Axisymmetric eddy current inspection of highly conducting thin layers via asymptotic models

Houssem Haddar (1) , Zixian Jiang (1)

1 DeFI - Shape reconstruction and identification

Thin copper deposits covering the steam generator tubes can blind eddy current probes in non-destructive testings of problematic faults and are therefore important to be identified. Existing methods based on shape reconstruction using eddy current signals encounter difficulties of high numerical costs due to the layer's small thickness and high conductivity. In this article, we approximate the axisymmetric eddy current problem with some appropriate asymptotic models using effective transmission conditions representing the thin deposits. In these models, the geometrical information related to the deposit is transformed into parameter coefficients on a fictitious interface. Standard iterative inversion algorithm is then applied to the asymptotic models in order to reconstruct the thickness of the thin copper layers. Numerical tests both validating the asymptotic model and benefit of the inversion procedure are provided.

Mots clés: axisymmetric eddy current inspection; asymptotic model; thickness reconstruction of thin layers

Domaines: Analyse numérique [cs.NA]

Houssem Haddar, Zixian Jiang. Axisymmetric eddy current inspection of highly conducting thin layers via asymptotic models. Inverse Problems, 2015. ⟨hal-01214308⟩