A 3D Reconstruction Method for Complex Defects Based on MFD and Multidirectional MFL

Magnetic flux leakage (MFL) detection is a widely used non-destructive testing method for ferromagnetic materials. In the field of research, 3D profile reconstruction is one of the most important concerns. In practice, the opening profile of de-fects is complex and can easily lead to information los...

Full description

Saved in:
Bibliographic Details
Published in2024 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) pp. 1 - 6
Main Authors Li, Shenaping, Zhang, Jie, Zhang, Xu, Feng, Chunrui, Bai, Libing
Format Conference Proceeding
LanguageEnglish
Published IEEE 20.05.2024
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Magnetic flux leakage (MFL) detection is a widely used non-destructive testing method for ferromagnetic materials. In the field of research, 3D profile reconstruction is one of the most important concerns. In practice, the opening profile of de-fects is complex and can easily lead to information loss and signal distortion. It is difficult to reconstruct the 3D profile without considering the opening profile and magnetization directions. In this paper, a 3D profile inversion method that combines magnetic field disturbance (MFD) and MFL is proposed. This method obtains an accurate defect opening profile by using MFD. Then, the depth profile is reconstructed using MFL signals acquired from two orthogonal magnetizations. The pseudo-inverse method is applied as the depth profile inversion algorithm. It combines the inversion equation of two directions, finding a solution that can fulfill the inputted two MFL signals simultaneously. Such a two-step process can help consider the influence of the opening profile and complete the information. Experiments show that the method can reconstruct the 3D profile of defects and reach to minimal 0.307 in reconstruction error. Which exhibits potential for nractical applications,
ISSN:2642-2077
DOI:10.1109/I2MTC60896.2024.10561243