Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network

Lars Eric Ertlmeier; Zhenyu Yang; Benjamin Refsgaard

doi:10.1007/978-981-99-5716-3_24

Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network

Lars Eric Ertlmeier, Zhenyu Yang^*, Benjamin Refsgaard

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

Abstract

To make better prediction and control of river water navigation in a Danish city - Vejle, the Long-Short-Term-Memory (LSTM) neural-network model is adopted to predict the water-level nearby a high flooding-risk area using correlated historical data. A set of feedback control solutions are developed based on the extension of the obtained LSTM model to automatically regulate a distribution-gate system, which guides the coming stream-flow into separated urban rivers. The proposed control solutions are tested in simulation based on four historic events, and it can be observed that two floods at the critical areas since 2017 could have been prevented by balancing flow-splits using automatic feedback control, which was manually controlled in the past. This study demonstrates a clear and promising potential to use machine learning methods for supporting development of smart cities and their climate adaption strategies.

Original language	English
Title of host publication	Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023
Editors	Chih-Huang Weng
Number of pages	18
Publisher	Springer Science+Business Media
Publication date	2024
Pages	278-295
ISBN (Print)	9789819957156
DOIs	https://doi.org/10.1007/978-981-99-5716-3_24
Publication status	Published - 2024
Event	5th International Conference on Advances in Civil and Ecological Engineering Research, ACEER 2023 - Macau, China Duration: 4 Jul 2023 → 7 Jul 2023

Conference

Conference	5th International Conference on Advances in Civil and Ecological Engineering Research, ACEER 2023
Country/Territory	China
City	Macau
Period	04/07/2023 → 07/07/2023

Series	Lecture Notes in Civil Engineering
Volume	336 LNCE
ISSN	2366-2557

Bibliographical note

Publisher Copyright:
© 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Keywords

flood control
flood prediction
LSTM
Urban rivers

Access to Document

10.1007/978-981-99-5716-3_24

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Ertlmeier, L. E., Yang, Z., & Refsgaard, B. (2024). Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network. In C.-H. Weng (Ed.), Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023 (pp. 278-295). Springer Science+Business Media. https://doi.org/10.1007/978-981-99-5716-3_24

Ertlmeier, Lars Eric ; Yang, Zhenyu ; Refsgaard, Benjamin. / Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network. Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023. editor / Chih-Huang Weng. Springer Science+Business Media, 2024. pp. 278-295 (Lecture Notes in Civil Engineering, Vol. 336 LNCE).

@inproceedings{3c473d9adf9e4b439a9561ce0d55c26e,

title = "Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network",

abstract = "To make better prediction and control of river water navigation in a Danish city - Vejle, the Long-Short-Term-Memory (LSTM) neural-network model is adopted to predict the water-level nearby a high flooding-risk area using correlated historical data. A set of feedback control solutions are developed based on the extension of the obtained LSTM model to automatically regulate a distribution-gate system, which guides the coming stream-flow into separated urban rivers. The proposed control solutions are tested in simulation based on four historic events, and it can be observed that two floods at the critical areas since 2017 could have been prevented by balancing flow-splits using automatic feedback control, which was manually controlled in the past. This study demonstrates a clear and promising potential to use machine learning methods for supporting development of smart cities and their climate adaption strategies.",

keywords = "flood control, flood prediction, LSTM, Urban rivers",

author = "Ertlmeier, {Lars Eric} and Zhenyu Yang and Benjamin Refsgaard",

note = "Publisher Copyright: {\textcopyright} 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 5th International Conference on Advances in Civil and Ecological Engineering Research, ACEER 2023 ; Conference date: 04-07-2023 Through 07-07-2023",

year = "2024",

doi = "10.1007/978-981-99-5716-3_24",

language = "English",

isbn = "9789819957156",

series = "Lecture Notes in Civil Engineering",

publisher = "Springer Science+Business Media",

pages = "278--295",

editor = "Chih-Huang Weng",

booktitle = "Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023",

address = "United States",

}

Ertlmeier, LE, Yang, Z & Refsgaard, B 2024, Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network. in C-H Weng (ed.), Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023. Springer Science+Business Media, Lecture Notes in Civil Engineering, vol. 336 LNCE, pp. 278-295, 5th International Conference on Advances in Civil and Ecological Engineering Research, ACEER 2023, Macau, China, 04/07/2023. https://doi.org/10.1007/978-981-99-5716-3_24

Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network. / Ertlmeier, Lars Eric; Yang, Zhenyu; Refsgaard, Benjamin.
Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023. ed. / Chih-Huang Weng. Springer Science+Business Media, 2024. p. 278-295 (Lecture Notes in Civil Engineering, Vol. 336 LNCE).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network

AU - Ertlmeier, Lars Eric

AU - Yang, Zhenyu

AU - Refsgaard, Benjamin

PY - 2024

Y1 - 2024

N2 - To make better prediction and control of river water navigation in a Danish city - Vejle, the Long-Short-Term-Memory (LSTM) neural-network model is adopted to predict the water-level nearby a high flooding-risk area using correlated historical data. A set of feedback control solutions are developed based on the extension of the obtained LSTM model to automatically regulate a distribution-gate system, which guides the coming stream-flow into separated urban rivers. The proposed control solutions are tested in simulation based on four historic events, and it can be observed that two floods at the critical areas since 2017 could have been prevented by balancing flow-splits using automatic feedback control, which was manually controlled in the past. This study demonstrates a clear and promising potential to use machine learning methods for supporting development of smart cities and their climate adaption strategies.

AB - To make better prediction and control of river water navigation in a Danish city - Vejle, the Long-Short-Term-Memory (LSTM) neural-network model is adopted to predict the water-level nearby a high flooding-risk area using correlated historical data. A set of feedback control solutions are developed based on the extension of the obtained LSTM model to automatically regulate a distribution-gate system, which guides the coming stream-flow into separated urban rivers. The proposed control solutions are tested in simulation based on four historic events, and it can be observed that two floods at the critical areas since 2017 could have been prevented by balancing flow-splits using automatic feedback control, which was manually controlled in the past. This study demonstrates a clear and promising potential to use machine learning methods for supporting development of smart cities and their climate adaption strategies.

KW - flood control

KW - flood prediction

KW - LSTM

KW - Urban rivers

UR - http://www.scopus.com/inward/record.url?scp=85171524006&partnerID=8YFLogxK

U2 - 10.1007/978-981-99-5716-3_24

DO - 10.1007/978-981-99-5716-3_24

M3 - Article in proceeding

AN - SCOPUS:85171524006

SN - 9789819957156

T3 - Lecture Notes in Civil Engineering

SP - 278

EP - 295

BT - Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023

A2 - Weng, Chih-Huang

PB - Springer Science+Business Media

T2 - 5th International Conference on Advances in Civil and Ecological Engineering Research, ACEER 2023

Y2 - 4 July 2023 through 7 July 2023

ER -

Ertlmeier LE, Yang Z, Refsgaard B. Flood Forecast and Control for Urban Rivers Using LSTM Neural-Network. In Weng CH, editor, Proceedings of The 5th International Conference on Advances in Civil and Ecological Engineering Research - Proceedings of ACEER2023. Springer Science+Business Media. 2024. p. 278-295. (Lecture Notes in Civil Engineering, Vol. 336 LNCE). doi: 10.1007/978-981-99-5716-3_24