RII Track-4: Understanding the Fundamental Thermal Physics in Metal Additive Manufacturing and its Influence on Part Microstructure and Distortion.

点击翻译按钮获取中文摘要

The 3D printing of metal parts promises to transform U.S. manufacturing. For example, metal additive manufacturing (AM) has the potential to reduce time-to-market for a new jet engine from five years to one year, while simultaneously increasing fuel efficiency and power by 10%. Poor consistency in part quality, however, limits the use of AM. As a result, safety-conscious industries (e.g., aerospace and biomedical fields) are reluctant to use AM processes to make mission-critical parts. The root cause for flaw formation in metal AM is the uneven temperature distribution inside the part during printing. To ensure a steady temperature distribution inside the part, practitioners currently use trial-and-error studies that require experimenting with different process settings and part designs – an expensive and time-consuming approach. A more efficient solution involves encapsulating the fundamental thermal physics of the printing process using computer simulation models. These simulation models can be used to identify and correct problems that can lead to an uneven temperature distribution in the part before it is built. The PI has advanced a new mathematical approach to predict the temperature distribution in AM parts that takes less than one-tenth of the time required by existing techniques and has an error of less than 10%. Rigorous validation of this concept with experimental data is the next step to scale this new concept to practice. The objective of this fellowship is to test the hypothesis that the instantaneous spatiotemporal distribution of temperature generated in a metal AM part as it is being deposited layer-upon-layer is predicted by invoking the novel theory of heat dissipation on planar graphs (spectral graph theory) with an accuracy comparable to existing finite element techniques but within a fraction of the computation time (less than 1/10th). To realize this objective, this fellowship provides the PI access to the Open Architecture Laser Powder Bed Fusion metal AM system at the Edison Welding Institute (EWI). This system has eight different sensors and allows the in-situ measurement of thermal signatures at scales ranging from 5 micrometer to 400 micrometers. Access to this unique apparatus will allow the PI to measure the instantaneous temperature distribution in a part and track changes in its shape with unprecedented precision. Using data obtained from experiments on the open architecture metal AM system at EWI, the PI will: (1) explain and an quantify the causal factors governing the temperature distribution in metal AM parts and link it to part quality; (2) achieve near real-time prediction of the temperature distribution, which will significantly reduce the experimental tests needed to optimize the part geometry and process parameters; and (3) establish the digital twin concept for qualification of metal AM parts by augmenting in-situ sensor data with physical process models. This work will result in experimentally validated, physics-based tools to aid rapid optimization of process settings and part geometry, which in turn will shorten time-to-market for AM parts and reduce scrap rates by up to 80%.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.