A Process Planning Software for Modeling Parameter Behavior in Automated Fiber Placement

The manufacturing of large-scale, geometrically complex composite structures is often accomplished today using the Automated Fiber Placement (AFP) process. AFP utilizes a fiber placement end effector and a gantry or robotic kinematic system to lay up groups of composite tows, iteratively building the complete structure. The reduced width of each tow allows the deployment of AFP for builds with significant curvature, unlike other automated methods such as automated tape laying. Despite its proven track record and widespread use, the current AFP process contains inefficiencies and suffers from workflow bottlenecks that significantly increase cycle time, material wastage, and overall cost. A primary example of this is the process parameter selection phase, where engineers utilize experimental trials and legacy-based configurations to pinpoint optimal process parameters for new builds. The novel strategies afforded through Industry 4.0 technologies provide tools to help perform this phase digitally, in turn eliminating the aforementioned inefficiencies. As such, this work presents a virtual simulation tool to model compaction pressure and heat application behavior during AFP manufacturing. A preliminary literature review was first conducted to establish foundations from various modeling strategies and assess frequent gaps in current research. Theoretical approaches for each process parameter model are then described. This consists of physics based approaches through Hertzian contact theory for compaction pressure and the finite difference method for heat dissipation through the substrate, as well as data-driven approaches for effective contact stiffness and surface temperature predictions. Next, experimental procedures are outlined for both capturing necessary data for model construction and validation. The structure and hierarchy of the digital platform for AFP process simulation, SmartPath , is presented as well along with specific augmentations to implement each model into SmartPath. This tool is demonstrated via specific virtual case studies for the compaction pressure, applied heating, and through thickness models. Finally, validation results for each model are presented, including error analysis and hypothesized correction efforts. Overall, the SmartPath tool serves as an initial prototype for providing a comprehensive, digital solution for AFP process parameter selection.