Linear and nonlinear model reduction in structural dynamics with application to model updating

Date

2000-08

Journal Title

Journal ISSN

Volume Title

Publisher

Texas Tech University

Abstract

The objective of the model updating/damage detection is to improve or "'update" the structural dynamics model by comparing experimental modal properties to those from the model. The problem is that the number of measured coordinates is generally much fewer than the number of degrees of freedom (DOF) in the model. Two methods are presented to reconcile the differences: (1) "expanded mode shape*' method, in which the measured mode shapes are expanded to the full model dimension, and (2) model reduction, in which the large model is reduced to the measurement dimension. Singular Value Decomposition (SVD) is utilized in the iterative optimization process used to obtain the updated model.

Numerical results illustrate that the model updating/damage detection based on the expanded model shape or on the model reduction is feasible and is effective in terms of accuracy for the example systems given in this dissertation.

Description

Keywords

Engineering models, Linear systems, Nonlinear systems

Citation