Nondestructive detection of CFRP subsurface defects using transient lock-in thermography

In this paper, transient lock-in (TLI) thermography is used as a nondestructive inspection technique to detect the subsurface defects in carbon fiber reinforced polymer (CFRP) composite. Initially, the transient thermal response is obtained by removing the DC term from the raw LI thermal image sequences using pixel-wise polynomial fits. Subsequently, the phase and amplitude images at the excitation frequency are extracted from the transient thermal response by the discrete Fourier transformation. Meanwhile, the two CFRP specimens with different sizes and depths of artificial flat-bottomed holes (FBHs) are used and the corresponding TLI tests are also carried out. Finally, to quantitatively demonstrate the superiority of the TLI thermography to inspect the CFRP subsurface defect, a comparative analysis of the TLI thermography and advanced defect detection methods, including sparse principal component thermography (SPCT), fast independent component thermography (FICT), and improved robust tensor PCT (IRTPCT), is also conducted; moreover, the signal-to-noise ratio (SNR), detection rate, and runtime are also given in detail. Results show that the TLI_phase image is effectively capable of eliminating uneven heating and gives a satisfactory SNR, highest DR, and lowest runtime overall. Therefore, transient lock-in thermography can be used as a rapid and robust NDT E approach for inspecting CFRP subsurface defects.