More attention has been drawn to ultrasonic guided waves (UGW) based damage detection method for its advantages of wide range inspection of large scale structures. However, complex propagation characteristics of guided waves as well as traditional contact ultrasonic transducers limit its application for the practical damage detection. By combining Scanning Laser Doppler vibrometer (SLDV) technology, Time-Reversal method in frequency-wavenumber domain (f-k RTM) can compensate for the dispersive nature of Lamb waves, localize multiple damage sites and identify their sizes without time consuming numerical calculation. In this work, we adopt f-k RTM for damage detection in plate-like structure. Instead of SLDV in experiment, 3D finite element numerical method is adopted to obtain scattered ultrasonic guided wavefield data with high spatial resolution. The direct path waves were extracted to obtain the incident wavefield while the scattered signals were used to calculate the scattering wave field. Damage imaging can also be achieved by introducing crosscorrelation imaging condition. Imaging results show that the method is very effective for crack localization and boundary shape-recognition. Numerical simulation results and imaging algorithm laid the foundation for the method applied in experiment and practice.
A spectral finite element method (SFEM) is developed to analyze guided ultrasonic waves in a delaminated composite beam excited and received by a pair of surface-bonded piezoelectric wafers. The displacements of the composite beam and the piezoelectric wafer are represented by Timoshenko beam and Euler Bernoulli theory respectively. The linear piezoelectricity is used to model the electrical-mechanical coupling between the piezoelectric wafer and the beam. The coupled governing equations and the boundary conditions in time domain are obtained by using the Hamilton’s principle, and then the SFEM are formulated by transforming the coupled governing equations into frequency domain via the discrete Fourier transform. The guided waves are analyzed while the interaction of waves with delamination is also discussed. The elements needed in SFEM is far fewer than those for finite element method (FEM), which result in a much faster solution speed in this study. The high accuracy of the present SFEM is verified by comparing with the finite element results.
The spectral finite element method (SFEM) is developed to predict guided ultrasonic waves in the surface-bonded piezoelectric wafer and beam structure. The Timoshenko beam theory, the Euler-Bernoulli beam theory and linear piezoelectricity are used to model the base beam and electric-mechanical behavior of the piezoelectric wafer respectively. Using Hamilton’s principle, the governing equations are obtained in the time domain, and then the SFEM are formulated from coupled differential equations of motion transformed into the frequency domain via the discrete Fourier transform. The SFEM is used to analyze the dispersion characteristics, mode conversion of guided waves and the interaction of waves and notch. The high accuracy of the present SFEM is verified by comparing with the finite element method results.
Ultrasonic guided waves are one of the most prominent tools for SHM in plate-like structure. However, complex propagation characteristics of guided waves as well as traditional contact ultrasonic transducers limit its application in the practical damage detection. Scanning Laser Doppler vibrometer (SLDV) technology is an effective non-contact method to obtain ultrasonic guided wavefield with ultra-high spatial resolution. Based on abundant wavefield data, wavenumber imaging algorithms are capable of not only damage location, but also assessment of damage characteristics such as size and shape. In this work, we adopt local wavenumber analysis method for horizontal crack detection in platelike structure. Instead of using SLDV in experiment, 3D finite element numerical method is adopted to obtain full ultrasonic guided wavefield data. Since the horizontal cracks result in decrease of local thickness, the wavenumber in corresponding area shows significant increase, which is used as indicators for crack imaging. The effects of different damage shapes, depths and spatial window sizes on imaging are also discussed. Numerical simulation results and imaging algorithm laid the foundation for the method applied in experiment and practice.