Search

Fiber Bragg gratings (FBG) sensor has attracted considerable attention for structural health monitoring (SHM) in aerospace aluminum alloy structures in recent years. The experimental FBG signals used to detect flaws contains noise. Therefore, in this paper, an algorithm that combines variational mode decomposition (VMD) with discrete wavelet transform (DWT) was proposed to deal with the signal denoising problem. Then, it is a challenge problem for structural damage detection to extract the features that sensitive to damage and robust to noise from the FBG signal response. This paper presents several damage features extraction algorithms, such as the multi-peaks central wavelength algorithm, full width at half Maximum (FWHM) algorithm, the spectral difference algorithm and the spectral area algorithm. Then, the denoised structure signal can be used to test the damage features extraction algorithm effectiveness and for future damage detection.

To apply Fiber Bragg Grating (FBG) sensors in the hole structural health monitoring of aircraft, this study investigates a temperature compensation algorithm based on Fourier curve-fitting, combined with an analysis of FBG layout. The results identify that the wavelengths have a linear relationship with stress and temperature and the linear correlation coefficients are all greater than 0.9. In the experiment, the temperature range was 17 °C-100 °C (step is 10 °C) and the tensile force changed from 0 N to 10 KN (step is 1 KN). With the Fourier curve-fitting of the wavelength variation, the temperature compensation can be actualized, and the average error of temperature is 0.17 °C. By integrating the algorithm in the self-developed FBG interrogation system, the real-time monitoring of stress is achieved with the average error of stress is 3.9 με.

This study investigated the reflected spectral deformation mechanism of fiber Bragg grating (FBG) sensors with crack propagation. This analysis was performed based on the simulated FBG response by applying modified-transfer matrix modeling (TMM) with the strain states, which were extracted by the finite element method (FEM) analysis. Experimental data were obtained from FBG sensors bonded in an aluminum alloy structure and subjected to multiple crack lengths, and the strain values were obtained by digital image correlation (DIC) technology. Based on the simulations and the experimental full spectral response, we compared the performance of two damage features: The full width at half maximum (FWHM) and the spectral difference. In addition, results showed that the two features were insensitive to experimental noise and were highly sensitive to the complex strain field caused by crack propagation. Moreover, the damage features changes in the crack propagation process also provided a way for crack position measurement. Ultimately, the 10 mm grating lengths sensors showed better performance to the crack detection with longer sensitivity distance. According to the research in this paper, the crack position was quantitatively determined by evaluating different damage features of the reflected spectrum.

The welded pore is one of common defects in titanium alloy electron-beam welded (EBW) joints and significantly influences its fatigue life. In this paper, effect of the welded pore on fatigue life of titanium alloy EBW joints were studied through an experimental investigation. Fatigue tests and fracture observation were conducted for EBW joints. Results showed that the welded pore is the most possible crack initiation. Through quantitative analysis on pore size and depth, a model was established to characterize the effect of initiation pore on fatigue life based on S-N model. Moreover, the effect of pore scatter on fatigue life scatter was investigated based on Monte-Carlo simulations, and the results indicated that scatter of pore size and depth is the key factor contributing to the wide scatter of fatigue life of welded joints.