Applications of Data Driven Methods in Computational Materials Design

Abstract

In today’s digitized world, large amounts of data are becoming available at rates never seen before. This holds true also for materials science where high-throughput simulations and experiments continuously produce new data. Data driven methods are required which can make best use of the information stored in large data repositories. In the present article, two of such data driven methods are presented. First, we apply machine learning to generalize and extend the results obtained from computationally intense density functional theory (DFT) simulations. We show how grain boundary segregation energies can be trained with gradient boosting regression and extended to many more positions in the grain boundary for a complete description. The second method relies on Bayesian inference, which can be used to calibrate models to give data and quantification of the model uncertainty. The method is applied to calibrate parameters in thermodynamic models of the Gibbs energy of Ti-W alloys. The uncertainty of the model parameters is quantified and propagated to the phase boundaries of the Ti-W system.

Probing the composition dependence of residual stress distribution in tungsten-titanium nanocrystalline thin films

“…4 , we carried out density functional theory (DFT) simulations of the Ti segregation the Σ3[110]( 11) grain boundary (GB) that is representative for high angle and high energy GBs 47 – 49 . Although the real coating contains many different grain boundary orientations which may exhibit relative differences in GB segregation state, we expect that the general segregation trend reported below remains the same because detailed analysis of segregation at various microstructure defects shows only small changes in the distribution of segregation energies 24 , 50 .…”

“…Machine learning (ML)-driven approaches foster extended possibilities with respect to automated acquisition and analysis 21 – 23 . Recently, remarkable results have been shown for instance with data-based prediction models 21 – 24 .…”