Grant: Engage Plus Grants Program
University of Alberta
$12,500.00
Control of a ultrafiltration-based water-treatment plant using reinforcement learning: testing on a bench-scale system
EF 2016-17
Résumé
Healthy, secure communities require clean drinking water. Water treatment is the process in which variousfilters and chemical treatments are applied to keep the water clean. However, water treatment is expensive asvarious parts of the process require much energy. The high level goal of the project will be to optimize theenergy efficiency of the treatment process to achieve higher sustainability of operating water treatment plants.The industrial partner, ISL Engineering and Land Services Ltd, is an emerging leader in water and wastewatermanagement for small-to-medium size municipalities. As every water treatment plant is uniquely configured,for cost efficiency of long term operations, the project will focus on developing and testing highly adaptive androbust reinforcement learning methods (a subclass of machine learning methods) that adjust automatically tonew settings. In particular, the project will focus on developing such strategies for maximizing energyefficiency while maintaining operational constraints, such as the production rate and various quality measures.The expected outcome is that the reinforcement learning platform will provide a robust methodology capableof exploring and exploiting process controls, improving the energy efficiency of the water treatment process.This research now requires a bench-scale mini-treatment plant to meet our goals of exploring improved energyuse linked to membrane processes, chemical treatment efficiencies and expanding the range of sensors neededto inform our RL system. Currently we are developing a testable RL control strategy for the UF membranesystem and now we have access to a bench-scale membrane water treatment system in the mechanicalengineering department at the University of Alberta. In this phase, we will fit and validate the RL controller onthis bench-scale UF membrane system.
Programme
Engage Plus Grants Program
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
505305-2016
Name nom
Szepesvari, Csaba
Department département
Computing Science
Organization id
9
Institution établissement
University of Alberta
Province en
Alberta
Province fr
Alberta
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2016
Competition year année de concours
2016
Award amount
12500
Program id
EGP2
Program name en
Engage Plus Grants Program
Program name fr
Programme de subvention d'engagement partenarial Plus
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1555
Committee name en
Prairie Internal Decision Committee
Committee name fr
Comité de décision interne pour les Prairies
Area of application code
702
Area of application group en
Transportation systems and services
Area of application group fr
Services et systèmes de transport
Area of application en
Water
Area of application fr
Maritime
Research subject code
2800
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Artificial intelligence (computer vision, use2603)
Research subject fr
Intelligence artificielle (vision artificielle, utiliser 2603)
Partie
1
Nb partie
1
Application title
Control of a ultrafiltration-based water-treatment plant using reinforcement learning: testing on a bench-scale system
Keyword
NSERC
Application summary
Healthy, secure communities require clean drinking water. Water treatment is the process in which variousfilters and chemical treatments are applied to keep the water clean. However, water treatment is expensive asvarious parts of the process require much energy. The high level goal of the project will be to optimize theenergy efficiency of the treatment process to achieve higher sustainability of operating water treatment plants.The industrial partner, ISL Engineering and Land Services Ltd, is an emerging leader in water and wastewatermanagement for small-to-medium size municipalities. As every water treatment plant is uniquely configured,for cost efficiency of long term operations, the project will focus on developing and testing highly adaptive androbust reinforcement learning methods (a subclass of machine learning methods) that adjust automatically tonew settings. In particular, the project will focus on developing such strategies for maximizing energyefficiency while maintaining operational constraints, such as the production rate and various quality measures.The expected outcome is that the reinforcement learning platform will provide a robust methodology capableof exploring and exploiting process controls, improving the energy efficiency of the water treatment process.This research now requires a bench-scale mini-treatment plant to meet our goals of exploring improved energyuse linked to membrane processes, chemical treatment efficiencies and expanding the range of sensors neededto inform our RL system. Currently we are developing a testable RL control strategy for the UF membranesystem and now we have access to a bench-scale membrane water treatment system in the mechanicalengineering department at the University of Alberta. In this phase, we will fit and validate the RL controller onthis bench-scale UF membrane system.