000013114 001__ 13114
000013114 005__ 20161114160328.0
000013114 04107 $$aeng
000013114 046__ $$k2009-06-22
000013114 100__ $$aGivoli, D.
000013114 24500 $$aReduction of dynamic systems and subsystems

000013114 24630 $$n2.$$pComputational Methods in Structural Dynamics and Earhquake Engineering
000013114 260__ $$bNational Technical University of Athens, 2009
000013114 506__ $$arestricted
000013114 520__ $$2eng$$aLarge dynamic systems appear in many applications, including the analysis of large structures, where the motivation for the present work arises. Often a detailed time-dependent computational model leads to such a large semi-discrete system that precludes solution in a reasonable time or poses unattainable memory requirements. In those cases it is standard to seek a reduced model. Model reduction is a very active and rich field of research, especially, but not exclusively, in the contexts of robust control and structural analysis. A special class of model reduction problems is that associated with the reduction of subsystems, where the goal is to reduce only certain parts of the system while affecting the dynamic behavior of the rest of the structure to the minimal extent possible. In this talk, which presents work done jointly with P. Barbone, I. Patlashenko and S. Tayeb, the general subject of model reduction will be briefly reviewed, and the special characteristics of the problem of subsystem reduction will be outlined. Then the method of Optimal Modal Reduction (OMR) of Barbone and Givoli will be discussed. In addition, recent advances related to this method will be presented, including an improved OMR formulation, comparison of OMR with a reduction based on mesh coarsening, and an extension to damped systems. Numerical examples will be used to demonstrate these advances.

000013114 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb.
000013114 653__ $$aModel Reduction, Modal Reduction, Subsystem, Dirichlet-to-Neumann Map, Rayleigh Damping, Coarsening. Abstract. Large dynamic systems appear in many applications, including the analysis of large structures, where the motivation for the present work arises. Often a detailed time-dependent computational model leads to such a large semi-discrete system that precludes solution in a reasonable time or poses unattainable memory requirements. In those cases it is standard to seek a reduced model. Model reduction is a very active and rich field of research, especially, but not exclusively, in the contexts of robust control and structural analysis. A special class of model reduction problems is that associated with the reduction of subsystems, where the goal is to reduce only certain parts of the system while affecting the dynamic behavior of the rest of the structure to the minimal extent possible. In this talk, which presents work done jointly with P. Barbone, I. Patlashenko and S. Tayeb, the general subject of model reduction will be briefly reviewed, and the special characteristics of the problem of subsystem reduction will be outlined. Then the method of Optimal Modal Reduction (OMR) of Barbone and Givoli will be discussed. In addition, recent advances related to this method will be presented, including an improved OMR formulation, comparison of OMR with a reduction based on mesh coarsening, and an extension to damped systems. Numerical examples will be used to demonstrate these advances.

000013114 7112_ $$aCOMPDYN 2009 - 2nd International Thematic Conference$$cIsland of Rhodes (GR)$$d2009-06-22 / 2009-06-24$$gCOMPDYN2009
000013114 720__ $$aGivoli, D.
000013114 8560_ $$ffischerc@itam.cas.cz
000013114 8564_ $$s448490$$uhttps://invenio.itam.cas.cz/record/13114/files/CD152.pdf$$yOriginal version of the author's contribution as presented on CD, section: Semi-plenary lectures.
000013114 962__ $$r13074
000013114 980__ $$aPAPER