Abstract
A new optimization technique for establishing control setpoints for an activated sludge process based on Activated Sludge Model No.2d (ASM2d) and Genetic Algorithm (GA) has been developed. Process performance predicted by ASM2d, which was compared with the data measured at an existing wastewater treatment plant and confirmed to be sufficiently reliable, was used as the objective function in determining new setpoints by the proposed optimization technique. It has been shown that this technique can determine the combinations of good setpoints that would yield effluent nitrogen and phosphorus concentrations below 10mg-N·l-1 and 1mg-P·l-1, respectively. Moreover, these setpoints were found to be economically advantageous in comparison with other operational conditions. The proposed new optimization technique is expected to make significant contributions to the enhancement of biological nutrient removal in the activated sludge process.
