A weld bead footprint locus model for predicting the overlap of weld beads in wire arc additive manufacturing

Abstract

Weld bead models are crucial in the process planning phase for wire arc additive manufacturing. By simulating the overlap between weld beads, the toolpath can be optimized to obtain a high-quality production process. The creation of datasets covering all possible combinations of process parameters and boundary conditions is hindered by a high and costly experimental effort. Accurate and numerically efficient modelling approaches that can be derived from simple and cost-effective experimental campaigns are needed. This article introduces a new weld bead footprint model to address this challenge. The proposed footprint locus model is used to predict the weld bead footprint, which is afterwards used to predict the weld bead shape. A calibration procedure is introduced, which allows for calibrating the model parameters for a reference situation. The geometrical interpretation of the model parameters is exploited for estimating how they should change when welding parameters or boundary conditions different from the nominal ones are used. The model is validated with experimental data and compared to a baseline model derived from assumptions commonly found in the literature. The deposition of two parts produced with different processing conditions could be predicted with an error inferior to 2% the deposited height.