Detection and quantification of artificial delaminations in CFRP composites using ultrasonic thermography

Delamination is inevitable during either fabrication or service of a carbon fiber reinforced polymer (CFRP) component. These flaws are major degradation phenomena that are difficult to be detected by nondestructive testing techniques, especially for closed delamination. This study performed an experimental investigation on delamination detection in CFRP plates using ultrasonic thermography with an ultrasonic transducer. In this work, a CFRP composite plate with delaminations was manufactured by a compression molding process. The ultrasonic was excited in the specimen and the heat was activated at the delaminations. Temperature distribution on the specimen surface was recorded by an infrared camera. The effects of wave propagation distance, delamination size, and depth were analyzed. Furthermore, the three most established signal processing techniques of thermography, namely thermal signal reconstruction (TSR), principal component analysis (PCA), and fast Fourier transform (FFT) were employed to deal with the subtracted-background thermal data. The performance of each algorithm was evaluated by temperature contrast, and global signal-to-noise ratio (SNR). Finally, the delamination area was quantitatively measured by Otsu’s thresholding method. The results indicate that the first and second principal components (PC1 and PC2) of PCA provide the highest SNR and the largest number of delaminations detected, respectively. The delamination of ϕ5 and a depth of 1.5 mm can be detected with excellent image quality. Hence, ultrasonic thermography can provide an efficient and cost-effective thermographic method for delamination detection in CFRP structures.