This paper discusses a new structural damage identification method where a grey relational analysis is applied to the measured frequency response function (FRF) matrix which has been subjected to a singular value decomposition (SVD). A number of vectors, which define the dynamic features of the system by means of normalized and truncated singular values, are used to form a reference matrix describing the variable states of the structure under study. A grey relational identification algorithm is utilized to investigate the discrepancies, i.e. damage, between the reference matrix and a number of test matrices formed using further measured FRF data. The methodology was applied to the healthy and damaged specimens of a railway wheel and the damage was located with good accuracy and repeatability. The FRFs of both healthy and damaged railway wheels were measured in the x, y, z directions at five measurement points along a radius, yielding 15 FRFs, each containing 4,096 spectral lines. It was concluded that the method was well suited to on-line structural healthy monitoring since neither a mathematical model nor a complete set of measurements are needed.
Structural damage identification via grey relational analysis of SVD-processed FRFs
Identifizierung von Strukturschäden mittels einer grauen Relationsanalyse von durch Singulärwertzerlegung behandelten Frequenzgangfunktionen
2000
7 Seiten, 7 Bilder, 4 Tabellen, 14 Quellen
Conference paper
English
Direct structural damping identification method using complex FRFs
Online Contents | 2015
|VIBRATION-BASED DAMAGE IDENTIFICATION USING RECONSTRUCTED FRFS IN COMPOSITE STRUCTURES
Online Contents | 2003
|Finite Element Structural Updating using FRFs (AIAA 2014-0492)
British Library Conference Proceedings | 2014
|Topology Optimization for Robust Damage Localization Using Aggregated FRFs Statistical Criteria
British Library Conference Proceedings | 2014
|