基于坐标偏移磁偶极子-牛顿-拉夫逊法的三维不规则缺陷重构方法
DOI:
CSTR:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TM153. 1 TH878

基金项目:


3D irregular defect reconstruction method based on coordinate offset magnetic dipole-Newton-Raphson method
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    漏磁检测因其操作方便、对检测环境要求低、自动化程度高等特点,被广泛应用于铁磁性材料的缺陷检测中。 而不规则 缺陷漏磁重构的病态性导致重构结果精度低,特别是三维不规则缺陷的漏磁重构。 因此,本文将三维缺陷的重构问题转化为二 维缺陷重构问题,提出了坐标偏移磁偶极子前向模型,可以快速精确地计算任意复杂缺陷的漏磁信号,并使用牛顿-拉夫逊法, 实现对三维不规则缺陷的重构。 仿真和实验结果表明,本文所提重构算法相比于 Levenberg-Marquardt 重构算法有明显的精度 提升,重构误差平均减少了约 41% ,最大深度误差平均减少了 62% ,实现了三维不规则缺陷的快速重构。

    Abstract:

    Magnetic flux leakage detection is widely used in the defect detection of ferromagnetic materials because of its convenient operation, low requirement of detection environment and high automation. However, the pathology of irregular defect leakage reconstruction leads to low accurate reconstruction results, especially the magnetic leakage reconstruction of 3D irregular defects. Therefore, the 3D defect reconstruction problem is transformed into the 2D defect reconstruction one in this study. A forward model based on coordinate offset magnetic dipole is proposed, which can quickly and accurately calculate the magnetic leakage signal of arbitrary complex defects, and the Newton-Raphson method is used to reconstruct three-dimensional irregular defects. The simulation and experimental results show that compared with the Levenberg-Marquardt reconstruction algorithm, the proposed reconstruction algorithm provides obvious accuracy improvement by reducing the average reconstruction errors of 41% , and average maximum depth errors of 62% , which realizes the rapid reconstruction of 3D irregular defects. Keywords:magnetic leakage detection; defect reconstruction; magnet

    参考文献
    相似文献
    引证文献
引用本文

王雨菲,韩文花.基于坐标偏移磁偶极子-牛顿-拉夫逊法的三维不规则缺陷重构方法[J].仪器仪表学报,2024,45(6):328-336

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2024-11-25
  • 出版日期:
文章二维码