Development of Innovative Predictive Control Strategies for Nutrient Removal

Water resource recovery facilities (WRRFs) are under pressure to improve effluent quality and reduce capital and operational costs. One emerging cost-saving strategy is the use of advanced process control concepts for nutrient management that combine online measurements and real-time process control. This project aims to develop and test a hybrid (machine learning + mechanistic model) nutrient management controller at three different WRRFs. The controller will demonstrate both short-term optimization functions and long-term predictive capabilities. The design of the controller, its performance at the WRRFs, and a description of future improvements will be documented. Research partner: Alexandria Renew Enterprises.