Damage Detection in Composite Structures Based on Optical Fibre Strain Sensing and Finite Element Model Updating
A prototype system for condition monitoring of composite structures is described that relies on the on-line measurement of dynamic strains in order to detect any deterioration in performance due to the accumulation of damage. Strain data from both long gauge and point optical fibre sensors are emplo...
Gespeichert in:
Veröffentlicht in: | Key engineering materials 2003-01, Vol.245-246, p.509-518 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | A prototype system for condition monitoring of composite structures is described that relies on the on-line measurement of dynamic strains in order to detect any deterioration in performance due to the accumulation of damage. Strain data from both long gauge and point optical fibre sensors are employed to update finite element models of the analysed structures. Together with Bragg grating point sensors, an innovative Fabry-Perot interferometric long gauge strain sensor is proposed. The cavity of these sensors consists of a pair of matched Bragg grating reflectors allowing a number of sensors of the desired length to be wavelength multiplexed on a single fibre, with the advantage of minimum intrusivity in composite materials while retaining sensing capability over the whole structure. A heterodyne based demodulation system, tailored for these sensors, has been employed. Gradient based optimisation algorithms have been utilised to update structural Finite Element models based on the output from the fibre optic sensors and strain based modal analyses. These procedures were used to detect the location of areas with known modifications of the stiffness properties in composite structures and showed good results, as the damaged areas have been correctly located. The comparison with results obtained using more conventional updating techniques shows the validity of the dynamic strain data approach in the damage detection field. |
---|---|
ISSN: | 1013-9826 1662-9795 1662-9795 |
DOI: | 10.4028/www.scientific.net/KEM.245-246.509 |