Optimization of biological nutrient removal in a SBR using simulation-based iterative dynamic programming

The purpose of this study was to simulate and optimize the nitrogen removal of a sequencing batch reactor (SBR) through the use of a simplified model derived from activated sludge model no. 1 (ASM1) and iterative dynamic programming (IDP), while meeting the treatment requirements. A new performance index for SBR optimization is proposed on the basis of minimum area criteria that consider the minimum batch time and the maximum nitrogen removal so as to minimize the energy consumption. Choosing area as the performance index simplifies the optimization problem and the use of appropriate weights in the performance index makes it possible to minimize the time and energy of the SBR simultaneously. In the optimized system, the optimal set-point of dissolved oxygen affects both the batch time and energy savings. For four different influent loadings, simulation results by IDP-based SBR optimizations suggest that batch scheduling, the set-point trajectory of dissolved oxygen concentration and the amount of external carbon all require supervisory control in order to achieve the optimal energy-saving concentration of total nitrogen in the effluent. Simulation results of the SBR show that the total energy cost can be reduced by up to 20% for the aerobic phase and 10% for the anoxic phase with maximum nitrogen removal.