A Parametric Study of Magneto-Optic Imaging Using Finite-Element Analysis Applied to Aircraft Rivet Site Inspection

Abstract

Magneto-optic/eddy current imaging (MOI) has become increasingly popular for inspecting aging aluminum airframes for cracks and corrosion due to its accuracy, reliability, and ease of use. As inspection requirements change, modifications to the MOI system must be made to improve sensitivity and resolution to reliably detect smaller and/or deeper defects in the aircraft structure. Incorporating such improvements by cut and try methods is time-consuming and expensive. Therefore, a numerical simulation model that produces quantitative values of the magnetic fields associated with induced eddy currents interacting with structural defects is an essential complement to the instrument development process. Such a model provides a convenient tool for parametrically evaluating the effectiveness of the MOI for detecting various structural defects. This paper presents a three-dimensional finite-element model of Maxwell's equations, utilizing the A-V formulation for numerical simulation of the MOI operation. The model is used to predict quantitative values of field distributions that produce the binary magneto-optic images of subsurface fatigue cracks at rivet sites in an aluminum airframe structure. A parametric study is performed to determine the effects of MOI operational parameters on the binary images. A skewness parameter based on the binary images is established to provide a measure of defect size. This parameter will prove useful for automatic detection and classification of defects. The model-generated images show good agreement with experimentally derived MOI images