Symposium QT03: Higher-Order Topological Structures in Real Space—From Charge to Spin
Symposium DS06: Integrating Machine Learning with Simulations for Accelerated Materials Modeling

Symposium DS06: Integrating Machine Learning with Simulations for Accelerated Materials Modeling

Pinar Acar, Virginia Tech

Materials Informatics for Computational and Machine Learning (ML)-Assisted Design: An Overview for Polycrystalline Metals and Mechanical Metamaterials

Written by Matthew Nakamura

Pinar Acar of Virginia Tech provided a comprehensive overview of computational methods developed by her group for optimizing metals and metamaterials at the micro-scale. The presentation began by outlining numerical approaches to assess the crystallographic texture and grain topology of polycrystalline metals, alongside a shape descriptor method for modeling mechanical metamaterials. These computational characterization techniques were seamlessly integrated into homogenization schemes for deriving mechanical properties. Acar then delved into the challenges posed by manufacturing-related uncertainties and defects, emphasizing the importance of design under uncertainty formulations. Acar discussed strategies for addressing forward and inverse design problems to enhance the elasto-plastic properties of materials. Notably, she talked about the integration of artificial intelligence/machine learning techniques into physics-informed materials models for accelerating design processes, showcasing applications in both conventional and additive manufacturing. The talk concluded with demonstrations of ML-driven design approaches for polycrystalline metals and mechanical metamaterials.

Comments

The comments to this entry are closed.