Intelligent Metal Deposition for Large Scale Part Fabrication

B. Jared
The University of Tennessee, Knoxville, Tennessee, United States

Poster stand number: T105

Keywords: metal deposition, automation, data analytics

The fabrication of large monolithic metal structures is a persistent challenge for America’s industrial base as delivery schedules are routinely defined in months and years; introducing unacceptable risk and cost for defense products. On-going research in wire-arc additive manufacturing, tungsten inert gas welding and hybrid manufacturing at the University of Tennessee (UTK) is addressing these challenges through the development of multiple technologies necessary for the intelligent deposition of high-quality metal materials. Different multi-axis robot systems are being used to provide automated deposition trajectories where-by complex geometries are generated using arc-welding torches oriented in both gravity and non-gravity aligned configurations. Process metrology on these systems provide multiple modalities as high-fidelity data from motion systems, arc conditions, boundary conditions, melt pool behavior and part geometry are captured in-situ and ex-situ. On-going investigations using advanced data analytics are exploring these temporal and spatial data streams to characterize underlying process physics, to develop process-structure material relationships and to identify critical signatures indicative of process, material and/or part anomalies. Through integration of these technologies, UTK is quantifying and leveraging process relationships to inform process advancements, to develop closed-loop process controls, and to establish autonomous, high-quality material deposition at melt pool, part and robot scales.