Calibration, Identifiability and Optimal Experimental Design of Activated Sludge Models Calibratie, Identificeerbaarheid En Optimale Proefopzet Voor Actief-slib Modellen

Thesis submitted in fulfillment of the requirements for the degree of Doctor (Ph.D) in Applied Biological Sciences Proefschrift voorgedragen tot het bekomen van de graad van Doctor in de Toegepaste Biologische Wetenschappen op gezag vanSummary The activated sludge process is one of the most widespread biological wastewater purification technologies. Especially during the last two decades modelling of biological degradation processes in activated sludge plants has been an important research topic. This thesis deals with calibration, identifiability and optimal experimental design of activated sludge models. The aim was to define a more methodological approach for model calibration with special focus on the investigation, illustration and solution of the problems-of both theoretical and practical origin-encountered when deriving information from lab-scale experiments. The focus in this thesis is on the Activated Sludge Model No. 1 (ASM1; Henze et al., 1987). However, the developed methodologies are transferable to model calibration applications in general. Generally speaking, a good model calibration exercise can benefit from the information derived from lab-scale experiments. In such lab-scale experiments it is tried to characterise (part of) the kinetics of the biomass present in the full-scale system under study. In the frame of this thesis a conceptual methodology to carry out lab-scale experiments was defined, investigated and illustrated, with the purpose of obtaining accurate parameter estimates of activated sludge reaction kinetics and wastewater component concentrations. For the determination of the optimal experiments the theory of optimal Experimental Design (OED), based on the Fisher Information Matrix (FIM), was applied. The literature review (chapter 2) begins with a detailed description of ASM1, and a discussion of the differences between ASM1 and the recently proposed ASM3 (Gujer et al., 1999). However, the main part of the literature review is focusing on ASM1 model calibration methodology and available methods for characterisation of wastewater components and of activated sludge reaction kinetics. The different methods were discussed and evaluated thoroughly. Furthermore, the relevance of characterising the different wastewater components and kinetic parameters via lab-scale experiments was discussed and a list of the most relevant parameters and components was defined. Based on the literature review it was concluded that especially respirometry, and to a lesser extent titrimetry and nitrate uptake rate measurements, are powerful methods that allow for the characterisation of several activated sludge kinetic parameters and wastewater components. With the literature review in mind, and to maximise the quality of the experimental data, a combined respirometric-titrimetric measurement …