Parameter Selection Strategies in Finite Element Model Updating

Abstract

In FE model updating, as in any identification procedure, we select some parameters in the model, and try to fine-tune them to minimize the discrepancy between the model predictions and the measured data. The paper compares the performance of the generic element matrices, recently introduced by the authors, with other selection strategies for finite element model updating. The updated models obtained from these methods are compared with the measured data and rated according to their ability to produce the measured data within and beyond the frequency range used in the updating, and more importantly, according to their ability to predict the effect of design changes.