Speaker
Description
Induction thermography is a type of active infrared thermography where eddy currents are induced in a specimen, leading to Joule heating of the specimen. The thermal image of the specimen captured by an infrared camera can be used for defect detection in non-destructive testing (NDT). Both surface cracks and subsurface defects can be detected due to the interaction between the defect and the eddy currents as well as the thermal diffusion, causing the surface temperature to deviate from the pattern obtained in sound areas. Pulse heating may be applied for defect detection, but amplitude-modulated heating can be applied for lock-in thermography to enhance the signal-to-noise ratio [1].
In order to study the effect of the many parameters that influence induction thermographic NDT, numerical modeling proves to be a valuable tool. Indeed, computer simulations avoid the costly and time-consuming task of producing samples with artificial defects and allow finer control over parameters such as defect dimensions than is possible with manufactured defects. Building on our long-standing experience in eddy current computations [2] within the CIVA simulation platform [3], dedicated to the simulation of inspection techniques by a variety of excitation sources, we are currently developing a 3D model for the complete induction thermography measurement chain: eddy current generation, thermal diffusion and thermal radiation towards the infrared camera. The goal is to give access to key quantities of interest in CIVA, such as the measured temperature evolution over time and the phase contrast, while keeping the overall simulation time limited.
In this contribution, we will present a coupled eddy current and thermal diffusion model and compare the simulation results with experimental data. The eddy current solver relies on the Boundary Element Method and computes the heating power at the surface of the specimen. The computation of the heating power in the volume of the specimen can be performed but is computationally expensive, and can only be afforded in a small region around the defect. Consequently, a simplified model for the volumetric heating is used in the defect-free region to provide the input to the (volumetric) thermal solver. The validity of these approximations will be demonstrated by comparing simulation results with data from typical induction thermography NDT setups. We will show that the numerical model allows to correctly predict the influence of NDT parameters (material properties, induction frequency, defect dimensions, etc.) on key observables such as the differential and phase contrast.
References
1. Oswald-Tranta, B.: Detection and characterisation of short fatigue cracks by inductive thermography. Quantitative InfraRed Thermography Journal 19(4), 239–260 (2022).
2. Bonnet, M., Demaldent, E.: Eddy-current asymptotics of the Maxwell PMCHWT formulation for multiple bodies and conductivity levels. Computers & Mathematics with Applications 141, 80-101 (2023).
3. EXTENDE: CIVA, NDT Simulation Software. Available at: https://www.extende.com/civa-ndt-simulation-software/ (Jan. 2026).
