Crack Damage Quantitative Detection Method Based on Dynamic Time Warping Correlation Feature

Abstract

The invention discloses a crack damage quantitative detection method based on dynamic time warping correlation features, comprising: A. building an optical fiber grating monitoring platform, distributing optical fiber grating sensors on the surface of the structure to be monitored, and recording the damage of the monitoring structure in a non-damaged state Fiber Bragg Grating Reflection Spectrum S 0 ; B. Get the smoothed reflectance spectrum S' 0 , intercept the fiber grating reflection spectrum S 0 the [l l , l r ] part as template signal X 1 ; C. Get the smoothed reflectance spectrum S' 1 , intercept S 1 [l' l , l' r ] part as real-time signal X 2 ; D, calculate the template signal X 1 Each element of the real-time signal X 2 The distance matrix C of each element in the distance matrix C; E, according to the regular path P of the distance matrix C, calculate the minimum cumulative cost matrix D, and obtain the degree of difference ξ; F, use the degree of difference ξ as a feature, and use the support vector regression algorithm to analyze the crack Quantitative calculation of length. The invention can improve the practicability and accuracy of crack damage quantitative monitoring.

Description

technical field [0001] The invention relates to a crack damage quantitative monitoring technology for flat structural materials, in particular to a crack damage quantitative detection method based on dynamic time warping correlation features. Background technique [0002] Due to the stringent requirements for material structure and equipment reliability in the aerospace field, the rapid development of structural health monitoring technology has been strongly promoted. [0003] In the field of structural health monitoring, real-time monitoring of structural cracks is a research hotspot. There are many methods in the prior art, which are realized by using different types of sensors. Among them, the fiber grating sensor is considered to be the most valuable and potential sensor because of its light weight, high precision, high sensitivity and many other advantages. In the use of FBG sensors, the characteristic parameters of the reflection spectrum of FBG sensors are usually e...

AI Technical Summary

Problems solved by technology

In addition to the phenomenon of spectral drift, there are also phenomena such as waveform broadening, and the alignment of the drifted reflection spectrum by direct translation has the problem of selecting the length of the spectral sequence, and the widening of the waveform will also lead to drastic changes in the correlation coefficient.

Therefore, using the existing correlation coefficient method to extract the features of the reflectance spectrum similarity has many complex and difficult problems.

Images