SKM 2023 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

CPP: Fachverband Chemische Physik und Polymerphysik

CPP 48: Data Driven Materials Science: Big Data and Work Flows – Microstructure-Property-Relationships (joint session MM/CPP)

CPP 48.1: Talk

Thursday, March 30, 2023, 10:15–10:30, SCH A 251

Orisodata: A methodology for grain segementation in atomistic simulations using orientation based iterative self-organizing data analysis — •Arun Prakash — Micro-Mechanics and Multiscale Materials Modeling (M5), TU Bergakademie Freiberg

Atomistic simulations of the molecular statics/dynamics kind have established themselves as a cornerstone in the field of computational materials science. Large scale simulations with tens to hundreds of millions of atoms are regularly used to study the behavior of nano-(poly)crystalline materials. Identifying grains a posteriori in such simulations is a challenging task, particularly for simulations at high temperatures or at large strains. In this work, we propose a methodology for grain segmentation of atomistic configurations using unsupervised machine learning [1]. The proposed algorithm, called OrISODATA, is based on the iterative self-organizing data analysis technique and is modified to work in the orientation space. The algorithm is demonstrated on a 122 grain nanocrystalline thin film sample in both undeformed and deformed states. The Orisodata algorithm is also compared with two other grain segmentation algorithms available in open-source visualization tool Ovito. The results show that the Orisodata algorithm is able to correctly identify deformation twins as well as regions separated by low angle grain boundaries. The intiutive model parameters relate to similar thresholds in experiments, which helps obtain optimal values and facilitates easy interpretation of results.

References: [1]: M. Vimal, S. Sandfeld and A. Prakash [2022]: Materialia, 21, 101314