PFI-TT: Ultrafast Thermal Simulation of Metal Additive Manufacturing

The broader impact/commercial potential of this Partnerships for Innovation - Technology Translation (PFI-TT) project is fast and accurate computer simulation software to predict when and why flaws are formed in metal parts made using additive manufacturing (3D printing). Given its singular design and material flexibility, metal additive manufacturing (metal AM) has the potential to revolutionize U.S. manufacturing by improving part performance and reducing waste and processing costs. However, safety-conscious industries, such as aerospace and biomedical, are hesitant to adopt AM processes due to the frequent occurrence of parts with hidden flaws. Traditional approaches for detecting and correcting flaws involve determining and adjusting the process parameters that lead to defects using a trial-and-error approach, which is expensive and time-consuming. This innovative project utilizes a computational simulation software to identify and correct design and processing problems before a part is printed. Importantly, this approach will provide scientific insights into why certain process parameters and part design features result in defect formation. This efficient and cost-effective method for detecting and correcting flaws in AM parts will enable their wide-spread commercialization and adoption. Ultimately, using AM processes rather than traditional manufacturing may save businesses time and resources while increasing part efficiency and reducing negative environmental impacts. This project will verify, validate, and commercialize a computational heat transfer modeling approach to simulate the temperature distribution in parts made using metal AM. This technology, which is based on the novel concept of heat diffusion on graphs (graph theory), aims to predict and correct design and processing problems before a part is printed. This capability would ultimately lead to improved AM part quality and increased use of AM processes in precision-critical industries. Existing simulation packages are expensive and incorporate proprietary assumptions. Non-proprietary approaches, in turn, take hours, if not days, to simulate the thermal history for a simple part. Prior work by the research team has demonstrated that the graph theory approach is approximately twenty times faster than non-proprietary methods and so computationally lightweight that it could be deployed on a laptop or smartphone. In moving toward commercializing the technology, the project team will employ practical use case samples produced by their industrial partners. The work will address two fundamental research questions: (1) What process conditions and part design features are linked to specific temperature patterns and why? (2) What is the influence of thermal history on flaw formation? The technical results from this project may include a rigorous, experimentally validated, computationally efficient, user-friendly, and industrially corroborated thermal simulation approach that can be used for rapid physics-based optimization of part design and process settings in metal AM. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.