A Study on Development of the Optimal Neural Network in GMA Welding Process

Qian Qian Wu; Ji Hye Lee; Jong Pyo Lee; Min Ho Park; Young Su Kim; Ill Soo Kim

doi:10.4028/www.scientific.net/AMR.936.1759

Paper Titles

Effect of Traverse Speed on the Microstructure and Mechanical Properties in Friction Stir Welded Dissimilar Barrier Butt Joints of AA2024 to AA6061/SiC/55p Composite
p.1735

Deformation Behavior of Aluminum Alloy Tube in Semi-Dieless Metal Bellows Forming Process with Local Heating Technique
p.1742

Properties of X80 Pipeline Girth Welds for Different Welding Procedures
p.1747

Investigation of Porous Defects in Spray Formed 7XXX Aluminum Alloy
p.1754

A Study on Development of the Optimal Neural Network in GMA Welding Process
p.1759

Potential Spark Erosion Approaches to Silicon Nanoparticles Production
p.1764

Study on CMT Welding of Stainless Steel Railway Vehicle Body
p.1769

Energy Consumption Discussions of Porcelain Produce with Operating Conditions in a Simplified Model
p.1775

Microstructure and Mechanical Properties of Submerged Arc Welding Filled with Ni Wire Dissimilar Joints Comprising SUS304 and Q245R
p.1780

HomeAdvanced Materials ResearchAdvanced Materials Research Vol. 936A Study on Development of the Optimal Neural...

A Study on Development of the Optimal Neural Network in GMA Welding Process

Abstract:

Gas Metal Arc (GMA) welding is considered as a multi-parameter process that it’s hard to find optimal parameters for good welding. To overcome the problem, an artificial neural network based on the backpropagation algorithm was built to realize the relationships between process parameters and welding quality as output parameter. In this study, Mahalanobis Distance (MD) was employed to evaluate the availability of a given welding parameters which was proved to performance well in multivariate statistics. Input parameters such as welding current and arc voltage were chosen due to their significant influence on the welding quality. To improve the precision of given parameters’ evaluation, neural networks with different configurations were verified. The analyses on the measured and predicted MD by the proposed neural network were conducted. The proposed neural network based on the error backpropogation algorithm was proved to have high reliability to evaluate process parameters, which further makes it available in on-line monitoring system.

You might also be interested in these eBooks

Materials Science and Engineering Technology

View Preview

Info:

Periodical:

Advanced Materials Research (Volume 936)

Pages:

1759-1763

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.936.1759

Citation:

Cite this paper

Online since:

June 2014

Authors:

Qian Qian Wu*, Ji Hye Lee, Jong Pyo Lee, Min Ho Park, Young Su Kim, Ill Soo Kim

Keywords:

Error Backpropagation Algorithm, GMA Welding Process, Mahalanobis Distance, Neural Network (NN)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] T. Melfi, New code requirements for calculating heat input, J. Welding 89(2010) 61-63.

Google Scholar

[2] D.E. Rumelhart, G.E. Hinton and R.J. Williams, Learning representations by back-propagating errors, J. Nature 323 (1986) 533-536.

DOI: 10.1038/323533a0

Google Scholar

[3] A.J. Torija, D.P. Ruiz and A.F. Ramos-Ridao, Use of back-propagation neural networks to predict both level and temporal-spectral composition of sound pressure in urban sound environments, J. Building and Environment 52 (2012) 45-56.

DOI: 10.1016/j.buildenv.2011.12.024

Google Scholar

[4] Jiří Šíma, Back-propagation is not efficient, J. Neural Networks 9 (1996) 1017-1023.

Google Scholar

[5] H.B. Kim, S.H. Jung, T.G. Kim and K.H. Park, Fast learning method for back-propagation neural network by evolutionary adaptation of learning rates, J. Neurocomputing 11 (1996) 101-106.

DOI: 10.1016/0925-2312(96)00009-4

Google Scholar

[6] Y. Fukuoka, H. Matsuki, H. Minamitani and A. Ishida, A modified back-propagation method to avoid false local minima, J. Neural Networks 11 (1998) 1059-1072.

DOI: 10.1016/s0893-6080(98)00087-2

Google Scholar

[7] S.Q. Feng, Off-line Programming of arc welding robot based on UG and welding qualification based on statistic methods, Ph.D., Tianjin University, (2010).

Google Scholar

[8] S.Q. Feng, O. Hiroyuki, T. Hidennori, K. Yuichi and S.S. Hu, Qualitative and quantitative analysis of gmaw welding fault based on Mahalanobis distance, Int. J. Precision Engineering and Manufacturing 12 (2011) 949-955.

DOI: 10.1007/s12541-011-0127-3

Google Scholar