Abstract
A balloon-type optical fiber sensor (OFS) for strain measurement was proposed, and the measurement range was significantly increased with the assistance of convolutional neural networks (CNN). The presented OFS was prepared by bending a Mach-Zehnder structure based on polarization-maintaining fibers into a balloon-type, which exhibits a sensitivity of 26.3 pm/με under the strain range from 0 to 300 με. The CNN-based signal analysis method can effectively extract the data information in the spectral, and the strain range that can be measured is greatly enlarged to 0–2000 με, while the accuracy of the measurement can be guaranteed. Further, the temperature sensitivity of the OFS is −0.0252 nm/°C; the lower temperature sensitivity can effectively suppress the crosstalk problem caused by temperature changes during strain measurement. The proposed method remarkably increases the strain measurement range of OFS, which is conducive to the application and development of OFSs.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call