Robust control synthesis for the activated sludge process

Abstract

Control of the activated sludge process (ASP) is a challenging problem due to the complexity of the biological and chemical reactions, and large variations in the influent flow. Herein, the ASP is presented as a linear parametric varying (LPV) model to account for the parameter changes in the system dynamics. Since the influent flow is the main source of parametric uncertainty but is variable and easily measured, it is chosen as the scheduling parameter. Based on this LPV model with a scheduling parameter, a robust gain-scheduled controller is synthesized to deal with the large uncertainties and neglected dynamic behaviors of the process. Accordingly, the dynamic performance of the controller changes with respect to the scheduling parameter. The ultimate goal is to regulate the dissolved oxygen (DO) concentration and control the biomass concentration. The extensive simulation results show that the robust controller can effectively deal with large uncertainties and unavoidable sensor noises for both linear and nonlinear ASP models, which helps to improve the system performance and saves energy. This study contributes a potential control approach for more robust wastewater treatment plants, which is currently an active research area.