Contributions à la modélisation géométrique du dépôt de matière pour l'élaboration d'un jumeau numérique du procédé WAAM

Wire and arc additive manufacturing processes are rising as they offer high deposition rates and minimize environmental and security issues due to the use of metal powders. Even if the process derived from welding is well-known, its use on robot to manufacture large parts or add features by refilling needs to elaborate manufacturing strategies to ensure process efficiency while minimizing deformations. The development of such strategies requires an in-depth knowledge of the manufacturing process, but also of the properties of the manufactured parts in order to move towards large-scale industrialisation. This thesis work focuses on the modelling of the WAAM process and the arc wire deposition process in order to contribute to the development of a digital twin of the production system. The implementation of such a twin aims to improve and control the geometry of the manufactured parts. The scientific approach is based on experimental campaigns to observe and characterise the influence of process parameters on the geometric behaviour of single beads manufactured on a flat substrate. Then, this study is extended to corners where the evolution and the behaviour of the bead pool is observed and modelled with the kinematic behaviour of the robot. Finally, the assembly of beads is studied through the manufacturing of simple geometric entities in order to establish a model allowing the prediction and simulation of the manufactured geometry. This model allows a better understanding of the essential parameters of the tool path such as the layer height and the inter-bead distance in order to limit geometric drift, guarantee feasibility and avoid any collision. All this work contributes to the development of a digital twin of the WAAM process, whose functional architecture is proposed based on concepts from the literature. The research work is part of the Ile-de-France additive manufacturing platform "Additive Factory Hub" involving industrial and academic actors.