Coordination of distributed MPC systems through dynamic real-time optimization with closed-loop prediction

Large plants, such as refineries, typically employ a distributed-MPC configuration positioned within a hierarchical automation architecture, with steady-state real-time optimization (RTO) providing set-points to the underlying distributed MPC system. This paper investigates coordination of distributed MPC systems through dynamic RTO (DRTO). A recent DRTO approach considered the use of closed-loop dynamics at the DRTO level in generating the predicted response of the plant under constrained MPC for computing economically optimal set-point trajectories. In this paper, a closed-loop DRTO strategy of this type is applied to determine set-point trajectories for distributed MPC controllers. The performance of the proposed scheme is assessed through evaluation on a case study, and compared to that of centralized and decentralized control.