Technical Briefs

Optimal Boundary Control of Reaction–Diffusion Partial Differential Equations via Weak Variations

[+] Author and Article Information
Scott J. Moura

UC President's Postdoctoral Fellow
Mechanical & Aerospace Engineering,
University of California, San Diego,
La Jolla, CA 92093
e-mail: smoura@ucsd.edu

Hosam K. Fathy

Assistant Professor
Mechanical & Nuclear Engineering,
Pennsylvania State University,
University Park, PA 16802
e-mail: hkf2@psu.edu

Contributed by the Dynamic Systems Division of ASME for publication in the JOURNAL OF DYNAMIC SYSTEMS, MEASUREMENT, AND CONTROL. Manuscript received January 19, 2012; final manuscript received November 21, 2012; published online February 21, 2013. Assoc. Editor: Rama K. Yedavalli.

J. Dyn. Sys., Meas., Control 135(3), 034501 (Feb 21, 2013) (8 pages) Paper No: DS-12-1022; doi: 10.1115/1.4023071 History: Received January 19, 2012; Revised November 21, 2012

This paper derives linear quadratic regulator (LQR) results for boundary-controlled parabolic partial differential equations (PDEs) via weak variations. Research on optimal control of PDEs has a rich 40-year history. This body of knowledge relies heavily on operator and semigroup theory. Our research distinguishes itself by deriving existing LQR results from a more accessible set of mathematics, namely weak-variational concepts. Ultimately, the LQR controller is computed from a Riccati PDE that must be derived for each PDE model under consideration. Nonetheless, a Riccati PDE is a significantly simpler object to solve than an operator Riccati equation, which characterizes most existing results. To this end, our research provides an elegant and accessible method for practicing engineers who study physical systems described by PDEs. Simulation examples, closed-loop stability analyses, comparisons to alternative control methods, and extensions to other models are also included.

Copyright © 2013 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

A visualization of the weak variations concept for optimal state and control trajectories

Grahic Jump Location
Fig. 2

Simulation example of state trajectories for the (a) open-loop and (b) closed-loop systems

Grahic Jump Location
Fig. 3

Boundary control input for the LQR controller

Grahic Jump Location
Fig. 4

Real part of eigenvalues for closed-loop spectrum of infinite-time LQR controlled system

Grahic Jump Location
Fig. 5

Block diagram for open-loop transfer function in negative feedback form to study stability robustness to uncertainty at the plant input

Grahic Jump Location
Fig. 6

Nyquist plot of G(s), defined in (46)

Grahic Jump Location
Fig. 7

Feedback control gains for the infinite-time LQR (solid lines) and backstepping (dashed) controllers. Parameter q corresponds to the state penalty: Q(x,y) = q sin(πx) sin(πy). Parameter γ is the reaction coefficient in the backstepping target system (55).




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In