Hybrid Intelligent Control Scheme for Activated Sludge Wastewater Treatment

This paper presents a neural network identification scheme to estimate substrate, biomass and dissolved oxygen concentrations in an activated sludge wastewater treatment. This scheme is based on a discrete-time high order neural network (RHONN) trained on-line with an extended Kalman filter (EKF)-based algorithm. Then, the identification scheme is associated with a fuzzy control to regulate the ratio between substrate and biomass concentrations. Obtained simulation results are very encouraging.