Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach
Abstract This study presents a multi-agent reinforcement learning (MARL) framework for integrated real-time control (RTC) of urban drainage systems (UDSs), coordinating sewers, wastewater treatment plants (WWTPs), and receiving waters. Trained within a hydraulic–water quality simulation environment...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | npj Clean Water |
| Online Access: | https://doi.org/10.1038/s41545-025-00512-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849226622736007168 |
|---|---|
| author | Zhenyu Huang Yiming Wang Xin Dong Wei Li Yangbo Tang Dazhen Zhang |
| author_facet | Zhenyu Huang Yiming Wang Xin Dong Wei Li Yangbo Tang Dazhen Zhang |
| author_sort | Zhenyu Huang |
| collection | DOAJ |
| description | Abstract This study presents a multi-agent reinforcement learning (MARL) framework for integrated real-time control (RTC) of urban drainage systems (UDSs), coordinating sewers, wastewater treatment plants (WWTPs), and receiving waters. Trained within a hydraulic–water quality simulation environment using QMIX, the framework enables facility-level decision-making and adaptive system coordination. Applied to Lu’an City, China, MARL achieved a 25.4% reduction in flooding and overflow volumes and an 18.0% decrease in river pollutants relative to benchmark strategies, while maintaining real-time control feasibility (6.35 s per 5-min interval). Under rainfall forecast and sensor noise uncertainty, MARL improved performance stability by 44.7–52.4%. Despite operational trade-offs, the framework supports integrated system optimization and consistent water quality improvements in urban settings. |
| format | Article |
| id | doaj-art-0d3ba1dc7ce34c048d41a2804e8960b3 |
| institution | Kabale University |
| issn | 2059-7037 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Clean Water |
| spelling | doaj-art-0d3ba1dc7ce34c048d41a2804e8960b32025-08-24T11:05:44ZengNature Portfolionpj Clean Water2059-70372025-08-018111710.1038/s41545-025-00512-zPollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approachZhenyu Huang0Yiming Wang1Xin Dong2Wei Li3Yangbo Tang4Dazhen Zhang5School of Environment, Tsinghua UniversitySchool of Environment, Tsinghua UniversitySchool of Environment, Tsinghua UniversityNational Engineering Research Center of Eco-Environment in the Yangtze River Economic BeltNational Engineering Research Center of Eco-Environment in the Yangtze River Economic BeltNational Engineering Research Center of Eco-Environment in the Yangtze River Economic BeltAbstract This study presents a multi-agent reinforcement learning (MARL) framework for integrated real-time control (RTC) of urban drainage systems (UDSs), coordinating sewers, wastewater treatment plants (WWTPs), and receiving waters. Trained within a hydraulic–water quality simulation environment using QMIX, the framework enables facility-level decision-making and adaptive system coordination. Applied to Lu’an City, China, MARL achieved a 25.4% reduction in flooding and overflow volumes and an 18.0% decrease in river pollutants relative to benchmark strategies, while maintaining real-time control feasibility (6.35 s per 5-min interval). Under rainfall forecast and sensor noise uncertainty, MARL improved performance stability by 44.7–52.4%. Despite operational trade-offs, the framework supports integrated system optimization and consistent water quality improvements in urban settings.https://doi.org/10.1038/s41545-025-00512-z |
| spellingShingle | Zhenyu Huang Yiming Wang Xin Dong Wei Li Yangbo Tang Dazhen Zhang Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach npj Clean Water |
| title | Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach |
| title_full | Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach |
| title_fullStr | Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach |
| title_full_unstemmed | Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach |
| title_short | Pollution-based integrated real-time control for urban drainage systems: a multi-agent deep reinforcement learning approach |
| title_sort | pollution based integrated real time control for urban drainage systems a multi agent deep reinforcement learning approach |
| url | https://doi.org/10.1038/s41545-025-00512-z |
| work_keys_str_mv | AT zhenyuhuang pollutionbasedintegratedrealtimecontrolforurbandrainagesystemsamultiagentdeepreinforcementlearningapproach AT yimingwang pollutionbasedintegratedrealtimecontrolforurbandrainagesystemsamultiagentdeepreinforcementlearningapproach AT xindong pollutionbasedintegratedrealtimecontrolforurbandrainagesystemsamultiagentdeepreinforcementlearningapproach AT weili pollutionbasedintegratedrealtimecontrolforurbandrainagesystemsamultiagentdeepreinforcementlearningapproach AT yangbotang pollutionbasedintegratedrealtimecontrolforurbandrainagesystemsamultiagentdeepreinforcementlearningapproach AT dazhenzhang pollutionbasedintegratedrealtimecontrolforurbandrainagesystemsamultiagentdeepreinforcementlearningapproach |