Share Email Print

Proceedings Paper

Conventional optical fiber current measurements improved by a high accuracy artificial neural network algorithm
Author(s): Antonio C. Zimmermann; Marcio Besen; Leonardo S. Encinas; Cristiano M. B. Cordeiro
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper relies on an experiment using a conventional fiber optical current measurement system with new concept for signal processing using Artificial Neural Networks. To achieve the high accuracy results, a class 0.1% current standard was used as reference for the ANN processing. Also a large data collection was conducted to assure the high variability of the measurement process behavior, including changes in temperature and other non-linear effects. The results show 0.2% maximum error for the developed system, demonstrating the robustness and high accuracy of the proposed method, being capable of mitigate some drawbacks of FOCS system, like non-idealities compensation.

Paper Details

Date Published: 17 October 2012
PDF: 4 pages
Proc. SPIE 8421, OFS2012 22nd International Conference on Optical Fiber Sensors, 84213G (17 October 2012); doi: 10.1117/12.975616
Show Author Affiliations
Antonio C. Zimmermann, Univ. Federal de Santa Catarina (Brazil)
Marcio Besen, Univ. do Estado de Santa Catarina (Brazil)
Leonardo S. Encinas, Univ. Federal de Santa Catarina (Brazil)
Cristiano M. B. Cordeiro, UNICAMP (Brazil)

Published in SPIE Proceedings Vol. 8421:
OFS2012 22nd International Conference on Optical Fiber Sensors
Yanbiao Liao; Wei Jin; David D. Sampson; Ryozo Yamauchi; Youngjoo Chung; Kentaro Nakamura; Yunjiang Rao, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?