Research On Localization Algorithm Of Acoustic Emission Signal Based On Fiber Grating Sensor

Acoustic emission technology is a kind of non-destructive testing technology,it can effectively detect the transient elastic wave emitted by various damage forms in the material,and predict the possible defects and damages,so as to ensure the safe operation of the equipment.Therefore,acoustic emission has a broad development trend in online fault detection.Acoustic emission source location technology is a key technology to find cracks and damage sources.Its core is to infer the location of the acoustic emission source by receiving signals.Starting from the basic theory of acoustic emission detection,this thesis studies the detection principle of fiber Bragg grating acoustic emission strain sensor,the selection of important parameters of Variational Mode Decomposition(VMD)algorithm and the fiber Bragg grating acoustic emission positioning based on VMD?GCC time delay estimation algorithm.Based on the principle of fiber grating and acoustic emission stress wave,an experimental platform of fiber grating acoustic emission positioning demodulated by narrow-band light source is constructed.Aiming at the problems of common delay estimation algorithms such as low accuracy and easy to be interfered by noise,a delay estimation algorithm based on VMD?GCC is proposed.The algorithm is based on the generalized weighted cross-correlation(GCC)algorithm and combined with the variational modal decomposition(VMD)to use characteristic frequency bands to extract the time difference,which effectively improves the signal-to-noise ratio and the accuracy of the algorithm's time difference extraction.The key to variational modal decomposition lies in the determination of decomposition scale K and penalty factor (?) parameters.This study uses waveform analysis to analyze the IMF components of different decomposition scales of VMD,and combines the IMF components under different K values with the original signal to determine the mean value of the correlation coefficient.Decompose the value of the scale K.Experiments show that the larger the penalty factor,the narrower the bandwidth of the IMF component,the better the effect,but the (?) size will affect the convergence speed of the algorithm.The speed of this thesis considers the effect of modal decomposition and the speed of algorithm convergence to determine the value of the penalty factor.Experiments show that,compared with the generalized cross-correlation time delay estimation algorithm,the second cross-correlation algorithm,and the LMS adaptive time delay estimation algorithm,the time delay estimation algorithm based on VMD?GCC can achieve better results for the acoustic emission signal of the fiber grating sensor.The processing effect can achieve good estimation accuracy even under the condition of low signal-to-noise ratio.The positioning algorithm error is greatly reduced compared with traditional algorithms.Under the condition of low signal-to-noise ratio,the correlation result has a single peak and high amplitude,and the positioning is accurate.The rate has increased significantly.In the fiber grating sensor acoustic emission source positioning experiment,the linear error is between 0.00355m-0.04105 m,the average error is 0.01477 m,and the error percentage is3.69125%.The maximum error of the results obtained in the planar positioning experiment is within 0.0422 m,absolute The average error is 0.01433 m,which meets the requirements of acoustic emission source positioning in engineering applications.