A proper lifting scheme wavelet transform for vibration-based damage identification in composite laminates

Abstract

Damage detection using the wavelet transform was investigated and appropriate approaches to raising the method’s sensitivity level were proposed. In addition, the current study attempted to implement the impulse wavelet design algorithm in order to present appropriate wavelet function with respect to the characteristics of the signal. The initial wavelet function corresponding to the impulse response of composite plate was achieved using impulse wavelet algorithm in time domain. The function was optimized using lifting scheme method. To detect damages, an appropriate signal was selected through applying wavelet transform. To enhance damage identification, first, the edges’ effect of wavelet transform was removed, then a higher accuracy was observed by summing the wavelet coefficients in all scale factors for each mode shape and the wavelet coefficients for all mode shapes. The article also presents a quantitative measure to compare different wavelets.

Identification of mechanical and damage parameters of composite laminates based on a CPAM method

“…For mode shapes, following relation can be used to simulate the contaminated modes where trueψ^r is the r th contaminated mode shapes and λ represents the noise level and randn is the random Gaussian number generator function in MATLAB. 17 Vibrational data polluted with disturbances (by importing Gaussian disorder to the experimental data) are employed to peruse how much the mechanical and damage parameters can be deviated. Different levels of disorder (1, 5, and 10%) are imported to the vibrational data of the specimens C4, C5 and C6 and the obtained results for them are presented in Tables 9to 14 by considering just the four unknown mechanical properties and in Tables 15 to 20 by considering the five ones.…”

“…The frequency's upper limit for this system is 25 kHz and for omitting the aliasing noise, an internal low pass filter is utilized. More detail about modal experiment procedure of these plates can be followed in Ashory et al 11,12,14 and Ghasemi-Ghalebahman et al 17 As can be seen from Figure 4, these plates were hanged by two strings. Because natural frequencies obtained experimentally, are independent of position and rigidities of these strings, boundary conditions of these plates are considered completely free.…”

“…18 Also, pattern of 12 Â 12 is considered for C6 to match the mesh size of this plate with its mesh size in Sahoo et al 18 Dimensions, density and fiber orientation of these plates are shown in Table 1 and their damage parameters are represented in Table 2. Also, the mechanical constants of them that were obtained in the previous studies 14,17,18 are shown in Table 3. Moreover, their obtained natural frequencies are presented in Table 4.…”

Identification of Damage in Steel‒Concrete Composite Beams Based on Wavelet Analysis and Deep Learning

“…Researchers can use it to perform multiscale joint analysis of data in the time domain and frequency [8]. Moreover, wavelet analysis accurately has revealed the location and extent of localized damage in plates and beams [9][10][11]. However, wavelet analysis for damage identification still requires manual the manual selection of features of the vibration signal, which does affect the accuracy of the identification.…”