Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning

Flooding in coastal cities is increasing due to climate change and sea-level rise, stressing the traditional stormwater systems these communities rely on. Automated real-time control (RTC) of these systems can improve performance, and creating control policies for smart stormwater systems is an acti...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Benjamin D. Bowes, Arash Tavakoli, Cheng Wang, Arsalan Heydarian, Madhur Behl, Peter A. Beling, Jonathan L. Goodall
Formato:	article
Lenguaje:	EN
Publicado:	IWA Publishing 2021
Materias:	real-time control reinforcement learning smart stormwater systems urban flooding Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066
Acceso en línea:	https://doaj.org/article/92bd11b5fa9c4d51ad6bec5092e0a374
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:92bd11b5fa9c4d51ad6bec5092e0a374
record_format	dspace
spelling	oai:doaj.org-article:92bd11b5fa9c4d51ad6bec5092e0a3742021-11-05T17:46:46ZFlood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning1464-71411465-173410.2166/hydro.2020.080https://doaj.org/article/92bd11b5fa9c4d51ad6bec5092e0a3742021-05-01T00:00:00Zhttp://jh.iwaponline.com/content/23/3/529https://doaj.org/toc/1464-7141https://doaj.org/toc/1465-1734Flooding in coastal cities is increasing due to climate change and sea-level rise, stressing the traditional stormwater systems these communities rely on. Automated real-time control (RTC) of these systems can improve performance, and creating control policies for smart stormwater systems is an active area of study. This research explores reinforcement learning (RL) to create control policies to mitigate flood risk. RL is trained using a model of hypothetical urban catchments with a tidal boundary and two retention ponds with controllable valves. RL's performance is compared to the passive system, a model predictive control (MPC) strategy, and a rule-based control strategy (RBC). RL learns to proactively manage pond levels using current and forecast conditions and reduced flooding by 32% over the passive system. Compared to the MPC approach using a physics-based model and genetic algorithm, RL achieved nearly the same flood reduction, just 3% less than MPC, with a significant 88× speedup in runtime. Compared to RBC, RL was able to quickly learn similar control strategies and reduced flooding by an additional 19%. This research demonstrates that RL can effectively control a simple system and offers a computationally efficient method that could scale to RTC of more complex stormwater systems. HIGHLIGHTS Reinforcement learning (RL) creates policies for real-time coastal stormwater system control.; RL's ability to mitigate flooding and manage ponds is compared to a passive system, model predictive control, and rule-based control.; RL was more efficient than model predictive control using a physics-based model and genetic algorithm.; RL's ability to mitigate flooding exceeded the passive system and rule-based control.;Benjamin D. BowesArash TavakoliCheng WangArsalan HeydarianMadhur BehlPeter A. BelingJonathan L. GoodallIWA Publishingarticlereal-time controlreinforcement learningsmart stormwater systemsurban floodingInformation technologyT58.5-58.64Environmental technology. Sanitary engineeringTD1-1066ENJournal of Hydroinformatics, Vol 23, Iss 3, Pp 529-547 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	real-time control reinforcement learning smart stormwater systems urban flooding Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066
spellingShingle	real-time control reinforcement learning smart stormwater systems urban flooding Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 Benjamin D. Bowes Arash Tavakoli Cheng Wang Arsalan Heydarian Madhur Behl Peter A. Beling Jonathan L. Goodall Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
description	Flooding in coastal cities is increasing due to climate change and sea-level rise, stressing the traditional stormwater systems these communities rely on. Automated real-time control (RTC) of these systems can improve performance, and creating control policies for smart stormwater systems is an active area of study. This research explores reinforcement learning (RL) to create control policies to mitigate flood risk. RL is trained using a model of hypothetical urban catchments with a tidal boundary and two retention ponds with controllable valves. RL's performance is compared to the passive system, a model predictive control (MPC) strategy, and a rule-based control strategy (RBC). RL learns to proactively manage pond levels using current and forecast conditions and reduced flooding by 32% over the passive system. Compared to the MPC approach using a physics-based model and genetic algorithm, RL achieved nearly the same flood reduction, just 3% less than MPC, with a significant 88× speedup in runtime. Compared to RBC, RL was able to quickly learn similar control strategies and reduced flooding by an additional 19%. This research demonstrates that RL can effectively control a simple system and offers a computationally efficient method that could scale to RTC of more complex stormwater systems. HIGHLIGHTS Reinforcement learning (RL) creates policies for real-time coastal stormwater system control.; RL's ability to mitigate flooding and manage ponds is compared to a passive system, model predictive control, and rule-based control.; RL was more efficient than model predictive control using a physics-based model and genetic algorithm.; RL's ability to mitigate flooding exceeded the passive system and rule-based control.;
format	article
author	Benjamin D. Bowes Arash Tavakoli Cheng Wang Arsalan Heydarian Madhur Behl Peter A. Beling Jonathan L. Goodall
author_facet	Benjamin D. Bowes Arash Tavakoli Cheng Wang Arsalan Heydarian Madhur Behl Peter A. Beling Jonathan L. Goodall
author_sort	Benjamin D. Bowes
title	Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
title_short	Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
title_full	Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
title_fullStr	Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
title_full_unstemmed	Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
title_sort	flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning
publisher	IWA Publishing
publishDate	2021
url	https://doaj.org/article/92bd11b5fa9c4d51ad6bec5092e0a374
work_keys_str_mv	AT benjamindbowes floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning AT arashtavakoli floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning AT chengwang floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning AT arsalanheydarian floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning AT madhurbehl floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning AT peterabeling floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning AT jonathanlgoodall floodmitigationincoastalurbancatchmentsusingrealtimestormwaterinfrastructurecontrolandreinforcementlearning
_version_	1718444113091624960

Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning

Ejemplares similares