Fiber bragg grating sensing network distortion spectrum demodulation method and system based on differential evolution improved extreme learning machine

Abstract

The invention provides a fiber bragg grating sensing network distortion spectrum demodulation method and system based on a differential evolution improved extreme learning machine; the method specifically comprises the following steps: converting a wavelength demodulation problem of a distortion spectrum into a nonlinear regression problem, and building a corresponding regression model; learning a distortion spectrum sample by utilizing an extreme learning machine improved based on differential evolution to obtain a trained model, then performing quick demudulation on the actual sampling spectrum by means of the trained model and identifying the corresponding distortion parameters. According to the invention, high-precision demodulation of a normal spectrum can be ensured, and adaptive demodulation of a distorted spectrum can be carried out to the greatest extent.

Description

technical field [0001] The invention relates to the field of fiber grating technology, in particular to a method and system for demodulating a distorted spectrum of a fiber grating sensor network based on a differential evolution improved extreme learning machine. Background technique [0002] As a new generation of sensors, Fiber Bragg Grating (FBG) is widely used due to its unique advantages of high sensitivity, low loss, low cost, anti-electromagnetic interference, anti-corrosion, high temperature resistance, and easy multiplexing and networking. In the sensing monitoring of various engineering fields such as power industry, civil engineering, petrochemical industry, etc., it has great application value especially in the field of structural health monitoring. Because the size and weight of the FBG sensor are very small and have strong waterproof performance, it is easy to embed inside the material or structure. By building a distributed fiber grating sensor network, the s...

AI Technical Summary

Problems solved by technology

[0003] Due to the unique advantages of FBG sensing technology, the working environment of FBG sensing network in practical applications is very harsh. Although FBG sensors are packaged and protected, long-term information monitoring will still cause them to be affected by factors such as ambient temperature, humidity, The impact of external factors such as chemical corrosion

Under the continuous action of the application environment, FBG sensors will inevitably experience varying degrees of performance degradation, causing distortion in the reflection spectrum of FBG, resulting in deviations in measurement results and reducing the reliability of the detection system.

In large-scale structural health monitoring systems, FBG sensors are usually embedded in the structure on a large scale in the form of network multiplexing. If only a single FBG sensor is replaced, it will not only destroy the overall network structure, but even need to replace the entire optical cable. It will inevitably cause huge economic losses. Therefore, for large-scale distributed FBG sensor networks, finding an effective FBG distortion spectrum demodulation method is a key technology to improve the reliability of the entire FBG sensor network.

[0004] Typical characteristics of FBG distorted spectra include broadening of the main peak of the spectrum, left-right asymmetry of the waveform of the spectrum, weakening of the light intensity of the spectrum, and increase of the peak of the side lobe of the spectrum. Crosstalk phenomenon, so that the sensing ability of FBG is greatly reduced

Images