Giga-Voxel Multiscale Composite Architecture Mirrored Through a Data-to-Model Closed-Loop Digital Twin.

Area of Science:

• Materials Science
• Computational Modeling
• Digital Twins

Background:

• Giga-voxel digital models offer high geometric detail but lack efficient distribution methods.
• Designing complex anisotropic composite materials is challenging due to current methodological limitations.

Purpose of the Study:

• To propose a systematic digital twin workflow for generating high-fidelity virtual representations of anisotropic composite microstructures.
• To create giga-voxel meso-structural models using integrated top-down and bottom-up approaches.

Main Methods:

• Utilized a digital twin workflow combining image-based modeling and data-driven generation.
• Employed X-ray tomography to characterize geometric attributes from 3D image data.
• Created digital unit twins with unique identities via random seed generation and hierarchical assembly.

Main Results:

• Demonstrated the efficacy of micro-digital representations as building blocks for mesostructural models.
• Successfully generated hierarchical models at the giga-voxel level with assembly relationships across multiple length scales.
• Preserved intrinsic microstructural details within the micro-to-macro geometric insights.

Conclusions:

• The proposed workflow enables efficient generation of detailed giga-voxel models for anisotropic composites.
• Digital unit twins serve as foundational elements for large-scale, high-fidelity virtual material representations.
• This approach offers intricate insights into micro-to-macro geometrics while maintaining microstructural fidelity.