Sparse Magnetic Array for the Imaging of Defects in Multilayer Metals

Wei-Chen Li,Chun-Yeon Lin

DOI: https://doi.org/10.1109/jsen.2024.3381623

IF: 4.3

2024-05-03

IEEE Sensors Journal

Abstract:This study presents a novel eddy current sensing method for imaging defect distribution in multilayered nonferrous metal plates using an array of magnetic sensors. This method involves dividing the metal plate into small voxels to facilitate reconstruction and uses multifrequency to excite the coil. For each frequency, reconstructing defects from magnetic flux density (MFD) measurements is formulated as a linear inverse problem. The absence or presence of a defect strongly suggests that the solution to the linear inverse problem is binary. This study develops an algorithm under a statistical framework to solve the linear inverse problem with binary constraints. The algorithm introduces a Bernoulli prior over the hidden variables and uses a variational Bayesian inference (VBI) to analytically approximate the posterior probability of the hidden variables conditioned on the observed data. The effectiveness of the proposed method is demonstrated using numerically simulated data and a prototype consisting of a coil and an array of magnetic sensors with 4 mm intervals. The results demonstrate that the method is feasible for imaging defects of 2 mm with a depth resolution of 0.5 mm.

engineering, electrical & electronic,instruments & instrumentation,physics, applied