Stochastic geometric model for overlapping beads fabricated using wire arc additive manufacturing under uncertainties

Abstract Wire-Arc Additive Manufacturing (WAAM) known for its applications in fabricating parts of big dimensions is gaining momentum with the recent optimisations and innovations. WAAM coupled with improved metal transfer mode called the Cold Metal Transfer (CMT) has improved the geometric quality of the parts produced. But the complex physics inherent to the process requires variable controls for the fabrication, generating surface irregularities on the beads which are difficult to predict. These irregularities cumulate as the fabrication is advanced from single beads to layers, and then to multiple layers forming the final part. The formation of layers involves the overlapping of adjacent beads which are spaced at an offset distance. But because of the irregular form of the beads, the offset distance also has to be controlled as a function of the form of the bead sections. Several existing models neglect the continuous variations and recommend deterministic values which are different from reality. Therefore, the current study is dedicated to developing a statistical model to predict the offset distance and its variations by considering the actual irregularities in the bead geometry. A total of 140 beads are fabricated using four different sets of input parameters, which enables studying the variation in bead geometry from thin to thick beads. This ensures that the maximum variations of bead geometry in the fabrication envelope of the process are studied and used for developing the prediction model. The results show that the overlapping distance is neither a deterministic value nor a generic value for the entire process. Furthermore, using the developed model of overlapping distance predictions are made which could be employed for the fabrication.