TY - JOUR
T1 - Fast inverse design of microstructures via generative invariance networks
AU - Lee, Xian Yeow
AU - Waite, Joshua R.
AU - Yang, Chih Hsuan
AU - Pokuri, Balaji Sesha Sarath
AU - Joshi, Ameya
AU - Balu, Aditya
AU - Hegde, Chinmay
AU - Ganapathysubramanian, Baskar
AU - Sarkar, Soumik
N1 - Publisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer Nature America, Inc.
PY - 2021/3
Y1 - 2021/3
N2 - The problem of the efficient design of material microstructures exhibiting desired properties spans a variety of engineering and science applications. The ability to rapidly generate microstructures that exhibit user-specified property distributions can transform the iterative process of traditional microstructure-sensitive design. We reformulate the microstructure design process using a constrained generative adversarial network (GAN) model. This approach explicitly encodes invariance constraints within GANs to generate two-phase morphologies for photovoltaic applications obeying design specifications: specifically, user-defined short-circuit current density and fill factor combinations. Such invariance constraints can be represented by differentiable, deep learning-based surrogates of full physics models mapping microstructures to photovoltaic properties. Furthermore, we propose a multi-fidelity surrogate that reduces expensive label requirements by a factor of five. Our framework enables the incorporation of expensive or non-differentiable constraints for the fast generation of microstructures (in 190 ms) with user-defined properties. Such proposed physics-aware data-driven methods for inverse design problems can be used to considerably accelerate the field of microstructure-sensitive design.
AB - The problem of the efficient design of material microstructures exhibiting desired properties spans a variety of engineering and science applications. The ability to rapidly generate microstructures that exhibit user-specified property distributions can transform the iterative process of traditional microstructure-sensitive design. We reformulate the microstructure design process using a constrained generative adversarial network (GAN) model. This approach explicitly encodes invariance constraints within GANs to generate two-phase morphologies for photovoltaic applications obeying design specifications: specifically, user-defined short-circuit current density and fill factor combinations. Such invariance constraints can be represented by differentiable, deep learning-based surrogates of full physics models mapping microstructures to photovoltaic properties. Furthermore, we propose a multi-fidelity surrogate that reduces expensive label requirements by a factor of five. Our framework enables the incorporation of expensive or non-differentiable constraints for the fast generation of microstructures (in 190 ms) with user-defined properties. Such proposed physics-aware data-driven methods for inverse design problems can be used to considerably accelerate the field of microstructure-sensitive design.
UR - http://www.scopus.com/inward/record.url?scp=85108284376&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85108284376&partnerID=8YFLogxK
U2 - 10.1038/s43588-021-00045-8
DO - 10.1038/s43588-021-00045-8
M3 - Article
AN - SCOPUS:85108284376
SN - 2662-8457
VL - 1
SP - 229
EP - 238
JO - Nature Computational Science
JF - Nature Computational Science
IS - 3
ER -