Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants
The treatments to be applied for water purification must be dynamically adaptable to the raw water conditions. Currently, treatments are applied based on standards that are not optimized for the circumstances of each drinking water treatment plant (DWTP), neither for critical events. This paper pres...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IWA Publishing
2024-01-01
|
| Series: | Water Supply |
| Subjects: | |
| Online Access: | http://ws.iwaponline.com/content/24/1/86 |
| _version_ | 1797200636360523776 |
|---|---|
| author | Aída Álvarez Díez Rocío Pena Rois Iulian Mocanu Claudia Orzan Cristian Brebenel Jiru Stere Santiago Muíños Landín Juan Manuel Fernández Montenegro |
| author_facet | Aída Álvarez Díez Rocío Pena Rois Iulian Mocanu Claudia Orzan Cristian Brebenel Jiru Stere Santiago Muíños Landín Juan Manuel Fernández Montenegro |
| author_sort | Aída Álvarez Díez |
| collection | DOAJ |
| description | The treatments to be applied for water purification must be dynamically adaptable to the raw water conditions. Currently, treatments are applied based on standards that are not optimized for the circumstances of each drinking water treatment plant (DWTP), neither for critical events. This paper presents a methodology for the creation of an Artificial Intelligence (AI) decision support system (DSS), encompassing the principal steps of the drinking water treatment processes (coagulation, sedimentation, filtration and disinfection), based on reinforcement learning techniques, that provides suggestions about the most efficient treatments (coagulant and chlorine dosages) for various raw water conditions, including critical events such as heavy rain and saline intrusions. Together with the model, a retraining strategy is included so the DSS adapts itself to the specific circumstances of each different DWTP. The model has been developed and validated in a DWTP replica. Furthermore, the model has been provided to a real DWTP to obtain feedback from experienced staff. The results and evaluation of the model are promising as a first approach on a DSS for drinking water treatments suggestion, although future versions might require more water quality parameters to characterize the raw water.
HIGHLIGHTS
AI decision support system (DSS) for the suggestion of the most efficient dosages (coagulant and chlorine) to be used in drinking water treatment plants (DWTP).;
Multi-armed bandits applied on the whole process of most common DWTPs (coagulation/filtration/disinfection).;
Combination of simulated data, data from a scaled-down replica DWTP, and a real DWTP.;
Auto-optimization routine.;
Outcomes validated by a real DWTP.; |
| first_indexed | 2024-03-08T00:26:30Z |
| format | Article |
| id | doaj.art-1ab8292401bc4110a3d1c18fd348ba16 |
| institution | Directory Open Access Journal |
| issn | 1606-9749 1607-0798 |
| language | English |
| last_indexed | 2024-04-24T07:34:48Z |
| publishDate | 2024-01-01 |
| publisher | IWA Publishing |
| record_format | Article |
| series | Water Supply |
| spelling | doaj.art-1ab8292401bc4110a3d1c18fd348ba162024-04-20T06:42:27ZengIWA PublishingWater Supply1606-97491607-07982024-01-012418610210.2166/ws.2023.328328Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plantsAída Álvarez Díez0Rocío Pena Rois1Iulian Mocanu2Claudia Orzan3Cristian Brebenel4Jiru Stere5Santiago Muíños Landín6Juan Manuel Fernández Montenegro7 AIMEN Technology Centre, Porriño 36410, Spain AIMEN Technology Centre, Porriño 36410, Spain CUP Dunarea, Brăila 555555, Romania CUP Dunarea, Brăila 555555, Romania CUP Dunarea, Brăila 555555, Romania CUP Dunarea, Brăila 555555, Romania AIMEN Technology Centre, Porriño 36410, Spain AIMEN Technology Centre, Porriño 36410, Spain The treatments to be applied for water purification must be dynamically adaptable to the raw water conditions. Currently, treatments are applied based on standards that are not optimized for the circumstances of each drinking water treatment plant (DWTP), neither for critical events. This paper presents a methodology for the creation of an Artificial Intelligence (AI) decision support system (DSS), encompassing the principal steps of the drinking water treatment processes (coagulation, sedimentation, filtration and disinfection), based on reinforcement learning techniques, that provides suggestions about the most efficient treatments (coagulant and chlorine dosages) for various raw water conditions, including critical events such as heavy rain and saline intrusions. Together with the model, a retraining strategy is included so the DSS adapts itself to the specific circumstances of each different DWTP. The model has been developed and validated in a DWTP replica. Furthermore, the model has been provided to a real DWTP to obtain feedback from experienced staff. The results and evaluation of the model are promising as a first approach on a DSS for drinking water treatments suggestion, although future versions might require more water quality parameters to characterize the raw water. HIGHLIGHTS AI decision support system (DSS) for the suggestion of the most efficient dosages (coagulant and chlorine) to be used in drinking water treatment plants (DWTP).; Multi-armed bandits applied on the whole process of most common DWTPs (coagulation/filtration/disinfection).; Combination of simulated data, data from a scaled-down replica DWTP, and a real DWTP.; Auto-optimization routine.; Outcomes validated by a real DWTP.;http://ws.iwaponline.com/content/24/1/86aidecision supportdrinking waterreinforcement learningtreatmentswater quality |
| spellingShingle | Aída Álvarez Díez Rocío Pena Rois Iulian Mocanu Claudia Orzan Cristian Brebenel Jiru Stere Santiago Muíños Landín Juan Manuel Fernández Montenegro Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants Water Supply ai decision support drinking water reinforcement learning treatments water quality |
| title | Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants |
| title_full | Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants |
| title_fullStr | Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants |
| title_full_unstemmed | Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants |
| title_short | Reinforcement learning-based DSS for coagulant and disinfectant dosage selection on drinking water treatment plants |
| title_sort | reinforcement learning based dss for coagulant and disinfectant dosage selection on drinking water treatment plants |
| topic | ai decision support drinking water reinforcement learning treatments water quality |
| url | http://ws.iwaponline.com/content/24/1/86 |
| work_keys_str_mv | AT aidaalvarezdiez reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT rociopenarois reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT iulianmocanu reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT claudiaorzan reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT cristianbrebenel reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT jirustere reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT santiagomuinoslandin reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants AT juanmanuelfernandezmontenegro reinforcementlearningbaseddssforcoagulantanddisinfectantdosageselectionondrinkingwatertreatmentplants |