Quantitative nondestructive estimation of deep defects in conductive structures

Issue title: Selected Papers from the 14th International Symposium on Applied Electromagnetics and Mechanics (ISEM 2009), Part II

Article type: Research Article

Authors: Li, Guohou^a | Huang, Pingjie^{a; *} | Chen, Peihua^a | Hou, Dibo^a | Zhang, Guangxin^a | Zhou, Zekui^a

Affiliations: [a] State Key Laboratory of Industrial Control Technology, Department of Control Science & Engineering, Zhejiang University, Hangzhou 310027, China

Correspondence: [*] Corresponding author. Pingjie Huang, Tel.: +86 57187952241; Fax: +86 57187951219; E-mail: vamadis@yahoo.com.cn

Abstract: Eddy current testing (ECT) is a widely used nondestructive testing method in many such fields as the power system, railway, machinery, aircraft and nuclear industries. Because of the limitation of penetration depth of the conventional eddy currents testing coils and other affecting factors existing in the eddy current response, inverse problems to evaluate the deep defects from ECT signals of the conductive materials have been a challenging task. The objectives of this study are to introduce methods based on the fuzzy Bayesian networks (BNs), and the giant magnetoresistance (GMR) sensors to estimate the deep defects from their ECT signals. The experimental validation carried out on the specimens with different known defects confirmed the suitability of the proposed approach for deep defects evaluation.

Keywords: Eddy current testing, deep defects evaluation, fuzzy bayesian networks, radial base function neural network, giant magnetoresistance

DOI: 10.3233/JAE-2010-1249

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 33, no. 3-4, pp. 1273-1278, 2010