• Project title: Ultra-sound imaging from active Infrared thermography for non-destructive testing of lightweight materials
• Acronym: USBIRT3
• Duration: 36 months (start: 01.10.2025)
• Funding: Austrian Research Promotion Agency FFG

Active thermography uses the illumination of a workpiece, e.g. with a flash lamp or a laser pulse, to detect defects below the surface. An infrared camera is used to measure the temperature development on the surface after illumination without contact.From the measured temperature development, the position, size or shape of the defects are reconstructed. For non-destructive testing (NDT) of lightweight materials, e.g. carbon fiber-epoxy composites (CFRP) for aircraft construction, the much faster active thermography is used in addition to ultrasound with water coupling. This project combines the advantages of thermographic measurement and ultrasonic reconstruction.

Thermography works well for defects whose depth below the surface is about the same as the diameter of the defect. In order to be able to reliably detect defects at somewhat greater depths, we have calculated a virtual ultrasonic wave from the measured thermal signal (Virtual Wave Concept - VWC), from which the position, size and shape of the defects can be reconstructed using ultrasonic reconstruction algorithms or deep learning algorithms. Common thermographic reconstructions usually work only in one dimension (1D) perpendicular to the surface (axial). The VWC also considers heat flows parallel to the surface (lateral - 2D or 3D). As a result, the same defect point inside the sample is detected from many different directions by many pixels of the infrared camera on the surface, which acts as a kind of averaging in the reconstruction and thus improves the signal-to-noise ratio (SNR),e.g. for 100 relevant pixels by a factor of 10. Thus, the maximum reconstruction depth for defects of a certain size can be increased.

In this project, the virtual wave concept (VWC) is significantly extended. So far, only homogeneous isotropic or anisotropic materials with a planar surface could be examined with it, and also only with single short pulses for illumination (pulse thermography). Now the VWC is to be extended to 2-component systems that are important for practical applications, e.g. foam core with a CFRP top layer. By excitation with several pulses (coded multi-pulse excitation) the possible depth range shall be extended and optimized for defect detection at a certain depth. This optimization of the pulse waveform shall be done sample specific with "reinforcement learning". This should also allow signals to be measured from below layers that are highly thermally reflective - e.g. in a foam core below a carbon fiber top layer. Finally, these methods are also to be used on curved surfaces, where the surface geometry is detected in a first step and then used for defect reconstruction in a second step.

• Research Center for Non-Destructive Testing GmbH (RECENDT)
• voidsy GmbH