Abstract:Humankind has been grappling with wastewater treatment for centuries. To ensure optimal operation and control of a wastewater treatment plant (WWTP), suitable advanced control strategies are required as they are inherently nonlinear in nature and subjected to different influent conditions. This paper proposes a novel supervisory control scheme for sequencing batch reactor (SBR)-based WWTP. It integrates hierarchical fuzzy control, based on ammonia and nitrate observations, in the presence of lower-level proportional integral (PI) and fractional-order PI (FPI) controllers, with the dual goal of aeration cost reduction and effluent quality enhancement. A modified ASM2d (activated sludge model No. 2d) framework is used as a model for SBR. In the hierarchical control system, variable dissolved oxygen (DO) trajectories are generated by the supervisory fuzzy logic controller and passed to the lower-level controller, according to ammonia and nitrate profiles within SBR. It is crucial to adjust this element properly in order to maximize wastewater treatment efficiency and reduce plant costs, especially for the aeration system. A notable aspect of nitrate-based hierarchical control scheme is to curtail the fresh oxygen use since nitrate (S NO ), a product of nitrification, is utilized for limiting aeration costs. Six distinct control techniques are implemented of which PI and FPI controllers for control of DO at the lower level. Four types of hierarchical ammonia and nitrate-based controllers employing intelligent fuzzy control are deployed. Addition of fuzzy controller contributes to an airflow reduction of 40.08% for ammonia control and 31.58% for nitrate control strategies. This study highlights the superiority of the ammonia-based control strategy, particularly coupled with lower-level FPI controller, based on its ability to minimize airflow without affecting effluent quality. These findings offer helpful insights for advancing the field of wastewater treatment, improving efficiency, and promoting cost-effective and sustainable practices in SBR.

XRL-FlexSBR: Multi-agent reinforcement learning-driven flexible SBR control with explainable performance guarantee under diverse influent conditions

Accelerating Reinforcement Learning with Local Data Enhancement for Process Control

A novel model-based adaptive control strategy for step-feed SBRs dealing with influent fluctuation.

Unified control of diverse actions in a wastewater treatment activated sludge system using reinforcement learning for multi-objective optimization

Intelligent sequencing batch reactor control from theory, through modelling, to full-scale application

Multi-agent reinforcement learning-enhanced autonomous calibration method for wastewater treatment modeling: Long-term validation of a full-scale plant

Optimal control towards sustainable wastewater treatment plants based on multi-agent reinforcement learning

An Influent Responsive Control Strategy with Machine Learning: Q-learning Based Optimization Method for a Biological Phosphorus Removal System.

Accelerating reinforcement learning with case-based model-assisted experience augmentation for process control

Optimization of Operation Parameters Towards Sustainable WWTP Based on Deep Reinforcement Learning.

Development of AI-based process controller of sour water treatment unit using deep reinforcement learning

Intelligent Control of Wastewater Treatment Plants Based on Model-Free Deep Reinforcement Learning

Intelligent Control/Operational Strategies in WWTPs through an Integrated Q-Learning Algorithm with ASM2d-Guided Reward

Dissolved oxygen concentration control in wastewater treatment process based on reinforcement learning

Effluent quality improvement in sequencing batch reactor-based wastewater treatment processes using advanced control strategies

Combined Sewer Overflow and Flooding Mitigation Through a Reliable Real‐Time Control Based on Multi‐Reinforcement Learning and Model Predictive Control

Aerobic Granular Sludge Operation and Nutrients Removal Mechanism in a Novel Configuration Reactor Combined Sequencing Batch Reactor and Continuous-Flow Reactor.

Reinforcement learning-based particle swarm optimization for sewage treatment control

Design of supervisory fuzzy control for enhanced energy saving in a sequencing batch reactor-based wastewater treatment plant

Artificial intelligence control of a sequencing batch reactor for nitrogen removal via nitrite from landfill leachate

Optimization of a Novel Engineered Ecosystem Integrating Carbon, Nitrogen, Phosphorus, and Sulfur Biotransformation for Saline Wastewater Treatment Using an Interpretable Machine Learning Approach.