多模光纤
光纤
光时域反射计
材料科学
降级(电信)
衰减
光纤传感器
单模光纤
拉曼散射
信号(编程语言)
光学
声学
拉曼光谱
计算机科学
物理
光纤分路器
电信
程序设计语言
作者
Gerald Kelechi Ekechukwu,Jyotsna Sharma
标识
DOI:10.1109/jsen.2023.3257264
摘要
Distributed fiber optic sensors (DFOSs) are viewed as a promising technology for efficient, cost-effective, and nonintrusive real-time monitoring for a variety of applications. The chemical passivity, nonmagnetic interference behavior, and ability to acquire continuous measurements at high spatial resolution have made DFOS very attractive for in situ measurement of temperature, strain, vibration, and so on. The accuracy of the measured parameter by the DFOS is highly dependent on the strength and quality of the optical signal returning from the point of interest. The signal degradation in DFOS results from optical losses due to attenuation, absorption, scattering, bending, dispersion, and coupling. This study analyzes the degradation observed in both the single-mode and multimode (MM) fibers installed in a 1574-m deep wellbore for acquiring distributed acoustic sensor (DAS) and distributed temperature sensor (DTS) data, respectively. The degradation was unexpected as it was observed within three years of DFOS installation in an experimental, nonproducing well that contains noncorrosive fluids and operates in a low-temperature ( $< 54.4 ^{\circ }\text{C}$ ) and low-pressure (< 3500 psi or 24 131 kPa) environment. A root cause analysis was performed to investigate the cause of this degradation using time-lapse optical loss measurements acquired at different wavelengths using an Optical time domain reflectometer (OTDR). Degradation of the multimode fiber was also examined by the time-lapse analysis of the Raman backscatter that is used for DTS measurement. Based on the investigation, microbending was found to be the main cause of the observed degradation. Its impact on the quality of the DAS and DTS measurements was also analyzed both qualitatively and quantitively. Recommendations for minimizing degradation are presented.
科研通智能强力驱动
Strongly Powered by AbleSci AI