Integration of model-predictive scheduling, dynamic real-time optimization and output tracking for a wastewater treatment process

Abstract

A hierarchical control architecture for the integration of scheduling decisions, dynamic real-time optimization and tracking control is proposed. It is shown how the discrete-continuous problem of simultaneous scheduling and trajectory optimization on a receding horizon can be integrated within the upper control layer of a two-layer architecture. On the upper layer, the optimal plant strategy defined by a sequence of control objectives and optimal control moves is computed on a long time horizon. The lower control layer consists of a state observer and a fast-acting tracking controller which operates on a short time horizon. The properties of this architecture are illustrated by a case study in which a wastewater treatment plant is operated under the influence of external disturbances. Conflicting operational objectives may be valid, depending on the state of the plant and the expected disturbances. By calculating an optimal sequence of operational strategies and control moves on the receding horizon, economically optimal operation of the plant can be achieved.