A neural network model is developed for on-line nugget size estimation in resistance spot welding. The variables used consist of features extracted from both controllable process input variables and on-line signals. A systematic signal and feature selection procedure is developed. The three commonly observed on-line signals, dynamic resistance, force, and electrode displacement, have been proven to carry similar information. Thus, only dynamic resistance is used in the model. The obtained model has been demonstrated to be robust over various welding conditions including electrode wear. [S1087-1357(00)01204-1]
Issue Section:
Technical Papers
1.
Stiebel, A., Ulmer, C., Kodrack, D., and Holmes, B., 1986, “Monitoring and Control of Spot Weld Operations,” SAE Technical Paper, No. 860579, International Congress and Exposition, Detroit, Michigan.
2.
Auto/Steel Partnership, 1995, Weld Quality Test Method Manual, Standardized Test Method Task Force.
3.
Takuma
, M.
, Shinke
, N.
, and Motono
, H.
, 1996
, “Evaluation of Function of Spot-Welded Joint using Ultrasonic Inspection
,” Trans. Jpn. Soc. Mech. Eng., Part A
, 62
, No. 595
, pp. 776
–780
.4.
Li
, W.
, Hu
, S. J.
, and Ni
, J.
, 1999
, “On-Line Expulsion Detection and Estimation for Resistance Spot Welding
,” Trans. NAMRI/SME
, XXVII
, pp. 123
–128
.Copyright © 2000
by ASME
You do not currently have access to this content.
