Metamodel-based Markov-Chain-Monte-Carlo parameter inversion applied in eddy current flaw characterization

Caifang Cai (1) , Roberto Miorelli (2) , Marc Lambert (3) , Thomas Rodet (4) , Dominique Lesselier (1) , Pierre-Emile Lhuillier (5)

1 L2S - Laboratoire des signaux et systèmes
2 LSME - Laboratoire de Simulation et de Modélisation Électromagnetique
3 GeePs - Laboratoire Génie électrique et électronique de Paris
4 SATIE - Systèmes et Applications des Technologies de l'Information et de l'Energie
5 EDF R&D MMC - Matériaux et Mécanique des Composants

Flaw characterization in eddy current testing usually requires to solve a nonlinear inverse problem. Due to high computational cost, Markov Chain Monte Carlo (MCMC) methods are hardly employed since often needing many forward evaluations. However, they have good potential in dealing with complicated forward models and they do not reduce to only providing the parameters sought. Here, we introduce a computationally-cheap surrogate forward model into a MCMC algorithm for eddy current flaw characterization. Due to the use of a database trained off-line, we benefit from the MCMC algorithm for getting more information and we do not suffer from the computational burden. Numerous experiments are carried out to validate the approach. The results include not only the estimated parameters, but also standard deviations, marginal densities and correlation coefficients between two parameters of interest.

Keywords: metamodel; Bayesian; flaw characterization; eddy-current; non-destructive testing; MCMC; Inversion

Domains: Engineering Sciences [physics]; Signal and Image processing; Engineering Sciences [physics]; Electromagnetism; Mathematics [math]; Mathematical Physics [math-ph]

Caifang Cai, Roberto Miorelli, Marc Lambert, Thomas Rodet, Dominique Lesselier, et al.. Metamodel-based Markov-Chain-Monte-Carlo parameter inversion applied in eddy current flaw characterization. NDT & E International, 2018, 99, pp.13-22. ⟨10.1016/j.ndteint.2018.02.004⟩. ⟨hal-01634071⟩