-------- Original Message -------- Subject: Fwd: Fwd: Abstract Date: Thu, 05 Sep 2013 13:59:06 +0200 From: Christian Rieger <rieger@ins.uni-bonn.de> To: ag-griebel@ins.uni-bonn.de, ag-schweitzer@ins.uni-bonn.de, ag-garcke@ins.uni-bonn.de, Nils Hornung <nils.hornung@scai.fraunhofer.de>, ag-griebel-stud@ins.uni-bonn.de Hallo, fuer den Vortrag morgen um 14:00 im Seminarraum 6.020: Vortragende: Sara Grundel Postdoctoral Research Fellow Computational Methods in Systems and Control Theory Max Planck Institute for Dynamics of Complex Technical Systems Sandtorstr. 1 D-39106 Magdeburg Germany Email: grundel@mpi-magdeburg.mpg.de Tel :+49 391 6110 805 Abstract (siehe auch den Anhang): Interpolation and Model Order Reduction for parametric systems Sara Grundel, Nils Hornung+ und Peter Benner MPI Magdeburg + Fraunhofer SCAI Model Order Reduction (MOR) Methods for linear systems are well studied and many successful methods exist. We will review the H2 optimal MOR and explain some recent advances in Parametric Model Order Reduction. The focus will be on a method where we interpolate the poles of the reduced order model by Radial Basis Function Interpolation. What this means is that the function we want to interpolate are eigenvalues of matrices depending on parameters. This can mean a variety of non nice behaviour in theory. It doesn’t actually show up very frequently in practice but we will still show a way how to deal with it. We will furthermore need to introduce a technique to create a medium size model for computational efficiency. Viele Gruesse, Christian