Modeling and prediction of weld strength in resistance spot welding (RSW) process using artificial neural network (ANN)

Direnç nokta kaynağı endüstride parça metal sacların birleştirilmesinde en yaygın kullanılan yöntemdir. Bu çalışmada 3 mm SAE 1010 çelik sacın nokta kaynağı yapay sinir ağları (NN) kullanılarak modellenmiştir. Geri yayılmalı ağlar kullanılarak (BPN) direnç nokta kaynağı parametreleri (elektrot baskı kuvveti, kaynak akımı ve elektrot çapı) ile kaynak mukavemeti arasındaki ilişki araştırılmıştır. Deneysel sonuçlar ile yapay sinir ağları bulunmuş sonuçlar arasında çok düşük farklar görülmüştür.

Direnç nokta kaynağında kaynak mukavemetinin modellenmesi ve tahmin edilmesinde yapay sinir ağlarının kullanılması

Resistance spot welding (RSW) is a process that is being widely used in the industry for sheet joining purposes. In this paper, use of artificial neural networks (ANN) to model the resistance spot welding process of 3mm SAE 1010 steel sheet is explored. Back-propagation neural networks (BPN) are used to associate the welding process parameters (electrode force, welding current and electrode diameter) with the features of the weld strength. These networks have achieved good agreement with the training data and have yielded satisfactory generalisation.

___

1. Lantz,J., “Principles of Resistance Welding”, Welding Design and Fabrication, 26-29, Feb 2000.
2. Tandoğan,A.B., “Determination of Spot Welding Parameters of Thick”, Heat Treated SAE 4140 Steel Parts, MSc Thesis, Middle East Technical University, The Department of Metallurgical and Materials Engineering, 89p, Ankara, 1998.
3. Crinon,E. and Evans,J.T., “The Effect of Surface Roughness, Oxide Film Thickness and Interfacial Sliding on the Electrical Contact Resistance of Aluminium”, Journal of Materials Science and Engineering, 242, 121-128, 1997.
4. Giroux,D., “Resistance Welding Manual (4th Edition)”, McGraw Hill Inc., 550p., New York, 1989.
5. Eşme,U., “The Effect of Welding and Design Parameters on the Quality of the Resistance Spot Welding (RSW) Joints”, MSc Thesis, Cukurova University The Department of Mechanical Engineering, 107p, 2002.
6. Nagesh,D.S. and Data,G.L., “Prediction of Weld Bead Geometry and Penetration in Shielded Metal-Arc Welding Using Artificial Neural Networks”, 123, 303–312, 2002.
7. Cook,G.E, Andersen,K., Karsai,G. and Ramaswamy,K., “Artificial neural networks applied to arc welding process modeling and control”, IEEE Trans. Ind. App. 26 (5), 824–830, 1990.
8. Cook,G.E. Barnett,R.J. Andersen,K. and Strauss,A.M., “Weld modeling and control using artificial neural networks”, IEEE Trans. Ind. App. 31(6), 1484–1491, 1995.
9. Brown,J.D and Rodd,M.G, “Application of artificial intelligence techniques to resistance spot welding”, Ironmaking & Steelmaking, 199-204, 1998.