A Machine Learning-Based Design Representation Method for Designing Heterogeneous Microstructures

Abstract

In designing microstructural materials systems, one of the key research questions is how to represent the microstructural design space quantitatively using a descriptor set that is sufficient yet small enough to be tractable. Existing approaches describe complex microstructures either using a small set of descriptors that lack sufficient level of details, or using generic high order microstructure functions of infinite dimensionality without explicit physical meanings. We propose a new machine learning-based method for identifying the key microstructure descriptors from vast candidates as potential microstructural design variables. With a large number of candidate microstructure descriptors collected from literature covering a wide range of microstructural material systems, a four-step machine learning-based method is developed to eliminate redundant microstructure descriptors via image analyses, to identify key microstructure descriptors based on structure–property data, and to determine the microstructure design variables. The training criteria of the supervised learning process include both microstructure correlation functions and material properties. The proposed methodology effectively reduces the infinite dimension of the microstructure design space to a small set of descriptors without a significant information loss. The benefits are demonstrated by an example of polymer nanocomposites optimization. We compare designs using key microstructure descriptors versus using empirically chosen microstructure descriptors as a demonstration of the proposed method.

Exploring the 3D architectures of deep material network in data-driven multiscale mechanics

“…Homogenization function h C . The analytical form of homogenization function h C is derived based on the interfacial equilibrium conditions 7) and the kinematic constraints…”

“…In aerospace and automotive industries, carbon fiber reinforced polymer (CFRP) composites have become attractive alternatives of metal materials due to their high strength/weight ratio induced by the interplay between carbon fibers and epoxy matrix [2,3,4]. Another typical example of heterogeneous materials is the particlereinforced rubber composite where nanoparticles are added into the matrix to manipulate the overall mechanical properties, such as stiffness and viscoelastic behaviors [5,6,7,8]. Moreover, in metal additive manufacturing, the performance of the final product is strongly affected by the microscopic polycrystalline microstructure, which could be controlled by the deposition and cooling processes [9,10].…”

Machine learning framework for analysis of transport through complex networks in porous, granular media: A focus on permeability

“…Ma et al [70] presented a machine learning framework to classify and predict fluid flow properties of stochastically reconstructed rocks, studying the relationships with geometry, topology and statistical correlation functions. Xu et al [71] predict the damping parameters of polymer nanocomposites, using correlation functions, particle shape descriptors and pore size descriptors. Feature selection for materials science has been explored by Ghiringhelli et al [72].…”

A computer vision approach for automated analysis and classification of microstructural image data

“…A classic example is the characterization of phase volume fraction in two-phase systems. Recent progress has focused on quantifying the phase morphology of fully segmented two-phase systems using higher order point statistics [2]; moment invariant shape descriptors [23,24]; and, more recently, sets of shape descriptors and shape correlation functions that are tailored to particular microstructure systems through a machine learning approach [22].…”

“…Likewise, although npoint correlation functions are theoretically able to exactly represent a microstructure, in practice the computational cost increases exponentially with each additional pixel state; performing even 2-point correlation analysis on images with 256 grayscale values is prohibitive. Therefore, these methods are typically applied only to a pre-selected set of micrographs, chosen by a human expert [2,3,6,25,26,22].…”