Abstract
An integrated approach based on the use of Design of Experiment (DOE), Artificial Neural Networks (ANN) and Genetic Algorithm (GA) for modeling of Gas Metal Arc Welding (GMAW) process has been explained in this paper. The effects on the five weld bead geometric descriptors by the five weld process variables has been initiated by means of 2n-1 fractional factorial experimental design technique. In this study, the 2n-1 fractional factorial experimental design method was applied on the data available from a published work (Wu in Weld J 43(4):179s–183s, 1964) to determine the main effects as well as 2-factor interaction effects. Using the results of 25-1 fractional factorial design experiments, multiple linear regression equations are postulated for both main as well as 2-factor interaction effects. It is observed that, main and 2-factor interaction linear equations have resulted in a small error between the estimated and experimental values while multiple linear regression equations have fitted the data very well. Back-propagation neural networks are used to associate the welding process variables with the features of the weld bead geometry. It is seen that neural network for estimating the weld bead geometric parameters can be effectively implemented, with little error percentage difference between the estimated and experimental results. In this study, Genetic Algorithms are used for optimizing the process parameters. The five process variables optimized by the GA are well within the vectors of minimum and maximum values of the controllable process variables of the experimental conditions in all the bead geometry descriptor cases. It can be concluded that genetic algorithms can able to optimize the process parameters for the desired weld bead geometric parameters.
Article PDF
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
References
Wu S.M.: Analysis of rail steel bar welds by two-level factorial design. Weld. J. 43(4), 179s–183s (1964)
Moon H., Na S.: Optimum design based on mathematical model and neural network to predict weld parameters for fillet joints. J. Manuf. Syst. 16(1), 13–23 (1997)
Wu S.M., Stegner D.A.J., Braton N.R.: Prediction of heat input for welding. Weld. J. 46(3), 122–128 (1967)
Bartee E.M.: Engineering Experimental Design Fundamentals. Prentice-Hall, Inc., Englewood Cliffs (1968)
Montgomery D.C.: Design and Analysis of Experiments, 4th edn. Wiley, NY (1997)
Karsai G., Cook G.E., Andersen K., Ramaswamy K.: Artificial neural networks applied to arc welding process modeling and control. IEEE Trans. Ind. Applicat. 26(5), 824–830 (1990)
Anderson, K.: Studies and implementation of stationary models of the gas tungsten arc welding process. Master’s thesis, Vanderbilt University (1992)
Hsieah C.C., Hunag M.-W., Arora J.S.: Genetic algorithm for sequence type problems in engineering design. Int. J. Numer. Methods Eng. 40, 3105–3115 (1997)
Kim D., Rhee S.: Optimization of arc welding process parameters using genetic algorithm. Weld. J. 80(7), 184–189 (2001)
Lippmann, R.P.: An introduction to computing with neural nets. IEEE ASSP Mag. 4–22 (1987)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Nagesh, D.S., Datta, G.L. Modeling of fillet welded joint of GMAW process: integrated approach using DOE, ANN and GA. Int J Interact Des Manuf 2, 127–136 (2008). https://doi.org/10.1007/s12008-008-0042-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12008-008-0042-8