Real-time Model Predictive Control of a Wastewater Treatment Plant based on Machine Learning

In wastewater treatment two separate goals should be jointly pursued: nutrient removal and energy conservation. An efficient controller performance should cope with process uncertainties, seasonal variations and process nonlinearities. This paper describes the design and testing of a model predictive controller (MPC) based on neuro-fuzzy techniques, capable of estimating the main process variables and provide the right amount of aeration to achieve an efficient and economical operation. The algorithm has been field-tested on a large-scale municipal WWTP of about 500.000 PE with encouraging results in terms of better effluent quality and energy savings.