A physics-informed statistical framework for predicting polymer degradation trajectories under sparse data condition

doi:10.1016/j.polymdegradstab.2026.112254

A physics-informed statistical framework for predicting polymer degradation trajectories under sparse data condition

2026-06-01

0

PRE

AI

DOI：10.1016/j.polymdegradstab.2026.112254

Original

Original request for help

Save

Abstract

En 中文

• A new PiS model predicts polymer aging with very limited experimental data. • Multi-index transfer learning links diverse indicators to enhance prediction. • Combines physics knowledge with data-driven learning for better accuracy. • Reveals how climate factors jointly shape degradation in outdoor conditions. • Achieves less than 12% prediction error across diverse environments.

A physics-informed statistical framework for predicting polymer degradation trajectories under sparse data condition

Abstract

Journal

Organization