More insight on function-weighted frequency response function sensitivity method for analytical model updating

doi:10.1016/j.jsv.2021.116143

2021-09-01

PRE

Akbar Esfandiari

Masoud Sanayei *

DOI：10.1016/j.jsv.2021.116143

原文链接

原文求助

摘要

En 中文

Structural model updating and parameter estimation by sensitivity-based methods are studied by many researchers. Extraction of accurate sensitivity relations have great im-pacts on parameter estimation accuracy. In this paper, assumptions made by different re-searchers and performance of the function-weighted frequency response function sensitiv-ities are investigated. The weighted form of the sensitivity function is reformulated. The findings demonstrate that careful consideration of alternate definitions of the weighted form of the sensitivity relations and solution techniques used are paramount in structural parameter estimation for finite element model updating. Different approaches for the def-inition of unknown parameters, weighting of sensitivity equation, and solution techniques are compared for structural model updating. The most accurate parameter estimates are obtained by the least squares method considering normalized structural parameters as un-knowns in structural parameter estimation. (c) 2021 Elsevier Ltd. All rights reserved. Structural parameter estimation based on measured responses is used for finite element model updating and damage detection of different types of structures. Hence, static and dynamic data are considered to extract damage sensitive features and indices by many researchers. The sensitivity-based methods formulate the change of structural responses and the extracted features as a function of the changes of structural parameters. The selected structural responses and features, definition of unknown parameters, measurement and excitation setup, accuracy of the sensitivity relations, and solution techniques used can highly affect the model updating accuracy. Frequency response function (FRF) describes structural behavior at different excitation frequencies and contain large

Keyword:

Structural vibrations

Sensitivity-based methods

Frequency response function

Receptance function

Finite element model updating

Condition assessment

Structural health monitoring