Mechanical property prediction of commercially pure titanium welds with artificial neural network
Y. H. Wei, H. K. D. H. Bhadeshia and T. Sourmail
Abstract
The factors that control the mechanical properties of commercially pure titanium welds have been investigated by interpreting published data from single-pass, autogenous welds using neural network analysis. The properties studied include the ultimate tensile yield strengths, elongation, reduction of area, Vickers and Rockwell B hardness values. Oxygen and nitrogen have the most significant effects on the strength of the welds while hydrogen is unimportant over the range investigated. The results have been compared against commonly used, linear oxygen equivalent equations.
Journal of Materials Science and Technology, Vol. 21, 2005, 403--407.
Dowload PDF file of paper.
