Fuzzy modeling of activated sludge wastewater treatment processes and predictive control of dissolved oxygen

In this paper, the problems of fuzzy modeling of activated sludge wastewater treatment processes and predictive control of dissolved oxygen are investigated. A structure identification algorithm is proposed, by which the structure of the fuzzy rules can be constructed according to the Activated Sludge Model No.1 (ASM1). Utilizing the fuzzy c-means cluster algorithm, the required fuzzy space of input variables is derived and the consequent parameters are identified using the method of least squares. Based on the obtained fuzzy model, the control signal is formulated by solving a convex optimization problem such that the dissolved oxygen concentration in an aerobic reactor of the wastewater treatment plant can be maintained at the set-point. Compared with the traditional PID control, the proposed fuzzy modelbased predictive control paradigm achieves satisfactory benefits in terms of both transient and steady performance.