Model-based evaluation of an integrated high-rate activated sludge and mainstream anammox system

Wastewater treatment plants of the future aim at energy autarky. This could potentially be realized through a high-rate activated sludge (HRAS) process for organics (expressed as Chemical Oxygen Demand, COD) redirection followed by a mainstream partial nitritation-anammox (PNA) process for nitrogen removal. This combination of processes was evaluated in this study, through modelling and simulation. The impact of operating conditions on the unit processes was investigated first. Operation of a HRAS stage often implied a trade-off between maximizing the COD capture for energy recovery and minimizing residual COD in the effluent fed to the subsequent autotrophic PNA process. Moderate DO concentrations (0.3-0.5 g O-2.m(-3)) and SRT (0.3-0.5 d) were suggested to balance these needs, whereas maximizing settling efficiency in the subsequent settler was always desirable. Regarding the mainstream PNA process, the optimal DO setpoint corresponding with maximum nitrogen removal decreased with increasing biomass concentrations. Anammox remained the dominating nitrogen removal process during long-term dynamic simulations with fluctuating HRAS stage effluent (1.3-4.3 g COD.g N-1, 10-20 degrees C), indicating the resilience and long-term stability of the PNA process at mainstream conditions. Overall, plant-wide evaluations revealed that the combined HRAS-PNA system could achieve a comparable effluent quality as the conventional activated sludge (CAS) system, complying with EU regulations, while allowing around 50% more influent COD to be redirected to the sludge line for energy recovery and over 60% savings in aeration energy. This illustrates the potential of being energy-neutral of the integrated HRAS-PNA system. However, the effluent quality of the HRAS-PNA system was found less satisfactory under dynamic conditions than under steady-state conditions.