基于SVD的BOTDA测温系统的解调算法研究

Research on Demodulation Algorithm of BOTDA Temperature Measurement System Based on SVD

基于布里渊光时域分析(Brillouin Optical Time Domain Analysis, BOTDA)的分布式光纤传感器具有测量范围远、精度高、可测温和测应变等优势,目前在大范围、长距离传感领域中广泛应用,如桥梁、建筑、道路等。然而目前大部分布里渊散射信号都是运用洛伦兹拟合算法进行寻峰解调,解调精度受噪声影响较大。提出一种基于奇异值分解(Singular Value Decomposition, SVD)的降噪寻峰算法,利用Hankel矩阵对测温信号进行升维,对BOTDA测温系统的仿真信号进行分解,并通过实验验证了利用SVD对测温信号降噪来提升测温精度的可行性。

Distributed optical fiber sensors based on Brillouin optical time domain analysis(BOTDA)have the advantages of long measurement range,high precision,capable of temperature measurement and strain measure-ment,etc.,are widely used in large-range,long-distance sensing fields,such as bridges,buildings,roads,etc.However,most of the current Brillouin scattering signals are using Lorentz fitting algorithm for peak finding de-modulation,and the demodulation accuracy is greatly affected by noise.A noise reduction peaking algorithm based on singular value decomposition(SVD)is proposed,and the temperature measurement signal is raised by using the Hankel matrix,and the simulation signal of the BOTDA temperature measurement system is de-composed,and the feasibility of using SVD to reduse the noise of the temperature measurement signal to im-prove the temperature measurement accuracy is verified through experiments.