We have already used this approach to identify several hundred new compounds We show that this approach is highly efficient in finding the ground states of binary metallic alloys and can be easily generalized to more complex systems. Knowledge is captured in a Bayesian probability network that relates the probability to find particular crystal structure at a given composition to structure and energy information at other compositions. An unusual, but efficient solution to this problem can be obtained by merging ideas from heuristic approaches and ab initio methods: In the same way that scientist build empirical rules by observation of experimental trends, we have developed machine learning approaches that extract knowledge from a large set of experimental information and a database of over 20,000 first principles computations, and used these to rapidly direct accurate quantum mechanical techniques to the lowest energy crystal structure of a material. Finding ground states by traditional optimization methods on quantum mechanical energy models is difficult due to the complexity and high dimensionality of the coordinate space. The premise of the approach is that for many materials chemistries standard computational quantum mechanics is highly accurate in selecting the true ground state of a system from a small set of candidate structures, though notable exceptions exist. We will present an ab initio approach that rapidly finds the stable crystal structure of materials with > 95% of success. Without detailed structure information the prediction of properties rapidly becomes irrelevant. The prediction of structure is a key problem in computational materials science that forms the platform on which rational materials design can be performed.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |