Finite element analysis of vibro-thermography nondestructive testing technique via local defect resonance

Vibro-thermography is an emerging and promising technique that uses ultrasonic elastic waves as an excitation source to detect and evaluate surface and subsurface defects. Friction of the edges of defects, viscoelastic behavior and non-linear vibrations of the defect region are the main sources of heating and the temperature gradient that shows up synchronously with variations of non-linear elastic energy. The temperature gradient in the defect region can be imaged by an infrared camera in order to estimate the location and size of the defects. In this paper, the vibro-thermography is simulated in COMSOL Multiphysics software. Lamb waves are used to excite an aluminum plate containing a flat-bottomed hole. First, the resonance frequency of the defect is found by means of the theory of vibrations and also by finite element method (FEM). An algorithm that incorporates frequency analysis as a function of out-of-plane displacements is used to verify this frequency and the results are compared with the eigenfrequency analysis results. The agreement observed between the theoretical and numerical models is found to be very good. The plate is then excited by an amplitude modulated sine-burst at the local defect resonance (LDR) frequency and a frequency related to the thermal penetration depth. Thermal image processing is carried out on the thermal waves to obtain their amplitude and phase images. By considering a four-point algorithm, the location, size and geometry of the defect is estimated with good accuracy.