Senvol (New York, NY) will demonstrate that data analytics can be applied to additive manufacturing (AM) data to establish Process-Structure-Property (PSP) relationships. Senvol ML, the company's data-driven machine learning software for AM, will be used to conduct the analyses. The data to be analyzed will come from physical sciences laboratory and non-regulatory agency NIST's (Gaithersburg, MD) various round-robin test studies, as well as from its AM Benchmark Test Series.
The software's capabilities to be utilized include model reliability, adaptive sampling, generative learning, hybrid modeling, and transfer learning. Additionally, Senvol will parameterize in situ monitoring data, non-destructive testing (NDT) data, and microstructure data so that these types of data can be incorporated into NIST's AM Material Database (AMMD). The project will culminate with an integration between Senvol ML and AMMD such that data stored within AMMD can be seamlessly analyzed by Senvol's machine learning software.
"The work in this project will demonstrate the power of a data-driven machine learning approach for additive manufacturing process understanding and material characterization," says Yan Lu, senior research scientist at NIST. "Furthermore, Senvol will showcase hybrid modeling, whereby physics-based models and data-driven models are joined under a single framework."
NIST's mission is to promote innovation and industrial competitiveness. Its activities are organized into laboratory programs that include nanoscale science and technology, engineering, information technology, neutron research, material measurement, and physical measurement.