Identification

Abstract

Rotordynamics tools for computer simulations are available nowadays, usually based on the Finite Element method. These routines allow to include all important components like shafts, impellers, bearings, seals etc. and take into consideration the corresponding effects like inertia, damping, stiffness, gyroscopics, unbalance and fluid structure interaction forces [7]. They predict modal parameters like natural frequencies, damping values, mode shapes and unbalance and transient vibrations as well. While these powerful tools themselves usually work without difficulties, problems more often occur in finding the correct input data. Particularly not all of the physical parameters are available from theoretical derivations. This is especially true for rotordesigns with complicated geometry, shrink fits, additional masses etc. and for the various fluid structure interactions in fluid bearings, squeeze film dampers, seals, impellers etc. described by rotordynamic coefficients. In such cases the required data have to be taken from former experience or have to be determined experimentally via identification procedures. © 2009 Springer-Verlag Berlin Heidelberg.