Research On Extraction Methods And Influencing Factors Of Characteristic Parameters Of Fiber Brillouin Scattering Spectrum

Fiber Brillouin distributed sensing technology can realize the measurement of temperature and strain information at different positions of the fiber by extracting the characteristic parameters of Brillouin scattering spectrum and using their relationship with temperature and strain.Therefore,the measurement accuracy of temperature and strain is closely related to the extraction accuracy of characteristic parameters of Brillouin scattering spectrum.At present,the extraction of characteristic parameters of Brillouin scattering spectrum is mainly realized by parameter fitting,pattern matching and optimization algorithm.The accuracy obtained by using these methods has a certain relationship with the performance of the method and different influencing factors.In order to use different Brillouin spectrum parameter extraction methods flexibly and achieve high precision parameters extraction,it is necessary to research on them comprehensively.Therefore,the characteristic parameters extraction methods and influencing factors are studied in this paper.The main contents are as follows:(1)The principle of optical fiber Brillouin scattering is studied.The simulation model of heterogeneous seven-core fiber is established,and the optical modes in different cores are solved by using finite element method.The characteristic parameters of Brillouin scattering spectrum of each core at different wavelengths are calculated.The results show that at the same wavelength,due to the different core structures in the heterogeneous seven-core fiber,the Brillouin frequency shift in the middle core is different from that in the outer core,and the linewidth and gain value at the peak are almost the same.(2)Based on artificial fish school algorithm(AFSA)and Levenberg-Marquardt(LM)algorithm,a characteristic extraction algorithm(AFSA-LM)of Brillouin scattering spectrum is proposed.AFSA algorithm and AFSA-LM algorithm are used to extract the characteristic parameters of Brillouin scattering spectrum of different cores in heterogeneous seven-core fiber respectively.The results show that the AFSA-LM algorithm can achieve high accuracy extraction of Brillouin frequency shift.There will be certain error in the extraction of Brillouin linewidth by using the two algorithms.(3)The principle of Brillouin frequency shift extraction based on similarity matching algorithm(SMM)is studied,the effectiveness and real-time advantage of this method are verified.The influence of the linewidth,signal to noise ratio,sweep step and sweep range on the accuracy of Brillouin frequency shift extraction is systematically studied under the condition that single factor changes.The frequency shift error is linear with the Brillouin linewidth.The difference of the linewidth between the selected detecting spectrum and selected reference spectrum has little impact on the accuracy of the algorithm.When the signal to noise ratio of the selected detecting spectrum remains constant,the frequency shift error decreases exponentially with the increase of the signal to noise ratio of the selected reference spectrum,and vice versa.When the frequency sweep step is constant,the frequency shift error increases with the increase of the selected reference spectrum sweep range,and the calculation time linearly increases.The frequency shift error is the smallest when the frequency range of the selected detecting spectrum is 2 times the linewidth.