Paper
6 June 1997 Condition monitoring of engineering structures with an optical fiber vibration sensor system
Author Affiliations +
Abstract
This paper presents preliminary work on a vibration monitoring system for assessing the condition of engineering structures or materials. It consists of an intensity-based fiber-optic vibration sensor, a fast Fourier transform pre- processing stage and a back-propagation neural network. The response of the vibration sensor to sinusoidal acceleration is compared with that of a piezoelectric accelerometer. The sensor was fixed to carbon-fiber composite panels and its response to acoustic transients was investigated. Signals from breaking a pencil lead or dropping a ball bearing on the panel were compared. The system was then trained to distinguish between the sensor response to dropping a ball bearing on panels with either real or simulated impact damage. The trained network then identified previously unseen examples of these signals with complete accuracy.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Crispin T. M. Doyle and Gerard Franklyn Fernando "Condition monitoring of engineering structures with an optical fiber vibration sensor system", Proc. SPIE 3042, Smart Structures and Materials 1997: Smart Sensing, Processing, and Instrumentation, (6 June 1997); https://doi.org/10.1117/12.275750
Lens.org Logo
CITATIONS
Cited by 1 scholarly publication.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Sensors

Optical fibers

Neurons

Neural networks

Data acquisition

Optical amplifiers

Composites

Back to Top