An Integrated Materials Informatics/Sequential Learning Framework to Predict the Effects of Defects in Metals Additive Manufacturing

In this project, Citrine Informatics and the ADAPT Center at the Colorado School of Mines propose to build an informatics-driven system to understand the effects of defects in additive manufactured parts. The entire history of each sample will be captured on this system; from specific printing parameters and details of precursor materials through to part characterizations and performance measurements. Researchers will be able to add new data, edit and append to records as new information is collected, search for data relevant to their projects, and visualize measurements across samples. The information that is collected will be used as input to machine learning models used to predict the defect properties and performance characteristics of parts before they are printed. The system that is developed in this project will be used to generate sets of printing parameters that will achieve targeted performance metrics through an understanding of the defects that are produced.