Prediction of tensile strain in mild steel tig weld using artificial neural networks

Abstract

Tensile strain is the relative length of deformation exhibited by a specimen subjected to a tensile force. An artificial neural network (ANN) was employed to predict the tensile strain of the weldment. One hundred welded specimens of mild steel, measuring 60mm x 40mm x10mm, were prepared and calculated using the VWACgauge. The results were employed to train ANN. The research produced an R² of 86% in comparison to the experimental result on a fitted line plot using regression analysis, while correlation analysis obtained in the training and validation exercise from ANN was over 90%. Results of the study have shown that ANN is a robust predictive tool in welding, which could help reduce trial and error in welding processes.