Bead Geometry Prediction for Multi-layer and Multi-bead Wire and Arc Additive Manufacturing Based on XGBoost

In the process of multi-layer and multi-bead wire and arc additive manufacturing (WAAM), the geometry of each layer has an important influence on the dimensional accuracy and surface quality of the final forming parts. In this paper, machine learning model is used to predict the geometrical morphology of multi-layer and multi-channel WAAM forming parts. A series of experiments of WAAAM under different parameters were carried out by rotating combination experiment, and the bead geometry of the forming parts were obtained by visual sensing system developed by ourselves. Aiming at the problem of less data samples which would lead to over-fitting in the process of model training, this paper introduces the XGBoost algorithm for modeling. Compared with the neural network algorithm, the regression prediction model of arc additive manufacturing morphology based on XGBoost has a higher prediction accuracy.