Electromagnetic Thermal Nondestructive Testing: Multiphysics mechanism, sensor structure and signal processing

Since 1960s, thermal testing has been successfully explored in NDT&E applications to measure the surface temperature variations in response to induced energy. Electromagnetic thermography (EMT) which combines eddy current (EC), magnetic and thermography, and involves the application for a short period of a high current electromagnetic pulse to the conductive material under inspection. In comparison with other thermography NDT&E techniques, the heat in EMT is not limited to the sample surface, rather it can reach a certain depth, which governed by the skin depth of eddy current.

Furthermore, electromagnetic thermography focuses the heat on the defect due to friction or eddy current distortion, and subsequently increase the temperature contrast between the defective region and defect-free areas. From adaptability in terms of defect orientation, electromagnetic thermography can enhance specific excitation direction to optimize the directional evaluation along the defect orientation which is more effective for geometrically complex components and showed more crack indication.

In electromagnetic thermography, the electromagnetic mechanism of Joule heating and thermal conduction on conductive material characterization broadens their scope for NDT&E by imparting sensitivity, conformability and allowing fast and imaging detection, which is necessary for efficiency. Methodologically, various EM-thermal NDT techniques like eddy current pulse thermography (ECPT), eddy current step thermography (ECST), eddy current pulse phase thermography (ECPPT), and pulsed inductive thermal wave radar (PITWR) are investigated. Time-domain, frequency-domain, time-frequency domain, statistical domain and logarithm-domain defect evaluation methods are described and analyzed. New sensing structure of a fusion of different physical phenomena for enhancing NDT sensitivity is investigated and the fusion progress includes 1) induced eddy current generates Joule heating, 2) alternating magnetization/ demagnetization produce hysteresis loss for heating, and 3) leakage magnetic flux with stray loss.

The constructs a physical time-dependent partition model to analyze the whole thermal transient process and considers characteristic times for separating Joule heating and thermal diffusion into four different stages for detectability discussion. Notwithstanding above, we bridge the gap between the physics world and mathematical modeling world. We generate physics-mathematical modeling and mining route in the spatial-, time-, frequency-, and sparse-pattern domains.

This is a significant step towards realizing the deeper insight in electromagnetic thermography and automatic defect identification. This renders the electromagnetic thermography a promising candidate for the highly efficient and yet flexible NDT&E technique. Advanced algorithms, such as principal components analysis (PCA), independent components analysis (ICA), Nonnegative matrix factorization, and multi-dimensional tensor decomposition are used which allows the detection is fully automated and does not require manual selection from the user of the specific thermal frame images. Image reconstruction, segmentation and enhancement is used to improve the detectability.

In view of applications, material properties variations including conductivity, permeability, and lift-off are evaluated; experimental studies for real damages including corrosion in steel, stress in aluminium, impact and delamination in carbon fiber reinforced polymer (CFRP) laminates, RCF cracks in rail, are abundant.