Parameter and state estimation of the activated sludge process—II. Applications

In a companion article (Kabouris and Georgakakos, 1996) the Linearized Maximum Likelihood (LML) method for the simultaneous estimation of activated sludge process states and parameters from noisy process measurements is presented. The present article describes the application of the LML method to the identification of a nitrifying activated sludge process, modelled by the IAWPRC Activated Sludge Model 1. It is concluded that identification of nine of the model parameters from process measurements is only possible when a sequential estimation scheme is applied and only influential measurements are considered.