(586e) Iterative Distributed Model Predictive Control of Nonlinear Systems: Handling Delayed Measurements

The development of distributed model predictive control (DMPC) schemes is of particular interest for the process industries because of the possibility of augmenting the sensor and actuation capabilities of control systems using hybrid communication networks that take advantage of an efficient integration of the existing, point-to-point communication networks (i.e., wired connections from each actuator or sensor to the control system using dedicated local area networks) and additional wired or wireless networked actuator or sensor devices. However, the design of networked control systems has to account for the dynamics introduced by the communication network which may include time-varying delays or data losses in feedback. On the other hand, measuring difficulties of some system states, for example, species concentrations, in chemical processes also introduce the presence of asynchronous and delayed feedback in control loops. Previous work on model predictive control (MPC) design for systems subject to asynchronous or delayed feedback has primarily focused on centralized MPC designs and little attention has been given to the design of DMPC for systems subject to asynchronous or delayed measurements. In a recent paper [1], the issue of delays in the communication between distributed controllers was addressed. In our previous work [2], we developed sequential DMPC schemes for nonlinear systems subject to asynchronous and delayed state feedback; however, we did not address the design of iterative DMPC for nonlinear systems subject to delayed state feedback.

Motivated by the above considerations, in this work, we propose iterative DMPC scheme for large scale nonlinear systems subject to delayed state feedback. Specifically, under the assumption that there exists an upper bound on the maximum feedback delay, we re-design the iterative DMPC scheme presented in [3] to take explicitly into account delays in feedback. This design takes advantage of the bi-directional communication network used in the iterative DMPC framework. Sufficient conditions under which the proposed distributed control design guarantees that the states of the closed-loop system are ultimately bounded in a region that contains the origin are provided. The theoretical results are illustrated through a catalytic alkylation of benzene process example.

[1] E. Franco, L. Magni, T. Parisini, M. M. Polycarpou, and D. M. Raimondo, ?Cooperative constrained control of distributed agents with nonlinear dynamics and delayed information exchange: A stabilizing receding-horizon approach,? IEEE Transactions on Automatic Control, vol. 53, pp. 324?338, 2008.

[2] J. Liu, D. Munoz de la Pena, and P. D. Christofides, ?Distributed model predictive control of nonlinear systems subject to asynchronous and delayed measurements,? Automatica, vol. 46, pp. 52?61, 2010.

[3] J. Liu, X. Chen, D. Munoz de la Pena, and P. D. Christofides, ?Sequential and iterative architectures for distributed model predictive control of nonlinear process systems,? AIChE Journal, in press.