Uncorrelated Modes Driven Inverse Eigensensitivity Method for Finite Element Model Updating

Abstract

The inverse eigensensitivity method is one of the most widely used approaches for finite element model updating. One of the requirements for updating using this approach is that the correspondence between the experimental and the analytical modes must be known. However, at times, it may not always be possible to establish with certainty, for some analytical modes, to which experimental modes they correlate or correspond. Eigenvalues and eigenvectors of such modes cannot be used in the updating process, and this represents a draw back of the inverse eigensensitivity method and other iterative methods based on the modal data. If a finite element model has an excessive modeling error, then the correlated mode pairs may not remain stable during the updating process and may undergo changes. This situation also cannot be handled effectively using the current techniques that implicitly assume stability of the mode pairs. This paper presents a new method, the uncorrelated-modes-driven inverse eigensensitivity method, that allows both the correlated as well as the uncorrelated modes to be used in the updating process and addresses the aforementioned shortcomings. Two numerical studies and an experimental study are presented to show the ability of the proposed method to accurately estimate the updating param eters and identify the correct mode pairs.

Finite element model updating of composite with adhesive jointed structure under built-up internal stress

“…Therefore, the modal analysis criterion (MAC) was used to determine the level of correlation between the analytical and measured mode shapes. It is calculated as (Modak, 2015)where boldΦp and boldΦm denote the mode shapes obtained from the predicted and measured results, respectively. The best correlation is expressed by a MAC value close to 1.…”