Adaptive optimal control for large-scale nonlinear systems

Adaptive optimal control for large-scale nonlinear systems

Michailidis, Lakovos (Democritus University of Thrace)
Baldi, S. (TU Delft Team Bart De Schutter; Centre for Research and Technology-Hellas)
Kosmatopoulos, Elias B. (Democritus University of Thrace)
Ioannou, Petros A. (University of Southern California)

2017

In this paper, we present an adaptive optimal control approach applicable to a wide class of large-scale nonlinear systems. The proposed approach avoids the so-called loss-of-stabilizability problem and the problem of poor transient performance that are typically associated with adaptive control designs. Moreover, it does not require the system model to be in a certain parameterized form, and most importantly, it is able to efficiently handle systems of large dimensions. Theoretical analysis establishes that the proposed methodology guarantees stability and exponential convergence to state trajectories that can be made as close as desired to the optimal ones. A numerical example demonstrates the capability of the proposed approach to overcome loss-of-stabilizability problems. Moreover, simulation experiments for energy-efficient climate control performed on a ten-office building demonstrate the effectiveness of the proposed approach in large-scale nonlinear applications.

http://resolver.tudelft.nl/uuid:d26de497-4222-4c52-b11d-aa147a41d339

https://doi.org/10.1109/TAC.2017.2684458

0018-9286

IEEE Transactions on Automatic Control, 62 (11), 5567 - 5577

Institutional Repository

journal article

© 2017 Lakovos Michailidis, S. Baldi, Elias B. Kosmatopoulos, Petros A. Ioannou