Research On Key Signal Processing Technology Of Distributed Optical Fiber Acoustic Sensing System

Distributed Optic Fiber Acoustic Sensing(DAS)system is widely used in perimeter security,oil and gas pipeline monitoring and other engineering fields due to its advantages of long-distance Sensing,high precision,high sensitivity and Sensing of related events on the Sensing Fiber.In this thesis,a series of researches are carried out on the detection optical signal regulation,data acquisition and demodulation of disturbance signal in DAS system from the point of engineering application.The main research contents are as follows:(1)A Field Programmable Gate Array(FPGA)based modulation pulse is designed according to the different requirements for detecting optical signals in different DAS applications.By controlling the pulse frequency and pulse width to control adjusting DAS system detecting light signal,thus the spatial resolution of the control system,the sensing distance,system response frequency and Rayleigh scattering light intensity,the DAS system has been greatly improved,and the performance of the system realized the spatial resolution of 1 m to 10 m controllable and below 5 k Hz frequency controlled the response of the system,So it can be applied in different engineering fields.(2)In conformity with the specific requirements of DAS system for data acquisition,the construction and implementation of the sensor signal data acquisition system are completed.Made of FPGA to generate modulation pulse external trigger source,DAS do synchronous processing system data acquisition system,set up the corresponding collection model and parameters,using FIFO Multi acquisition mode and ABA(A low speed,B high-speed)sampling mode mixing,the system output signal cycle collection,realized the function of the DAS system real-time acquisition.(3)The collected data are demodulated to extract Rayleigh scattering curve and the time domain curve of the disturbance signal(which is simulated by FPGA on PZT).After the moving average pretreatment,the Short Time Fourier Transform,STFT,wavelet transform,FAST Independent Component Algorithm(FAST ICA)and fractional domain spectrum subtraction are used to demodulate and restore the disturbance signals on the sensing fiber.(4)Hilbert Huang Transform(HHT)and its two improved algorithms are applied to DAS sensor signal demodulation.The two improved algorithms are Ensemble Empiricial Mode Decomposition-Hilbert Huang Transform,EEMD-HHT)and Complete Ensemble Empirical Mode Decomposition-Hilbert Huang Transform(CEEMD-HHT).By comparing the demodulation results,the advantages of the improved algorithm in DAS disturbance sensor signal demodulation are obtained,which not only achieves more intuitive time-frequency analysis,but also improves the Signal-Noise Ratio(SNR)of the modulated signal.Among them,CEEMD-HHT algorithm was used to demodulate 1k Hz disturbance signal,and its SNR reached7.33 d B.Compared with FAST ICA,fractional-order domain spectral subtraction,wavelet transform,HHT,EEMD-HHT,SNR respectively increased by 4.15 d B,2.58 d B,2.26 d B,2.15 d B and 0.94 d B.Finally,in order to solve the problem that traditional methods cannot demodulate the actual audio disturbance signal,EEMD and CEEMD-HHT are combined to demodulate the audio signal acting on the sensing fiber(the audio signal is played by the speaker).