Machine learning-based bead modeling of wire arc additive manufacturing (WAAM) using an industrial robot

The complex and intricate metallic parts can be made by wire arc additive manufacturing (WAAM) with advantages such as minimum time, and more accuracy than other deposition techniques. The influence of process parameters on the built wall geometry of multi-bead overlapped wire arc additive manufacturing with the assistance of the welding robot was focused on and studied in this work. The process parameters
namely weld travel speed (WTS), wire feed speed (WFS), tool waiting time (TW), input voltage (V), and shielding gas mixture (SGM) exposure were considered to be more predominant parameters in building the wall geometry. Hence, these parameters were considered to be optimized in the current paper and also discussed their effect on the mechanical properties, dimensional accuracy, and surface quality.
Further, with the optimized parameters, an ANN model was constructed and simulated with the specified boundary conditions. The simulation results were found to be a good agreement with the experimental inferences.