News & Highlights
New approach to accelerate complex chemical simulations
Machine-learning accelerates development of material microstructure evolution model
CMI postdoc Tim Hsu becomes LLNL staff member
Surrogate models for microstructure simulation
New “gold-standard” classification of atomic structure
Fei Zhou at Lawrence Livermore National Laboratory leads the CMI project "Machine learning materials design"
The goal of this project is to design and develop advanced alloy materials through computational modeling and analysis enabled by machine learning (ML), algorithms, and tools. In particular, ML tools will be developed and applied to accelerate the optimization of thermomechnical properties of cerium-aluminum alloys, a break-through technology with important impact on the U.S. supply chain for rare earths. ML models will be trained on large amount of experimental and first-principles data in order to make computational suggestions regarding additional dopants and new compositions for cerium-aluminum alloys with improved mechanical properties and reduced manufacturing cost.