Welding Defects Classification Based on Multi-Weights Neural Network

Liang Hua; Peng  Xue; Jin Ping Tang; Hui  Jin; Qi Zhang

doi:10.4028/www.scientific.net/AMR.820.130

Paper Titles

Effect of Buffer Layer Thickness on Fatigue and Residual Stress of Welded High-Strength Low-Alloy
p.110

Research on Hydraulic Control System of Straightening Machine with Nine Rolls for Plates Based on BP Neural Network PID Controller
p.117

Five-Axis NC Simulating Manufacturing of Titanium Alloy Integral Impeller
p.122

Study on the 5-Axis Linked-Blade Grinder Based on Online Detection Simulation Technology
p.126

Welding Defects Classification Based on Multi-Weights Neural Network
p.130

Tool Wear and Surface Integrity in End Milling of Ti6Al4V with Polycrystalline Diamond Tools
p.134

The Influence of Material Parameters on Penetrator Embedding Pre-Perforated Steel Plate Vertically
p.138

Experimental Study of Assembly Stress Based on Manufacturing Characteristics
p.141

A Combined Error Model for Thermal Error Compensation of Machine Tools
p.147

HomeAdvanced Materials ResearchAdvanced Materials Research Vol. 820Welding Defects Classification Based on...

Welding Defects Classification Based on Multi-Weights Neural Network

Abstract:

Incomplete fusion and incomplete penetration are two types of damage serious welding defects. These two kinds of defects have the similarity in the features in X-ray imaging. Identifying the two kinds of defects automatically and accurately can improve the welding technology and improve the quality of welding effectively. The causes of defects and features of X-ray images are described in the paper. The welding defects calssification method based on multi-weights neural network is put forward in the paper. The multi-weights neural network based on graphic geometry theory is introduced, which uses the geometrical shape in high dimensional space to cover the same class defect samples via constructing multi-weights neural network. The experimental results proved the effectiveness of the algorithm.

You might also be interested in these eBooks

Metallic Materials and Manufacturing Technology

View Preview

Info:

Periodical:

Advanced Materials Research (Volume 820)

Pages:

130-133

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.820.130

Citation:

Cite this paper

Online since:

September 2013

Authors:

Liang Hua, Peng Xue, Jin Ping Tang, Hui Jin, Qi Zhang

Keywords:

Incomplete Fusion, Incomplete Penetration, Multi-Weights Neural Network

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Xu Ke , Pu Leiye and Duan Honghong: Identification of incomplete penetration and incomplete fusion in weld nondestructive testing. Welded Pipe and Tube，Vol. 34(2011), pp.51-53, in Chinese.

Google Scholar

[2] Lin Shiyi: Incomplete penetration and incomplete fusion defect image recognition on the X-ray films. Mechanical & Electrical Technology, vol. 2(2007), pp.72-74, in Chinese.

Google Scholar

[3] Sun zheng: Study on image-based extraction and recognition of welding defects(PhD thesis). China University of Mining and Technology, Mechanical and Electrical Engineering Institute, 2010, in Chinese.

Google Scholar

[4] Information on http: /www. maihanji. com/bbs/showcontent. asp?id=26472.

Google Scholar

[5] Jianhui Hu: Studies on iris recognition algorithm based on the biomimetic pattern recognition (Master's degree thesis). Zhejiang university of technology, College of Information Engineering, 2006, in Chinese.

Google Scholar