This paper addresses the impact of decomposition on the closedloop performance and computational efficiency of model predictive control (MPC) of nonlinear process networks. Distributed MPC structures with different communication strategies are designed for regulation of an integrated reactor−separator process. Different system decompositions are also considered, including decompositions into local controllers with minimum interactions obtained via community detection methods. The closed-loop performance and computational effort of the different MPC designs are analyzed. Through such a comprehensive comparison, tradeoffs between performance and computation effort, and the importance of systematic choice of the system decomposition, are documented.

No abstract

We focus on output feedback control of distributed processes whose infinite dimensional representation in appropriate Hilbert subspaces can be decomposed to finite dimensional slow and infinite dimensional fast subsystems. The controller synthesis issue is addressed using a refined adaptive proper orthogonal decomposition (APOD) approach to recursively construct accurate low dimensional reduced order models (ROMs) based on which we subsequently construct and couple almost globally valid dynamic observers with robust controllers. The novelty lies in modifying the data ensemble revision approach within APOD to enlarge the ROM region of attraction. The proposed control approach is successfully used to regulate the Kuramoto‐Sivashinsky equation at a desired steady state profile in the absence and presence of uncertainty when the unforced process exhibits nonlinear behavior with fast transients. The original and the modified APOD approaches are compared in different conditions and the advantages of the modified approach are presented. © 2013 American Institute of Chemical Engineers AIChE J, 59: 4595–4611, 2013

A combined distributed moving horizon estimation and distributed model predictive control architecture is proposed to address the distributed output−feedback control problem for nonlinear process systems. Community detection based on modularity maximization is used to generate separate optimal decompositions for the estimation and control problems on the basis of suitable graphs. The process of benzene alkylation with ethylene is used as a case study to illustrate the application and computational advantages of the proposed control strategy.

No abstract