Machine learning-based accelerated property prediction of two-phase materials using microstructural descriptors and finite element analysis

This study explores the use of supervised machine learning (ML) to predict mechanical properties a family two-phase materials using their microstructural images. Random microstructures with diversity inclusion volume fractions, size distributions, and/or shapes are input into finite element analysis program determine elastic modulus, Poisson’s ratio, and phase stresses. The results establish “ground truth” train ML models. Two-point correlation (TPC) functions principal component (PC) applied before training testing artificial neural network (ANN) forest ensemble MLs. chosen methods found accurately homogenized properties. Although PCs for each set unique, recognizable patterns detected that signify features key microstructure-based property prediction. work enables development algorithms complex, multi-phase composites, based on