September 19, 2016
Fast Robust Model Predictive Control of Advanced Manufacturing Systems
Abstract: Motivated by the needs of advanced manufacturing systems, this presentation describes real-time optimal control design algorithms for dynamical systems that have (1) high to infinite state dimension, (2) parameter uncertainties that can be deterministic or probabilistic, (3) time delays, (4) unstable zero dynamics, (5) actuator, state, and output constraints, (6) stochastic noise and disturbances, and (7) phenomena described by combinations of algebraic, ordinary differential, partial differential, and integral equations (that is, generalizations of descriptor/singular systems). Robust model predictive control (RMPC) formulations are presented that have the flexibility to handle linear dynamical systems with these characteristics, while employing projections and shifting of the most expensive calculations offline so that the online computational cost is low. Implementation to a detailed mechanistic model of an advanced pharmaceutical manufacturing plant demonstrates an order-of-magnitude improved robustness of the product quality to model uncertainties while having an online optimization cost of less than 1 second. Some extensions to nonlinear dynamical systems are discussed.
Bio: Dr. Richard D. Braatz is the Edwin R. Gilliland Professor at the Massachusetts Institute of Technology (MIT) where he does research in control theory and its application to advanced manufacturing. He received an MS and PhD from the California Institute of Technology. Honors include the Donald P. Eckman Award from the American Automatic Control Council, the Curtis W. McGraw Research Award from the Engineering Research Council, the Antonio Ruberti Young Researcher Prize, and the IEEE Control Systems Society Transition to Practice Award. He is a Fellow of IEEE, IFAC, and the American Association for the Advancement of Science.