Nonlinear optimal control of an alternating activated sludge process in a pilot plant

Abstract The application of a nonlinear optimal control technique to an alternating activated sludge process is presented. Faced with the time-varying features of the process, a parameter estimation procedure is designed and implemented based on a relatively simple process model established by the authors. New process variables are found and defined and a simplified state space model is developed to describe the nitrification and denitrification dynamics. The optimal control problem is formulated based on a criterion to minimize the daily average effluent nitrogen content in the face of a typical diurnal load variation. As the model and control problem exhibits strong nonlinearities, an iterative Newton-Raphson optimization method is applied to the problem. Simulation results and experiments in a pilot plant show that the new model and optimal control approach is successful and effective for improved nitrogen removal in the waste water treatment plant.