Numerical power prediction for waDED produced aluminium ER5183 wall structures on vertical substrate plates

Due to high melting rates, wire and arc directed energy deposition (waDED) enables the efficient production of large metallic components. Compared to other additive manufacturing processes, operating and material costs are relatively low. The used arc welding process, however, leads to high heat input into the component. To mitigate heat accumulation, cooling strategies can be employed. It is also possible to reduce heat accumulation by adjusting the welding power according to the current component temperature. In this study, a proportional–integral–derivative (PID) control function provided by the finite element software LS-DYNA is utilized to numerically predict optimized waDED welding powers for the first time. Experiments using tungsten inert gas (TIG) welding and aluminium substrate plates are conducted to obtain data required for the calibration of the finite element method (FEM) model. Following calibration, the maximum power deviation between experiment and simulation is 76W, corresponding to an error of only 5%. The calibrated FEM-model is used to predict optimized welding powers for a multi-layer wall structure. Good seam qualities are achieved using the numerically predicted welding powers, effectively preventing variations in seam width caused by insufficient or excessive heat input.