Identification of Defects Using Multi-Frequency Eddy Current Technique and Artificial Neural Networks

被引：0

作者：

Lopato, Przemyslaw ^{[1
]}

Chady, Tomasz ^{[1
]}

Sikora, Ryszard ^{[1
]}

机构：

[1] Zachodniopomorski Uniwersytet Technol, Katedra Elektrotech Teoretycznej & Informatyki, PL-70310 Szczecin, Poland

来源：

PRZEGLAD ELEKTROTECHNICZNY | 2009年 / 85卷 / 08期

关键词：

nondestructive testing; artificial neural networks; neural identification; eddy current testing; RECOGNITION;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Eddy current (EC) testing is a commonly used technique for nondestructive evaluation of conducting materials. Identification of flaws, cracks and other sorts of discontinuities may be realized using Automatic Defect Recognition (ADR) algorithms. This paper presents flaws identification algorithms based on artificial neural networks. Three different approaches are proposed. All presented methods were evaluated using results of measurements achieved for the Inconel 600 plates with the flaws having rectangular and non-rectangular profiles. In all cases reliable results of identification were achieved.

引用

页码：18 / 22

页数：5

共 4 条

[1] Natural crack recognition using inverse neural model and multi-frequency eddy current method [J].