A reliability evaluation of linear frequency modulated square wave thermography using low peak power laser

This study presents a reliability assessment of Digital Frequency Modulated Thermal Wave Imaging (DFMTWI) for detecting subsurface defects in carbon fiber reinforced polymer (CFRP) composites. Three CFRP specimens, containing 27 artificial flat-bottom hole defects with varying diameter-to-depth ratios to simulate delamination, were inspected using a low-power laser excitation system. To enhance defect recognition accuracy, four post-processing algorithms, fast Fourier transform (FFT), fractional Fourier transform (FrFT), principal component analysis (PCA) and fitting correlation coefficient (FCC), were comparatively evaluated. Crucially, the detectability of each algorithm was quantitatively assessed using the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), while the overall inspection reliability was evaluated through Probability of Detection (POD) using a hit/miss method. Experimental results demonstrate that while all four algorithms successfully identify defects with large diameter-to-depth ratios, the FrFT algorithm significantly enhances the SNR and CNR for deep-seated, low-contrast defects. Furthermore, the POD assessment reveals that the highest reliability is achieved through the specific combination of the DFMTWI and the FrFT algorithm, as FrFT optimally matches the time-frequency characteristics of the modulated thermal wave. This provides a reliable technical solution for comprehensive defect evaluation in CFRP materials under low peak power conditions.