Microstructural Materials Design Via Deep Adversarial Learning Methodology

TLDR

Identifying key microstructure representations is crucial for computational materials design, yet existing microstructure characterization and reconstruction techniques lack suitable design variables or introduce significant information loss. The study introduces a deep adversarial learning methodology to overcome limitations of existing microstructure characterization and reconstruction techniques. Generative adversarial networks learn a mapping from latent variables to microstructures, and the resulting low‑dimensional latent variables serve as design variables that are optimized via Bayesian optimization to produce microstructures with targeted material properties, while the network architecture captures complex microstructural characteristics. Numerical tests on a synthetic dataset and a case study optimizing optical performance for energy absorption demonstrate the method’s validity, scalability, transferability, and its end‑to‑end capability to preserve microstructural detail and improve design exploration efficiency.

Abstract

Identifying the key microstructure representations is crucial for computational materials design (CMD). However, existing microstructure characterization and reconstruction (MCR) techniques have limitations to be applied for microstructural materials design. Some MCR approaches are not applicable for microstructural materials design because no parameters are available to serve as design variables, while others introduce significant information loss in either microstructure representation and/or dimensionality reduction. In this work, we present a deep adversarial learning methodology that overcomes the limitations of existing MCR techniques. In the proposed methodology, generative adversarial networks (GAN) are trained to learn the mapping between latent variables and microstructures. Thereafter, the low-dimensional latent variables serve as design variables, and a Bayesian optimization framework is applied to obtain microstructures with desired material property. Due to the special design of the network architecture, the proposed methodology is able to identify the latent (design) variables with desired dimensionality, as well as capturing complex material microstructural characteristics. The validity of the proposed methodology is tested numerically on a synthetic microstructure dataset and its effectiveness for microstructural materials design is evaluated through a case study of optimizing optical performance for energy absorption. Additional features, such as scalability and transferability, are also demonstrated in this work. In essence, the proposed methodology provides an end-to-end solution for microstructural materials design, in which GAN reduces information loss and preserves more microstructural characteristics, and the GP-Hedge optimization improves the efficiency of design exploration.