Computational Methods for Reconstructing and Analyzing Polycrystalline Microstructures

The ability to predict the performance and behavior of metals and other polycrystals in different conditions has huge impacts in real life. Besides phenomenological

studies, finding the relationship between material microstructure and macroscopic properties is also indispensable, which relies on modern characterization techniques and complex computational tools. This thesis describes some recent developments of the computational tools for microstructure reconstruction and analysis. The grain coarsening phenomenon is studied in an alpha-phase Iron sample. We designed a new method to determine grain boundary energies (GBE) from triple junction geometries.

The synchrotron based imaging technique Near-field High Energy X-ray Diffraction Microscopy (nf-HEDM) is extended to reconstruct intragranular strains. Meanwhile, A neural network based method is developed for Electron Backscatter Diffraction (EBSD) indexing, which is another popular characterization technique