Abstract
Appropriate selection of welding conditions to guarantee requisite weld joint mechanical properties is ever difficult because of their complex interactions. An approach is presented here to identify suitable welding conditions in typical two-wire tandem submerged arc welding (SAW-T) that involves many welding variables. First, an objective function is defined, which depicts the squared error between the mechanical properties of weld joint and of base material. A set of artificial neural network (ANN)-based models are developed next to estimate the weld joint properties as function of welding conditions using experimentally measured results. The neural network model-based predictions are used next to create a set of process map contours that depict the minimum achievable values of the objective function and the corresponding welding conditions. In typical SAW-T of HSLA steel, welding speed from 9.0 to 11.5 mm/s, leading wire current from 530 to 580 A, and trailing wire negative current from 680 to 910 A are found to be the most optimal.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have