Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 106
PROCEEDINGS OF THE TWELFTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STRUCTURES TECHNOLOGY Edited by:
Paper 124
Application of a Bounded-Variable Least-Squares Algorithm for Damage Detection based on Natural Frequencies Changes M. Caretto, A. Messina and N.I. Giannoccaro
Dipartimento di Ingegneria dell'Innovazione, Università del Salento, Lecce, Italy M. Caretto, A. Messina, N.I. Giannoccaro, "Application of a Bounded-Variable Least-Squares Algorithm for Damage Detection based on Natural Frequencies Changes", in , (Editors), "Proceedings of the Twelfth International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 124, 2014. doi:10.4203/ccp.106.124
Keywords: damage detection, modal analysis, structural dynamics, finite element, inverse problem, constrained least-square algorithms..
Summary
The aim of this paper is to test the validity of a bounded variable least-squares
(BVLS) algorithm in solving the problem of detecting the location and the extent of
potential damaged sites in elastic structures by using the changes in their natural
frequencies. In the work, described in this paper, the BVLS algorithm has been
implemented and then tested as a new tool for identifying and quantifying damaged
sites on the finite element model of a one side clamped beam. Several tests are
illustrated evaluating the performances of this innovative approach in comparison
with the NNLS and MDLAC algorithms in different conditions (including different
extents of damage, different numbers and locations of damaged sites and different
numbers of discretization elements). The results have been presented and discussed
enabling the field of applicability to be clarified, the improvements and the limits of
this new procedure in comparison with the established techniques.
purchase the full-text of this paper (price £20)
go to the previous paper |
|