Inherent Robustness of MPC and the Integration of Scheduling and Control

Douglas A. Allan

A robust model predictive controller is one in which small amounts of model-system mismatch or noise do not destabilize the controlled system. MPC is often inherently robust (Pannochia, Rawlings, and Wright, 2011) and therefore doesn't need to be explicitly designed to tolerate error. I am investigating the conditions under which MPC is inherently robust.

Integration of production scheduling and process control is a subject rich with research opportunities. The high level, low-detail models often used by schedulers can result in suboptimal process economics because they do not take into account detailed process dynamics, and sometimes the resulting schedules are found to be infeasible at the control level.

Recently, a state space model for scheduling has been proposed in collaboration with the Maravelias group (Subramanian, Maravelias, and Rawlings, 2012). This is the first step towards applying results from control theory to scheduling problem. In collaboration with Exxon, I am investigating these properties in a combined scheduling and control problem.


G. Pannocchia, J. B. Rawlings, and S. J. Wright. Conditions under
which suboptimal nonlinear MPC is inherently robust.
System and Control Letters, 60:747:755,2011.

K. Subramanian, C. T. Maravelias, and J. B. Rawlings.
A state-space model for chemical production scheduling.
Computers and Chemical Engineering, 47:97-110, 2012.