Dynamic GMAW Process Model for Layer Geometry Control in Wire Arc Additive Manufacturing

Wire Arc Additive Manufacturing (WAAM) is a large-scale metal AM technology that uses an electric arc to melt a metallic filler wire in order to produce near net-shape metal parts by depositing layers of molten metal on top of each other. The accuracy of the layer geometry during deposition leads to decreased material consumption and post processing machining costs, as well as reducing the occurrence of internal voids in the produced part. Closed-loop control can be used to guarantee a more precise deposition geometry. Thus, the relationship between the deposition parameters and the layer geometry (layer height and wall width) needs to be established in order to control the WAAM process. This paper focuses on modelling this complex mass and heat transfer problem for control design. It proposes a static model for predicting the deposited layer geometry of thin walls using process variables (wire feed speed, travel speed and contact tip to workpiece distance) and physical variables (arc power, inter-pass temperature) as inputs. The model is based on the well known Rosenthal solution for the temperature distribution due to a moving heat source in combination with a geometric parameterization of layer geometry. This layer geometry model is incorporated in a dynamic model of the GMAW process for control design and simulation. The proposed model is compared to experimental data produced by a GMAW power source using carbon steel filler wire for model parameter identification. Finally, PID controllers are proposed for the regulation of both layer height and wall width. Numerical simulation results illustrate the efficacy of the proposed control methods.