Abstract
In this paper, a fast and accurate optimization framework is proposed to compute optimal aeration policies in SBR processes under partial nitrification. The optimization framework aims to determine an optimal intermittent aeration profile which minimizes both the operation time of the SBR cycle and the energy required for aeration. Special consideration is given to the fact that the results not only need to be accurate but also to converge within a short time. Moreover, methods to avoid nitrate formation are analyzed and implemented. It is demonstrated that the implementation of a nonlinear model "5-state" and the reduction of the optimization problem to three control variables are the keystones to an efficient solution strategy which achieves fast, robust, and accurate computation of the optimal intermittent aeration profile for any given conditions of the process. The optimization approach is so efficient that it can also be implemented with more complex models such as the ASM3 extended for a two-step nitrification-denitrification process.