Comparison of Two Model Concepts for Simulation of Nitrogen Removal at a Full-Scale Biological Nutrient Removal Pilot Plant

The experimental studies conducted at the Hanover-Gümmerwald pilot wastewater treatment plant (WWTP) focused on minimizing nitrogen loads discharged during stormwater events. The data collected during the plant operation were used for a long-term process simulation. The aim of this study was to compare predictive capabilities of two different mechanistic models (ASM2d and ASM3P) in terms of nitrogen removal. The influent wastewater composition was generated using on-line measurements of only three parameters (COD, $\mathrm{N}–\mathrm{N}{\mathrm{H}}_4^{+}$ , $\mathrm{P}–\mathrm{P}{\mathrm{O}}_4^{3-}$ ) and the model predictions were primarily compared with on-line data (concentrations of $\mathrm{N}–\mathrm{N}{\mathrm{H}}_4^{+}$ , $\mathrm{N}–\mathrm{N}{\mathrm{O}}_3^{-}$ ) originating from the aerobic zone of the bioreactor. The simulation results confirmed the experimental data concerning the capabilities of the system for handling increased flows during stormwater events. The predicted peaks of $\mathrm{N}–\mathrm{N}{\mathrm{H}}_4^{+}$ at the line with the quadruple dry weather flow rate were normally exceeding $8\mathrm{g}\mathrm{N}\bullet {\mathrm{m}}^{-3}$ (similar to the observations), whereas no (or minor) peaks of $\mathrm{N}–\mathrm{N}{\mathrm{H}}_4^{+}$ were predicted for the line with the double dry weather flow rate. The relationships between ASM2d and ASM3P predictions for $\mathrm{N}–\mathrm{N}{\mathrm{H}}_4^{+}$ and $\mathrm{N}–\mathrm{N}{\mathrm{O}}_3^{-}$ were highly correlated $(r^2=0.83–0.99)$ with the slopes remaining close to 1.0. Both models appear to be equally suitable for practical applications in common municipal WWTPs.