Agentic data science with the Data Agent Kit
cloud.google.com·7h ago
AI Research
arXiv CS.AI·Sep 7·5 min read
Data-Driven Discovery of Composition-Dependent Constitutive Models for Hyperelasticity and Viscoelasticity of Digital Materials
TL;DR
A data-driven framework was developed to create composition-dependent constitutive models for hyperelastic and viscoelastic behaviors in digital materials. This approach effectively captures rate-dependent stiffness and hysteresis across various material compositions while maintaining thermodynamic consistency.
✦ Why It Matters
Consider adopting this data-driven framework to enhance your material modeling capabilities in 3D printing applications.
Key Takeaways
- Explore the use of neural ordinary differential equations in material modeling.
- Utilize the framework for predicting material behavior across compositions.
- Implement the model to capture hysteresis in digital materials.
- Adopt this approach to improve the design of multi-material 3D printed components.
How It Works
The framework employs neural ordinary differential equations (NODEs) to derive strain-energy functions and model parameters based on the composition of materials. This allows for a flexible representation of both equilibrium and nonequilibrium behaviors, capturing the complex interactions between different material constituents.