An efficient optimization based microstructure reconstruction approach with multiple loss functions

TL;DR

This paper presents an efficient optimization-based microstructure reconstruction method that combines statistical descriptors and deep neural network features to accurately replicate material properties with reduced computational cost.

Contribution

It introduces a novel loss function integrating statistical and deep learning features for microstructure reconstruction, improving efficiency and accuracy.

Findings

Achieves significant computational efficiency in microstructure reconstruction.

Successfully captures key physical properties of target microstructures.

Demonstrates potential for extension to 3D microstructure reconstruction.

Abstract

Stochastic microstructure reconstruction involves digital generation of microstructures that match key statistics and characteristics of a (set of) target microstructure(s). This process enables computational analyses on ensembles of microstructures without having to perform exhaustive and costly experimental characterizations. Statistical functions-based and deep learning-based methods are among the stochastic microstructure reconstruction approaches applicable to a wide range of material systems. In this paper, we integrate statistical descriptors as well as feature maps from a pre-trained deep neural network into an overall loss function for an optimization based reconstruction procedure. This helps us to achieve significant computational efficiency in reconstructing microstructures that retain the critically important physical properties of the target microstructure. A numerical example for the microstructure reconstruction of bi-phase random porous ceramic material demonstrates the efficiency of the proposed methodology. We further perform a detailed finite element analysis (FEA) of the reconstructed microstructures to calculate effective elastic modulus, effective thermal conductivity and effective hydraulic conductivity, in order to analyse the algorithm's capacity to capture the variability of these material properties with respect to those of the target microstructure. This method provides an economic, efficient and easy-to-use approach for reconstructing random multiphase materials in 2D which has the potential to be extended to 3D structures.