Revisiting the separation principle for improved transition control

Feedback control can be used to suppress transient energy growth and delay turbulent transition in shear flows. The separation principle of modern control theory is commonly invoked to design observer-based control laws, whereby a full-state feedback controller and a state estimator are designed independently, then combined to achieve an output feedback flow control law. In previous work, we established that transient energy growth can never be fully eliminated by observer-based control, even when an associated full-state feedback law can fully suppress such growth. In this paper, we use a linearized channel flow to show that observer-based feedback will lead to higher levels of transient energy growth than if no control is used at all. Further, we show that transient energy growth can actually be reduced via optimal static output feedback controllers, thus overcoming the performance limitations of observer-based designs. We introduce a modified AndersonMoore algorithm for efficiently computing optimal static output feedback controllers, then show that the resulting controllers reduce the worst-case transient energy growth relative to the uncontrolled system and to observer-based designs. Further, our results indicate that optimal static output feedback exhibits robustness to Reynolds number variations and modeling uncertainty. The result of this study highlight the advantages of optimal static output feedback control over observer-based designs and create opportunities for realizing improved transition control strategies in the future.

[1]  George Papadakis,et al.  A linear state-space representation of plane Poiseuille flow for control design: a tutorial , 2006, Int. J. Model. Identif. Control..

[2]  D. S. Henningson,et al.  Transition delay using control theory , 2011, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[3]  Pertti M. Mäkilä,et al.  A descent anderson-moore algorithm for optimal decentralized control , 1985, Autom..

[4]  W. Yan Static Output Feedback Stabilization , 2006, TENCON 2006 - 2006 IEEE Region 10 Conference.

[5]  James Lam,et al.  Static Output Feedback Stabilization: An ILMI Approach , 1998, Autom..

[6]  Ekkehard W. Sachs,et al.  Computational Design of Optimal Output Feedback Controllers , 1997, SIAM J. Optim..

[7]  Anne E. Trefethen,et al.  Hydrodynamic Stability Without Eigenvalues , 1993, Science.

[8]  Thomas Bewley,et al.  A Linear Systems Approach to Flow Control , 2007 .

[9]  W. Brogan Modern Control Theory , 1971 .

[10]  James F. Whidborne,et al.  On the Minimization of Maximum Transient Energy Growth , 2007, IEEE Transactions on Automatic Control.

[11]  Kathryn M. Butler,et al.  Three‐dimensional optimal perturbations in viscous shear flow , 1992 .

[12]  P. Dorato,et al.  Static output feedback: a survey , 1994, Proceedings of 1994 33rd IEEE Conference on Decision and Control.

[13]  John C. Doyle,et al.  Guaranteed margins for LQG regulators , 1978 .

[14]  J. Whidborne,et al.  Computing the maximum transient energy growth , 2011 .

[15]  Thomas Bewley,et al.  Flow control: new challenges for a new Renaissance , 2001 .

[16]  Maziar S. Hemati,et al.  Dynamic mode shaping for fluid flow control: New strategies for transient growth suppression , 2017 .

[17]  James F. Whidborne,et al.  MINIMIZATION OF MAXIMUM TRANSIENT ENERGY GROWTH BY OUTPUT FEEDBACK , 2005 .

[18]  Brian D. O. Anderson,et al.  Linear Optimal Control , 1971 .