UV-Induced Intrinsic Fabry-Perot Interferometric Fiber Sensors and Their Multiplexing for Quasi-Distributed Temperature and Strain Sensing

Abstract

Distributed temperature and strain sensing is demanded for a wide range of applications including real-time monitoring of industrial processes, health monitoring of civil infrastructures, etc. Optical fiber distributed sensors have attracted tremendous research interests in the past decade to meet the requirements of such applications.

This research presents a multiplexed sensor array for distributed temperature and strain sensing that can multiplex a large number of UV-induced sensors along a single fiber. The objective of this research is to develop a quasi-distributed sensing technology that will greatly increase the multiplexing capacity of a sensor network and can measure temperature and strain with a high accuracy and high resolution.

UV-induced intrinsic Fabry-Perot interferometric (IFPI) optical fiber sensors, which have low reflectance and low power loss, are good candidates for multiplexed sensors networks. Partial reflectors are constructed by irradiating photosensitive fiber with a UV laser beam. A pair of reflectors will form a Fabry-Perot interferometer that can be used for temperature and strain sensing. A sensor fabrication system based on a pulsed excimer laser and a shadow mask is developed.

A spectrum-based measurement system is presented to measure the interference fringes of IFPI sensors. A swept coherent light source is used as the light source. The spectral responses of the IFPI sensors at different wavelengths are measured. A frequency division multiplexing (FDM) scheme is proposed. Multiple sensors with different optical path differences (OPD) have different sub-carrier frequencies in the measured spectrum of the IFPI sensors. The multiplexing capacity of the sensor system and the crosstalk between sensors are analyzed.

Frequency estimation based digital signal processing algorithms are developed to determine the absolute OPDs of the IFPI sensors. Digital filters are used to select individual frequency components and filter out the noise. The frequency and phase of the filtered signal are estimated by means of peak finding and phase linear regression methods. The performance of the signal processing algorithms is analyzed.

Experimental results for temperature and strain measurement are demonstrated. The discrimination of the temperature and strain cross sensitivity is investigated. Experimental results show that UV-induced IFPI sensors in a FDM scheme have good measurement accuracy for temperature and strain sensing and potentially have a large multiplexing capacity.