On an inexact gradient method using Proper Orthogonal Decomposition for parabolic optimal control problems

This paper is devoted to a numerical solution technique for linear quadratic parabolic optimal control problems using the model order reduction technique of Proper Orthogonal Decomposition (POD). The proposed technique is an inexact gradient descent method where the step size is determined with a line-search algorithm evaluating the state and adjoint equations with POD. The gradient is evaluated with a Finite Element method which allows for a recently developed a posteriori error estimation technique to rate the error in the control. The method is compared to another algorithm presented by Tröltzsch and Volkwein (Comput. Optim. Appl. 42(1):43–63 2009).

[1]  J. Hesthaven,et al.  Reduced Basis Approximation and A Posteriori Error Estimation for Parametrized Partial Differential Equations , 2007 .

[2]  B. Moore.  Principal component analysis in linear systems: Controllability, observability, and model reduction , 1981 .

[3]  Fredi Tröltzsch,et al.  A regularization method for the numerical solution of elliptic boundary control problems with pointwise state constraints , 2009, Comput. Optim. Appl..

[4]  Stefan Volkwein,et al.  Galerkin proper orthogonal decomposition methods for parabolic problems , 2001, Numerische Mathematik.

[5]  Stefan Volkwein,et al.  Error estimates for abstract linear–quadratic optimal control problems using proper orthogonal decomposition , 2008, Comput. Optim. Appl..

[6]  Jacques-Louis Lions,et al.  Mathematical Analysis and Numerical Methods for Science and Technology : Volume 4 Integral Equations and Numerical Methods , 2000 .

[7]  Mark Kärcher,et al.  Reduced basis a posteriori error bounds for parametrized linear-quadratic elliptic optimal control problems , 2011 .

[8]  Stefan Volkwein,et al.  POD A-POSTERIORI ERROR BASED INEXACT SQP METHOD FOR BILINEAR ELLIPTIC OPTIMAL CONTROL PROBLEMS , 2010 .

[9]  P. Holmes,et al.  Turbulence, Coherent Structures, Dynamical Systems and Symmetry , 1996 .

[10]  Serkan Gugercin,et al.  H2 Model Reduction for Large-Scale Linear Dynamical Systems , 2008, SIAM J. Matrix Anal. Appl..

[11]  Enrique S. Quintana-Ortí,et al.  Model Reduction Based on Spectral Projection Methods , 2005 .

[12]  D. Rovas,et al.  Output bounds for reduced-basis approximations of symmetric positive definite eigenvalue problems , 2000 .

[13]  Stefan Volkwein,et al.  Model reduction techniques with a-posteriori error analysis for linear-quadratic optimal control problems , 2012 .

[14]  Angelika Bunse-Gerstner,et al.  h2-norm optimal model reduction for large scale discrete dynamical MIMO systems , 2010, J. Comput. Appl. Math..

[15]  F. Tröltzsch.  Optimal Control of Partial Differential Equations: Theory, Methods and Applications , 2010 .

[16]  Stefan Ulbrich,et al.  Adaptive Multilevel Inexact SQP Methods for PDE-Constrained Optimization , 2011, SIAM J. Optim..

[17]  Stefan Volkwein,et al.  POD a-posteriori error estimates for linear-quadratic optimal control problems , 2009, Comput. Optim. Appl..

[18]  Jan Ziems.  Adaptive multilevel SQP-methods for PDE-constrained optimization , 2010 .

[19]  O. C. Zienkiewicz,et al.  A simple error estimator and adaptive procedure for practical engineerng analysis , 1987 .

[20]  Ekkehard W. Sachs,et al.  A priori error estimates for reduced order models in finance , 2013 .

[21]  Stefan Volkwein,et al.  A method of a-posteriori error estimation with application to proper orthogonal decomposition , 2012 .

[22]  C. DeWitt-Morette,et al.  Mathematical Analysis and Numerical Methods for Science and Technology , 1990 .