基于单斜坡的分布式光纤温度传感适用性研究

Study of Availability of Distributed Optical Fiber Temperature Sensing Based on Slope-assisted

传感器是配电物联网的基础,尤其是基于布里渊散射的分布式光纤传感器应用日益广泛。为了提高基于布里渊散射的分布式光纤传感的测量速度,为基于温度的电气设备状态感知奠定基础,对单斜坡法在光纤沿线温度测量的适用性进行了研究。编程实现了基于伪Voigt模型的谱拟合法和单斜坡法的计算机程序,针对不同信噪比的布里渊谱分别采用谱拟合法和单斜坡法计算光纤沿线的温度。结果表明,在一定的信噪比和光纤沿线布里渊频移波动程度范围内,单斜坡法的计算准确性可以达到与谱拟合法相近程度,但谱测量时间可以缩短为后者的几十分之一,对应的计算时间仅为谱拟合法的1/758。研究了信噪比、布里渊频移和线宽不确定度对误差的影响,结果表明,随信噪比(以dB为单位)增加,温度误差快速下降,然后逐渐趋于稳定值;随布里渊频移与工作点差距的增加,温度误差增大,且加速度随二者差距的增大而增大;温度误差随线宽偏差的增大而增大。

Sensors form the basis for the distribution Internet of Things in electricity, especially the distributed optical fiber sensors based on Brillouin scattering are widely used in this field. To improve the measurement speed of distributed optical fiber sensing based on Brillouin scattering and to lay the foundation for the condition sensing of electrical equipment based on temperature, the adaptability of the optical fiber temperature measurement method based on the slope-assisted technique is investigated. The computer programs of the spectral fitting method and the slope-assisted technique based on the pseudo-Voigt model are written. The temperature along the optical fiber for the Brillouin spectra with different signal-to-noise ratios(SNRs) are estimated by the above spectral fitting method and the slope-assisted technique. Results reveal that within a certain signal-to-noise ratio and BFS fluctuation range along the optical fiber, the accuracy of the slope-assisted technique is close to that of the spectral fitting method. However, the spectrum measurement time of the former can be shortened to a few tenth of the latter, and the corresponding computation time is only 1/758 of the latter. The influence of SNR and BFS on the error is studied. Results show that the temperature error decreases rapidly with SNR(in dB), and then it gradually tends to a stable value. The temperature error increases with the difference between the Brillouin frequency shift and the working point, and the accelerated speed also increases with the difference. The temperature error increases with the linewidth error.