首页> 外文OA文献 >Neural Network Based Pattern Recognition for Defect Detection of Load/Lock Slots
【2h】

Neural Network Based Pattern Recognition for Defect Detection of Load/Lock Slots

机译:基于神经网络的模式识别用于加载/锁定槽缺陷检测

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务,本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文,但由于OA文献来源多样且变更频繁,仍可能出现获取不到、文献不完整或与标题不符等情况,如果获取不到我们将提供退款服务。请知悉。

摘要

The Load (Lock) Slot viewed from the top is shaped like a running track around a soccer field (see Figure 1.1). The minimum size of the crack required to be detected is 0.005″×0.010″. For modeling cracks, there are altogether 9 notches in the Slots as marked in the Figure 1.1. The minimum notch at the 9th stage is 0.005″×0.010″. For inspection of this complex geometry, an eddy current probe with a double reflection coil was developed. Scanning occurs in the minor elliptical axis direction while indexing takes place along the major axis. The split of the coil is perpendicular to the scan direction (or in another word, parallel to the Slot edge). The challenge here is the changing scan surface length due to the elliptic shape and the small size (the minor axis is about 1/8″ long) of the inspection area.
机译:从顶部看的“加载(锁定)”插槽的形状像足球场周围的跑道(见图1.1)。需要检测的裂纹的最小尺寸为0.005″×0.010″。对于裂纹建模,如图1.1所示,槽中共有9个槽口。第九阶段的最小缺口为0.005“×0.010”。为了检查这种复杂的几何形状,开发了具有双反射线圈的涡流探头。扫描沿椭圆的短轴方向进行,而分度沿主轴进行。线圈的分割垂直于扫描方向(或换句话说,平行于插槽边缘)。这里的挑战是由于椭圆形和检查区域的小尺寸(短轴大约为1/8英寸长)而导致扫描表面长度的变化。

著录项

  • 作者

    Xu, J.-H.; Birx, D. L.;

  • 作者单位
  • 年度 1995
  • 总页数
  • 原文格式 PDF
  • 正文语种 en
  • 中图分类

相似文献

  • 外文文献
  • 中文文献
  • 专利
代理获取

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号