Pulsed Phase Thermography Inversion Procedure using Normalized Parameters to Account for Defect Size Variations

摘要

Pulsed Phase Thermography (PPT) is rapidly evolving as a solid NDT&E technique. Acquisition is accomplished in a similar way as in classical Pulsed Thermography, thermal data is processed afterward using a transformation algorithm, e.g. the Fourier Transform (FT), providing amplitude and phase delay data. The authors have recently presented an extended review on PPT theory, as well as a new inversion technique for depth retrieval using phase. Furthermore, an automatic defect depth retrieval algorithm had also been presented. Due to the Time-Frequency Duality of the discrete FT, PPT sampling and truncation parameters should be carefully selected to produce the desired frequency response. An interactive methodology for the optimal selection of these parameters has been proposed. Nevertheless, this is not always a simple task. On one hand, there exists stored data for which sampling and truncation was performed without considering the time-frequency relationship; and on the other hand, there is not always possible to produce the desired frequency output because of equipment limitations. In this paper, two situations are considered. First, two composites plates (CFRP and GFRP), for which adequate parameters have been used. In this case, we demonstrate that depth can be directly estimated from the diffusion length equation as is done by Lock-In Thermography. Secondly, an aluminum specimen that has been incorrectly sampled is considered. In this case, we propose the normalized diffusion length μ_n, and the normalized diameter D_N, to account for defect size variation.