Abstract

This dissertation presents the development and application of a mechanistic mathematical model in an activated sludge BNR system, where the assessment of a conventional two-step nitrification model was extended. The primary goal was to identify the bottlenecks in modelling full-scale wastewater treatment technologies, focusing on nitrite shunt and comammox. Major challenges (kinetic parametrization and calibration of models) were examined. The research utilized kinetic-based competition mechanisms between AOB and NOB, with the nitrogen removal process controlled by DO concentration and aerobic SRT. A conceptual two-step nitrification model, incorporating comammox activity, was evaluated during the start-up and operation of SBR, operated at elevated temperatures, under limited DO, and fed with ammonia-rich influent wastewater. Local sensitivity analysis (LSA) was applied to identify the most critical kinetic parameters of the newly developed model. Correlation matrix was used to identify pairs of the most sensitive parameters. The model performance was comprehensively evaluated using statistical measures, including RSME, MAE, and NSE. The study highlighted the significant impact of DO on the accuracy of two-step nitrification model predictions. It emphasized the importance of accounting for the distinct comammox pathways. Special attention was given to substrate limitation mechanisms, inhibition processes, the adaptive properties of unacclimatized biomass, and biomass washout.

Author (1)