Application of back-propagation neural networks to defect characterization using eddy current testing

被引：2

作者：

Zhou, Xinwu ^{[1
,2
]}

Urayama, Ryoichi ^{[2
]}

Uchimoto, Tetsuya ^{[2
,3
,4
]}

Takagi, Toshiyuki ^{[3
,4
,5
]}

机构：

[1] Tohoku Univ, Grad Sch Engn, Dept Mech Syst Engn, Sendai, Miyagi, Japan

[2] Tohoku Univ, Inst Fluid Sci, Sendai, Miyagi, Japan

[3] Univ Lyon, CNRS, ELyTMax UMI 3757, Lyon, France

[4] Tohoku Univ, Sendai, Miyagi, Japan

[5] Tohoku Univ, Org Res Promot, Tohoku Forum Creat, Sendai, Miyagi, Japan

来源：

INTERNATIONAL JOURNAL OF APPLIED ELECTROMAGNETICS AND MECHANICS | 2020年 / 64卷 / 1-4期

关键词：

Eddy current testing; artificial intelligence; back-propagation neural network;

D O I：

10.3233/JAE-209394

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 testing is widely used for the automatic detection of defects in conductive materials. However, this method is strongly affected by probe scanning conditions and requires signal analysis to be carried out by experienced inspectors. In this study, back-propagation neural networks were used to predict the depth and length of unknown slits by analyzing eddy current signals in the presence of noise caused by probe lift-off and tilting. The constructed neural networks were shown to predict the depth and length of defects with relative errors of 4.6% and 6.2%, respectively.

引用

页码：817 / 825

页数：9

共 9 条

[1]

[Anonymous], 2010, 42172010 JEAG

[2] ELECTRICAL RESISTIVITY OF SOME ENGINEERING ALLOYS AT LOW TEMPERATURES [J].

CLARK, AF ;

CHILDS, GE ;

WALLACE, GH .

CRYOGENICS, 1970, 10 (04) :295-&

[3] Fast Eddy Current Testing Defect Classification Using Lissajous Figures [J].

D'Angelo, Gianni ;

Laracca, Marco ;

Rampone, Salvatore ;

Betta, Giovanni .

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2018, 67 (04) :821-830

[4]

Deng Weiquan, 2019, METALS, V9, P1

[5]

Kang Hyouk L., 2012, IEEE T MAGN, V48, P3965

[6]

Le Ber L, 2006, AIP CONF PROC, V820, P684

[7] Quantitative evaluation of crack depths and angles for pulsed eddy current non-destructive testing [J].

Nafiah, Faris ;

Sophian, Ali ;

Khan, Md Raisuddin ;

Abidin, Ilham Mukriz Zainal .

NDT & E INTERNATIONAL, 2019, 102 :180-188

[8] An application of back propagation neural network for the steel stress detection based on Barkhausen noise theory [J].

Wang, Ping ;

Zhu, Lei ;

Zhu, Qiujun ;

Ji, Xiaoli ;

Wang, Haitao ;

Tian, Guiyun ;

Yao, Entao .

NDT & E INTERNATIONAL, 2013, 55 :9-14

[9]

Zhang Q., 2018, SCI REP-UK, V8, P1, DOI [10.1038/s41598-017-17765-5, DOI 10.1038/S41598-017-17765-5]

← 1 →