Graph-Based Stochastic Learning to Build Physics-Informed AI for Discovering Defect Mechanisms in Powder-based Fabrication

Abstract

The proposed research aims to integrate computational material sciences, multi-physics simulations, statistical modeling, in-situ sensing, and Graphnets to create physics-informed AI software that accelerates the discovery of defect formation. This project will establish a theoretical framework to study combinatorial generalization problems (CG) that can empower AI to make inferences on new scenarios given piecewise knowledge learned from multiple data sources, such as multi-physics simulation and experimental data. The research will develop theories and algorithms based on physics-informed Graphnets and generative AI to solve CG problems in three thrust areas, helping discover unmeasurable defect mechanisms in powder-based fabrication.

Document Details

Document Type
DoD Grant Award
Publication Date
Mar 14, 2024
Source ID
FA95502310739

Entities

People

  • Hui Wang

Organizations

  • Air Force Office of Scientific Research
  • Florida A&M University
  • United States Air Force

Tags

Fields of Study

  • Physics

Readers

  • Computational Fluid Dynamics (CFD)
  • Defense Technology Research and Development.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML