Quantitative Inversion of Stress and Crack in Ferromagnetic Materials Based on Metal Magnetic Memory Method

Abstract

Ferromagnetic materials are widely used in engineering structures. Damages caused during manufacture and the use of ferromagnetic materials may seriously undermine the safety of the engineering structure, and even lead to serious industrial accidents. For the early damage testing problem of ferromagnetic materials, it is imperative to propose and adopt new nondestructive testing methods. The metal magnetic memory (MMM) method, which is also known as the micromagnetic testing method, can achieve the early detection of damages in ferromagnetic materials such as steels. However, due to the lack of the quantitative research, the MMM method is only applicable for damage localization. In this paper, the MMM method is utilized to quantitatively evaluate the stress and crack in ferromagnetic materials. The inverse model including the objective function and optimization parameters is proposed. The reconstruction approach is established based on the conjugate gradient inversion method and the exact line search algorithm. In the theoretical analysis, the effectiveness of the MMM method in quantitative determination regarding the position and size of defects is verified with the experimental signal of a hole defect. Following this, the theoretical analyses of localization and sizing of stress and crack are conducted by using the simulation results as the measured signals. It can be found from the results that the MMM method can be applicable for quantitative evaluation of surface rectangular cracks, surface stress concentration zone, convex-type defect, and so on. In addition, the influences of the signal selection, sampling rate, lift-off value, stress level, and noise amplitude on the quantitative analysis of early stress concentration are discussed in details.