Automatic Classification of Welding Flaws in Ultrasonic Testing Using the Layer Multi-Class Classifier Based on v-SVM

Yuan Chen; Hong Wei Ma

doi:10.4028/www.scientific.net/AMR.490-495.3215

Paper Titles

Design of CO₂ Heat Pump Water Temperature Control System Based on C8051F500 SCM and Electronic Expansion Valve
p.3197

Constant-Wavelength Synchronous Fluorescence Spectrometry for the Determination of Polycyclic Aromatic Hydrocarbons in Water Samples
p.3202

Development of Experimental System in Organic Synthesis
p.3207

Low Temperature Large Scale CVD Synthesis of Nano Onion-Like Fullerenes
p.3211

Automatic Classification of Welding Flaws in Ultrasonic Testing Using the Layer Multi-Class Classifier Based on v-SVM
p.3215

Extremely High Efficiency Orange Phosphorescent Organic Light-Emitting Devices
p.3221

Nonlocal Size Dependence of a Softness Nanobeam with Large Axial Tension under Various Boundary Conditions
p.3226

Effect of Acid Pretreatment on Photocatalysis Bleaching
p.3231

Melt Flow Type and Physical Model of Semisolid Rheo-Casting Process by Cooling Sloping Plate
p.3237

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 490-495Automatic Classification of Welding Flaws in...

Automatic Classification of Welding Flaws in Ultrasonic Testing Using the Layer Multi-Class Classifier Based on v-SVM

Abstract:

Aiming at the difficult question of qualitative analysis for welding flaws in ultrasonic testing, a novel method of flaw classification utilizing the combination of wavelet packet transform (WPT) with the layer multi-class classifier based on -SVM is presented in this paper. First, WPT is applied to extract features for ultrasonic echo-signals of welding flaws, and flaws are automatically classified by using the layer multi-class classifier based on -SVM. Then, its classification results are compared with that of several widely used multi-class SVM classification methods. To validate the method above, a series sample of butt girth welds of seamless steel tube with four types of welding flaws is made, and some experiments are performed. The results show that the proposed method is superior to other multi-class SVM classification methods in the aspects of classification accuracy, training and testing time.

You might also be interested in these eBooks

Mechatronics and Intelligent Materials II

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 490-495)

Pages:

3215-3220

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.490-495.3215

Citation:

Cite this paper

Online since:

March 2012

Authors:

Yuan Chen, Hong Wei Ma

Keywords:

Multi-Class Classification, Ultrasonic Testing (UT), V-SVM, Wavelet Packet Transform, Welding Flaws

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] S.J. Song, H.J. Kim, and H. Cho: Nuclear Engineering and Design, Vol. 212 (2002), p.307.

Google Scholar

[2] F. Bettayeb, T. Rachedi and H. Benbartaoui: Ultrasonics, Vol. 41 (2004), p.853.

Google Scholar

[3] S. Sambath, P. Nagaraj and N. Selvakumar: Journal of Nondestructive Evaluation, Vol. 30 (2011) No. 1, p.20.

Google Scholar

[4] Ó. Martín, M López and F. Martín: Journal of Materials Processing Technology, Vol. 183 (2007), p.226.

Google Scholar

[5] T.S. Li: Computers & Industrial Engineering, Vol. 56 (2009), p.1154.

Google Scholar

[6] X.F. Zhao, G.F. Yin, G.Y. Tian, et al: Chinese Journal of Scientific Instrument, Vol. 29 (2008) No. 5, p.908, in Chinese.

Google Scholar

[7] B. Dai, J. Zhao and Y. Zhou: Journal of Chemical Industry and Engineering, Vol. 59 (2008) No. 7, p.1812, in Chinese.

Google Scholar

[8] G. Li and M. Huang: Journal of Hefei University of Technology, Vol. 29 (2006) No. 2, p.246.

Google Scholar

[9] Information on http: /www. csie. ntu. edu. tw /~cjlin/papers/nusvmtutorial. pdf.

Google Scholar