Benchmarking Biological Nutrient Removal in Wastewater Treatment Plants: Influence of Mathematical Model Assumptions

This paper examines the effect of different model assumptions when describing biological nutrient removal (BNR) by the activated sludge models (ASM) 1, 2d & 3. The performance of a nitrogen removal (WWTP1) and a combined nitrogen and phosphorus removal (WWTP2) benchmark wastewater treatment plant was compared for a series of model assumptions. Three different model approaches describing BNR are considered. In the reference case, the original model implementations are used to simulate WWTP1 (ASM1 & 3) and WWTP2 (ASM2d). The second set of models includes a reactive settler, which extends the description of the nonreactive TSS sedimentation and transport in the reference case with the full set of ASM processes. Finally, the third set of models is based on including electron acceptor dependency of biomass decay rates for ASM1 (WWTP1) and ASM2d (WWTP2). The results show that incorporation of a reactive settler: 1) increases the hydrolysis of particulates; 2) reduces the SNO3 concentration (bottom of clarifier); 3) increases the oxidation of COD compounds; 4) increases XB and XA decay; and, finally, 5) increases the growth of XPAO and formation of XPHA for ASM2d. Introduction of electron acceptor dependent decay leads to a substantial increase of the concentration of XA, XB and XPAO in the bottom of the clarifier. The paper ends with a critical discussion of the influence of the different model assumptions, and emphasizes the need for a model user to understand the significant differences in simulation results that are obtained when applying different combinations of ‘standard’ models.