The Paton Welding Journal, 2002, №04

2002 №04 (13)

2002 №04 (02)

---

The Paton Welding Journal, 2002, #4, 2-5 pages

Selection of input variables and structure of neural network to evaluate the quality of resistance spot welding

N.V. Podola, V.S. Gavrish, P.M. Rudenko

E.O. Paton Electric Welding Institute of the NASU 11 Kazymyr Malevych Str., 03150, Kyiv, Ukraine.

Abstract
Analysis of input parameters of neural networks to control the quality of spot welding was conducted. The principal requirements to neural networks have been defined. It is shown that the proposed neural network ensures the required accuracy of predicting the weld spot nugget diameter and can be built into the microcontroller to control the spot welding process
Keywords: resistance spot welding, quality control, neural networks, microcontrollers

References

1. Podola, N.V., Rudenko, P.M. (1988) Microcomputer-based automated systems for monitoring and control of resistance and arc welding on the base of a microcomputer. Moscow: Mashinostrocnic.
2. Paton, B.E., Podola, N.V., Gavrish, V.S. et al. (1998) Evaluation of the quality of resistance spot welding, using neural networks. Avtomatich. Svarka, 12, 3−10.
3. Irving, B. (1997) Neural networks are raying off on the pro¬duction line. Welding J., 10, 59−64.
4. Monari, G., Dieraert, O., Oberle, II. (1982) Predication of spot welding diameter using neural networks. IIW Doc. 111-1108-98.
5. Zhawg, Z., Li, Y., Пе, X. et al. (1997) Artificial neural networks applied to spot welding process modeling. China Welding, 1, 41−48.
6. Matsuyama, K. (1982) Nugget size sensing of spot weld based on neural network learning. IIW Doc. III-1081-97.
7. Dilthey, V., Dickersbach, J. (1982) Quality assurance of re¬sistance spot welding bv emplovment of neural networks. IIW Doc. III-1O93-97.
8. Brown, J., Jobling, C., Williams, N. (1982) Optimization of signal inputs to a neural network for modeling spot welding of zinc coated steels. IIW Doc. III-117-98.
9. Satoh, T., Abe, II., Suzuyama, S. (1982) A trial of quality assurance in resistance spot welding by aid of neural net¬work and fuzzy reasoning. IIW Doc. III-1083-97.

---

Suggested Citation

N.V. Podola, V.S. Gavrish, P.M. Rudenko (2002) Selection of input variables and structure of neural network to evaluate the quality of resistance spot welding. The Paton Welding J., 04, 2-5.

---