Development of a Reduced-Order Model to Identify Ammonia-Based Aeration Control Proportional-Integral Tuning...

While ammonia-based aeration control (ABAC) significantly improves process efficiency in water resource recovery facilities (WRRFs), its performance is often limited by the difficulty of tuning proportional-integral (PI) controllers amidst dynamic loads and nonlinear reaction kinetics. This study proposed a systematic tuning approach that derives first-order plus deadtime (FOPDT) parameters from a reduced-order model based on empirical reactor hydraulics. Furthermore, the nonlinearity of Monod saturation kinetics, which describe the impact of dissolved oxygen (S_O2) on nitrification rates, is explicitly integrated into the feedback control structure to linearize the system response. The approach was validated via both a model-based simulator and full-scale implementation. In the model-based simulation, both controller structures provided stable performance, but the direct S_O2 controller showed nonlinear overshoot during high load periods, while the inclusion of Monod kinetics in ABAC linearized the response, particularly when tuned with the reduced-order model (mean absolute error (MAE) 0.09 mg N/L). At the full-scale plant, when tuned using the proposed method, the controller demonstrated stable performance and successfully attenuated dynamic loads to achieve a low 0.16 mg N/L MAE. These results demonstrate that combining reduced-order modeling with kinetic-based control structures offers a robust automatable alternative to heuristic tuning methods.