Integration of Design and Control using an MPC-based superstructure for control structure selection.

Abstract A MINLP framework is used to solve the simultaneous process design and control of a waste water treatment plant including the selection of the control structure. The novelties of the work are that MPC controllers with different interconnection structures are considered and that the resulting closed loop variability costs are calculated using analytical bounds derived from robust control tests. Both centralized and decentralized MPC controllers involving different combinations of manipulated and controlled variables are considered as candidates within the methodology. The proposed simultaneous design and control methodology was applied to a waste-water treatment industrial plant.