Advanced Search

Query:

Fill in the form and click »Search«...

Format:

Display: entries per page entries

Zbl 1091.65061
Biros, George; Ghattas, Omar
Parallel Lagrange-Newton-Krylov-Schur methods for PDE-constrained optimization. I: The Krylov-Schur solver. (English)
[J] SIAM J. Sci. Comput. 27, No. 2, 687-713 (2005). ISSN 1064-8275; ISSN 1095-7197/e

Summary: Large-scale optimization of systems governed by partial differential equations (PDEs) is a frontier problem in scientific computation. Reduced quasi-Newton sequential quadratic programming (SQP) methods are state-of-the-art approaches for such problems. These methods take full advantage of existing PDE solver technology and parallelize well. However, their algorithmic scalability is questionable; for certain problem classes they can be very slow to converge. In this two-part article we propose a new method for steady-state PDE-constrained optimization, based on the idea of using a full space Newton solver combined with an approximate reduced space quasi-Newton SQP preconditioner. The basic components of the method are a Newton solution of the first-order optimality conditions that characterize stationarity of the Lagrangian function; the Krylov solution of the Karush-Kuhn-Tucker (KKT) linear systems arising at each Newton iteration using a symmetric quasi-minimum residual method; preconditioning of the KKT system using an approximate state/decision variable decomposition that replaces the forward PDE Jacobians by their own preconditioners, and the decision space Schur complement (the reduced Hessian) by a Broyden-Fletcher-Goldfarb-Shanno approximation initialized by a two-step stationary method. Accordingly, we term the new method Lagrange-Newton-Krylov-Schur (LNKS). It is fully parallelizable, exploits the structure of available parallel algorithms for the PDE forward problem, and is locally quadratically convergent. In part I of this two-part article, we investigate the effectiveness of the KKT linear system solver. We test our method on two optimal control problems in which the state constraints are described by the steady-state Stokes equations. The objective is to minimize dissipation or the deviation from a given velocity field; the control variables are the boundary velocities. Numerical experiments on up to 256 Cray T3E processors and on an SGI Origin 2000 include scalability and performance assessment of the LNKS algorithm and comparisons with reduced SQP for up to $1,000,000$ state and 50,000 decision variables. In part II of the article [ibid. 27, No.~2, 714--739 (2005; reviewed below)], we address globalization and inexactness issues, and apply LNKS to the optimal control of the steady incompressible Navier-Stokes equations.

MSC 2000:: *65K10 Optimization techniques (numerical methods)
65Y05 Parallel computation (numerical methods)
76D05 Navier-Stokes equations (fluid dynamics)
49J20 Optimal control problems with PDE (existence)
49M37 Methods of nonlinear programming type

Keywords: sequential quadratic programming; nonlinear equations; parallel algorithms; adjoint methods; PDE-constrained optimization; optimal control; Lagrange-Newton-Krylov-Schur methods; Navier-Stokes equation; finite elements; preconditioners; indefinite systems; numerical experiments

Citations: Zbl 1091.65063

Comment on this Item PDF MathML XML ASCII DVI PS BibTeX Online Ordering Article Journal

	Scientific prize winners of the ICM 2010
	Overhang
	Lie groups, physics and geometry. An introduction for physicists, engineers and chemists.

Advanced Search

Zentralblatt MATH Berlin [Germany]