Diffusion-compensated correlation analysis of frequency-modulated thermal signal for quantitative infrared thermography

Application of a broadband modulated excitation signal and its cross-correlation with the responding thermal response, so-called thermal wave radar or pulse compression thermography, is a well-known technique in active infrared thermography. The technique benefits from both broadband spectrum and the pulse compression efficiency of excitation signal, providing lag phase quantities which can be used for characterization of defects. However, the significant distortion of the thermal response due to heat diffusion makes the cross-correlation analysis inefficient, particularly for the detection of deep defects or the inspection of a material with thermal diffusivity. To tackle this issue, diffusion-compensated correlation analysis (DCCA) thermal signal is proposed for quantitative infrared thermography, using a frequency-modulated sweep signal as excitation waveform. DCCA is based on the correlation of the measured thermal response with a library of template thermal responses calculated using the 1D analytical solution of the heat diffusion problem. DCCA can then more reliably resolve the thermal response presence of measurement noise, and depending on the definition of the library, directly map the corresponding depth or diffusivity. Two inspection modes are studied: (i) transmission of from defects which act as sub-surface heat sources (e.g. vibrothermography), and (ii) reflection an externally applied heat flux from sub-surface defects (e.g. optical infrared thermography reflection mode). Outperformance of DCCA to thermal wave radar is analytically substantiated and verified by finite element simulation on a carbon fibre reinforced polymer plate. Further-more, the technique is experimentally studied for thermographic inspection of a carbon reinforced polymer (CFRP) coupon including artificial defects, and the advantages as well limitations of the technique are discussed.

Automated summary

The technique of thermal wave radar or pulse compression thermography, which utilizes a broadband modulated excitation signal and its cross-correlation with the thermal response, is widely used in active infrared thermography for defect characterization. However, the distortion of the thermal response due to heat diffusion affects the efficiency of cross-correlation analysis, especially for deep defects or materials with different thermal diffusivity. To overcome this issue, diffusion-compensated correlation analysis (DCCA) thermal signal is proposed, using a frequency-modulated sweep signal as an excitation waveform. DCCA can accurately analyze the thermal response in the presence of measurement noise, and can directly map the corresponding depth or diffusivity based on a library of template thermal responses. The performance of DCCA is analytically substantiated and verified through simulations and experiments on carbon fiber reinforced polymer plates, showing its superiority over thermal wave radar. The technique has potential for thermographic inspection of materials with artificial defects.