A framework for automated analysis and simulation of 3D polycrystalline microstructures.: Part 1: Statistical characterization

Abstract

This is the first of a two-part paper aimed at developing a robust framework for the collection, quantification and simulation of 3D polycrystalline microstructures. Serial-sectioning methods are used to generate data that characterize the microstructural morphology and crystallography of grains. The microstructure simulation model and codes take as input a series of electron backscatter diffraction (EBSD) patterns from the serial-sectioning experiments. Robust statistical analysis of the grain-level microstructures in 3D is conducted in this part of this paper. This analysis can provide necessary information for modeling and simulation efforts in the form of a highly refined and unbiased description of specific features, such as the distribution of grain size, shape and orientation.