A Bayesian Network Method for Automatic Classification of Eddy Current NDE Signals

Bo Ye; Fang Zeng; Ming Li

doi:10.4028/www.scientific.net/AMM.291-294.2775

Paper Titles

The Metal Foams Modelling with Anisotropic Cells
p.2754

Synthesis and Characterization of ZnAPSO-34 Membrane on Porous α- Al₂O₃ Support
p.2758

Combination of Carbon Nanotube and Alumina by Plasma Chemical Coating
p.2765

The Software Design of Aging Test System for Rechargeable Battery Based on LabVIEW
p.2771

A Bayesian Network Method for Automatic Classification of Eddy Current NDE Signals
p.2775

NPD Management Model Based on Ecological Lifecycle Theory
p.2780

An Improved Frequency Conversion Constant Pressure Water Supply Algorithm
p.2784

Comprehensive Evaluation Method for the Product’s Disassembly Sequences Based on Information Entropy
p.2789

Power Grid Business-Oriented Management Tools of Application Service
p.2794

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 291-294A Bayesian Network Method for Automatic...

A Bayesian Network Method for Automatic Classification of Eddy Current NDE Signals

Abstract:

Eddy current nondestructive evaluation (ECNDE) techniques are widely used in structural integrity and health monitoring. A novel algorithm was proposed for characterizing eddy current (EC) signals. In scanning inspection, the EC signals responding to impedance change were pre-processed for noise elimination and feature extraction. After feature extraction, Bayesian networks (BNs) were carried out to classify EC signals. It is shown by extensive experiments that kernel principal component analysis (KPCA) is better than principal component analysis (PCA) for feature extraction. The methods using BNs by KPCA feature extraction can perform better than the other classification methods.

You might also be interested in these eBooks

Advances in Energy Science and Technology

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 291-294)

Pages:

2775-2779

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.291-294.2775

Citation:

Cite this paper

Online since:

February 2013

Authors:

Bo Ye, Fang Zeng, Ming Li

Keywords:

Bayesian Network (BN), Classification, Eddy Current Nondestructive Evaluation, Inverse Problem

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] L. Oukhellou and P. Aknin: Modified fourier descriptors: a new parametrization of eddy current signatures applied to the rail defect classification, in III International Workshop on Advances in Signal Processing for Nondestructive Evaluation of Materials, Quebec, (1997).

Google Scholar

[2] J. Grman, R. Ravas and L. Syrova: Application of wavelet transformation in eddy current testing of steam generator tubes, in IEEE Instrumentation and Measurement Technology Conference, Budapest, Hungary, (May 2001), pp.392-396.

DOI: 10.1109/imtc.2001.928846

Google Scholar

[3] G.Y. Tian, A. Sophian, D. Taylor and J. Rudlin: IEE Proc. -Sci. Meas. Technol., Vol. 152, No. 4, (Jul. 2005), pp.141-148.

Google Scholar

[4] B. Ye, J. Cai, P. Huang, M. Fan, X. Gong, D. Hou, G. Zhang and Z. Zhou: Nondestructive Testing and Evaluation, Vol. 24, No. 1-2, (Feb. 2009), pp.19-37.

Google Scholar

[5] J. Pearl: Probabilistic Reasoning in Intelligent Systems, Morgan, Kaufmann, (1988).

Google Scholar

[6] N. Friedman, D. Geiger and M. Goldszmidt: Machine Learning, Vol. 29, (1997), pp.131-163.

Google Scholar