Research On The Fiber Grating Based High-accuracy Temperature/Strain Sensing System

Fiber Bragg grating (FBG) sensing is an emerging sensing technology since 1970s, and this technology has been widely used in civil and military fields, for example, construction industry, power industry, aerospace industry, scientific research and so on. The sensing function of FBG is achieved by the change of the reflection wavelength with the change of the external physical quantity, so the accuracy of wavelength detection is an important index for FBG sensing system. The wavelength demodulation technology of FBG is researched in this thesis, the factors that affect the accuracy of wavelength demodulation are analyzed, and the methods for improving the accuracy are given.Firstly, this thesis researches the FBG coupled mode theory; and the theories of temperature and strain sensing of FBG, as well as the issue of temperature-strain cross-sensitivity; through comparing the advantages and disadvantages of several common-used wavelength demodulation technologies, a four-channel FBG wavelength demodulation system based on Fabry-Perot (F-P) tunable filter is designed. Secondly, this thesis studies the influence of cavity length drift of F-P filter and nonlinearity of scanning voltage on measurement accuracy, and the reference FBG and F-P etalon are applied to calibrate the wavelength; then from the aspect of fiber dispersion and transmission cable length in actual project, the impact of transmission delay on system calibration is discussed, and the influence level is analyzed quantitatively via numerical simulation. Thirdly, this thesis analyzes the effect of the finite-impulse-response (FIR) and infinite-impulse-response (IIR) filter when using them to filter the spectrum signal, then the filter type and parameters, which are used in the wavelength demodulation system, are determined; in order to obtain high detection accuracy and stability, four peak searching algorithms have been analyzed and compared, including differential method, centroid method, cross-correlation method, and Gaussian fitting method, then the peak searching algorithm and its parameter settings used in the wavelength demodulation system is determined. Finally, according to the above study, this thesis realizes the FIR low-pass filtering and Gaussian fitting algorithm by LabWindows/CVI, and the software of four-channel FBG wavelength synchronous demodulation system is completed.