Estimating the unbalance of a rotating machine from a single run-down

Reliable estimation of a flexible foundation model and state of unbalance (both amplitude and phase) of a turbogenerator from measured vibration data on a machine during run-down is an active research area. Earlier studies on the estimation of both these parameters used the whole frequency range of the run-down in a single band. However, such identification may be in error for large flexible foundations having many numbers of modes in the run-down frequency range. For reliable identification in such cases, the whole frequency range has to be divided into a number of frequency bands and the frequency dependent foundation models have to be estimated along with the unbalance. The theoretical concept, the complete computational implementation used and the various limitations observed during estimation of foundation models and state of unbalance (both amplitude and phase) are presented in this paper. The method is demonstrated using experimental data from a machine with 4 journal bearings.