization for flood mitigation under meteorological and hydrological forecast uncertainty,” Water Resour. Manag., vol. 29, no. 5, pp. 1635–1651, 2015.10.1007/s11269-014-0899-1Search in Google Scholar
[10] R. G. Owens and T. D. Hewson, ECMWF Forecast User Guide, UK, ECMWF: Reading, 2018. Available at: https://www.ecmwf.int/en/elibrary/81307-ecmwf-forecast-user-guide Accessed: Nov. 27, 2023.Search in Google Scholar
[11] U. Pfeifroth, et al.., Surface Radiation Data Set – Heliosat (SARAH) – Edition 3, Satellite Application Facility on Climate Monitoring, 2023, https://doi.org/10.5676/EUM_SAF_CM/SARAH/V003.Search in Google Scholar
[12] K. Ludwig and M. Bremicker, “The water balance model LARSIM, design, content and applications,” in Freiburger Schriften zur Hydrologie, Institut für Hydrologie, C. Leibundgut, S. Demuth, and J. Lange, Eds., Freiburg, Universität Freiburg im Breisgau, 2006, p. 141.Search in Google Scholar
[13] C. Elpers and M. Hunger, Erstellung und Kalibrierung Wasserhaushaltsmodell LARSIM-Weser, Schlussbericht, Federal Institute of Hydrologie, 2022. Available at: vorhersage@bafg.de.Search in Google Scholar
© 2024 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Editorial
- Advanced water management
- Survey
- Advances in dissolved oxygen prediction and control methods in aquaculture: a review
- Methods
- Probabilistic multi-step ahead streamflow forecast based on deep learning
- A new assessment method on co-occurring mountain and plain floods based on copula functions
- Application
- Prescriptive analytics for low and high water management in German waterways
- Tools
- Integration of selected AI methods into a simulation tool for urban wastewater systems – towards practical application
- NiMo 4.0 – Enabling advanced data analytics with AI for environmental governance in the water domain
Articles in the same Issue
- Frontmatter
- Editorial
- Advanced water management
- Survey
- Advances in dissolved oxygen prediction and control methods in aquaculture: a review
- Methods
- Probabilistic multi-step ahead streamflow forecast based on deep learning
- A new assessment method on co-occurring mountain and plain floods based on copula functions
- Application
- Prescriptive analytics for low and high water management in German waterways
- Tools
- Integration of selected AI methods into a simulation tool for urban wastewater systems – towards practical application
- NiMo 4.0 – Enabling advanced data analytics with AI for environmental governance in the water domain
