Structure-Preserving Model Order Reduction of Dynamical SystemsOberseminar Analysis

by Prof. Matthias Voigt (FernUni Schweiz)

Thursday May 7, 2026, 2:00 PM → 4:00 PM Europe/Berlin

Endenicher Allee 60/1-016 - Lipschitzsaal (Mathezentrum)

Model order reduction is a field of mathematics that deals with the complexity reduction of mathematical models in order to reduce the computational costs of tasks such as numerical simulation, optimization, or control. In this talk, first an overview of the field will be given and two of the classical systems theoretic reduction methods (balanced truncation and IRKA) will be reviewed. The second part of the presentation will focus on structure-preserving reduction methods for structured models. This includes a variant of balanced truncation for symmetric second-order systems and a rational fitting method for the class of port-Hamiltonian systems. If time allows, recent directions in the direction of kernel methods will be presented.