Classification of composite damage from FBG load monitoring signals

Classification of composite damage from FBG load monitoring signals

Rajabzadeh, Aydin (TU Delft Signal Processing Systems)
Hendriks, R.C. (TU Delft Signal Processing Systems)
Heusdens, R. (TU Delft Signal Processing Systems)
Groves, R.M. (TU Delft Structural Integrity & Composites)

2017

This paper describes a new method for the classification and identification of two major types of defects in composites, namely delamination and matrix cracks, by classification of the spectral features of fibre Bragg grating (FBG) signals. In aeronautical applications of composites, after a damage is detected, it is very useful to know the type of damage prior to determining the treatment method of the area or perhaps replacing the part. This was achieved by embedding FBG sensors inside a glass-fibre composite, and analysing the output signal from the sensors. The glass-fibre coupons were subjected to mode-I loading under tension-compression and static tests, in order to induce matrix cracks and delamination damages respectively. Afterwards, using wavelet features extracted from spectral measurements of the FBG sensors, classification of the damage type was carried out by means of support vector machines as a general classification tool with a quadratic kernel.

classification
delamination
Fibre Bragg grating (FBG)
load monitoring
matrix crack
support vector machines (SVM)
wavelet features

http://resolver.tudelft.nl/uuid:2535d0c8-fbaa-4103-9970-09d1f5540c0e

https://doi.org/10.1117/12.2257660

SPIE, Bellingham, WA

9781510608214

Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2017

Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2017, 2017-03-26 → 2017-03-29, Portland, United States

Proceedings of SPIE, 0277-786X, 10168

Institutional Repository

conference paper

© 2017 Aydin Rajabzadeh, R.C. Hendriks, R. Heusdens, R.M. Groves