A Material Agnostic Physics-Aware, Interpretable AI Framework for Fatigue Life Prediction in TPMS Structures
Triply periodic minimal surface (TPMS) structures are mathematically defined structures with a geometry that enables lightweight materials that are exceptionally strong and provide large internal surface areas for heat transfer, chemical reactions and fluid flow. These properties may advance batteries, catalysts, aerospace, biomedical and automotive components. Reliable fatigue-life prediction of TPMS structures is challenging due to complex interactions among geometry, defects, surface roughness, and material properties, requiring extensive and costly experimental testing.
The project“A Material Agnostic Physics-Aware, Interpretable AI Framework for Fatigue Life Prediction in TPMS Structures,” led by Auburn University, will develop a physics-aware, interpretable AI framework that transfers fatigue knowledge from solid materials to TPMS structures while quantifying uncertainty. By integrating modeling, AI, active learning, additive manufacturing, and experimental validation, the approach could substantially reduce testing requirements and accelerate reliable TPMS design for energy and extreme-environment applications.
