Research On Strain Measurement Error Analysis And Correction About Fiber Extrinsic Fabry-Perot Interferometer

Optical fiber sensors have been widely concerned applied in the field of temperature and strain sensing because of its small size,anti-electromagnetic interference and high sensitivity.It is difficult to measure the strain in harsh environments such as high temperature,high pressure and high speed rotation.The optical fiber sensor can be applied to the measurement of these harsh environments according to its advantages.However,the optical fiber sensor working in high temperature and harsh environment will be disturbed seriously in the actual measurement process,resulting in errors,including noise in the transmission process of optical path signal,demodulation error and measurement random error.Therefore,it is of great significance to improve the measurement accuracy of optical fiber sensors by analyzing the error sources and correcting the error from each source that may produce the measurement strain error.In this paper,aiming at the error generated in the measurement process of optical fiber sensor,the spectral signal is firstly reconstructed and denoised,and then demodulated.Finally,the measurement error of the overall strain is corrected.The main research contents are as follows:(1)Strain Measurement Experiment and Error Source Analysis of Extrinsic Fabry-Perot Interferometer(EFPI).Research on the principle,structure,material and packaging method of optical fiber EFPI strain sensor.The mathematical expression of interference spectrum of optical fiber EFPI strain sensor is deduced.The measurement experiment of fiber optic EFPI strain sensor is designed and executed,including high temperature thermal strain measurement,room temperature mechanical strain measurement and simultaneous measurement at high temperature.The sources of possible errors under different measurement conditions are analyzed,and different types of errors are analyzed.(2)Research on denoising and demodulation methods of fiber EFPI strain sensor signal.Aiming at the spectral signal at high temperature,the variational mode decomposition(VMD)method is used to decompose the spectral signal,and the effective components after decomposition are extracted to reconstruct the signal to achieve the denoising effect.The interference spectrum signal of fiber EFPI strain sensor is demodulated by using the method of Fast Fourier Transform(FFT)combined with frequency domain interpolation.The minimum mean-square error(MMSE)estimation is added to further improve the demodulation accuracy.Finally,the accuracy of the demodulation method is verified by simulation and experiment.The results show that the proposed method has higher demodulation accuracy than other demodulation methods.(3)Research on measurement error correction method.Aiming at the random error produced in the measurement process of optical fiber EFPI strain sensor,a correction process suitable for the measurement error of the sensor is proposed.Adaptive-Network-based Fuzzy Inference System(ANFIS)is used to build the mapping relationship between the measured strain value and the reference strain value,so as to realize the correction of the measured strain.The network parameters are optimized to verify the accuracy of the network,so as to provide a more accurate error correction network.Finally,the actual measurement data are substituted into the error correction network to realize the correction of the measurement error of the fiber EFPI strain sensor.