Model-based nonlinear filtering for Activated Sludge Process

Abstract

The paper address the design of new nonlinear filtering algorithm for the estimation of the non-measurable states in wastewater treatment plants (WWTPs). Two approaches are considered based on the state-dependent formulation and the H-infinity control design technique. It differs from other well-known methods for state estimation, because not only nonlinear, but also the robustness and the load estimation are considered. The filter is designed and implemented on a benchmark model of an Activated Sludge Process (ASP). The different nonlinear filters presented have been tested using the benchmark's data of different weather conditions, i.e. dry, rainy and storm, showing good agreement in all the estimated states. The simulation study presents a comparison of the Extended Kalman Filter (EKF), the state-dependent Riccati filter (SDRF), the extended H-infinity filter (EHF) and finally the new proposed state-dependent H-infinity filter (SDHF).