Office of Naval Research Quality Made
Carnegie Mellon researchers are a part of a collaborative effort to develop a framework that couples modeling tools, in-situ process measurements, real-time closed-loop control, and machine learning to meet performance requirements for additive manufacturing (AM) parts in support of U.S. Navy plans to use AM to supply out-of-production and long lead time metal components.
Predictive modeling tools are being implemented for specific AM process variables and resulting microstructure predictions. Additional models for predicting mechanical properties based on microstructure characteristics are being verified with metallurgical testing and evaluations of Ti-6Al4V AM specimens. Sensors are to be selected and tested for in-situ process monitoring with the goal of implementing these in-situ data into a closed-loop control system. The end goal is a framework that can provide real-time monitoring of the build, identify material variability and potential anomalies, and capture a build process history.