An adaptive physics-informed graph neural network-based method for aero-engine gas path estimation

doi:10.1016/j.ast.2026.113604

An adaptive physics-informed graph neural network-based method for aero-engine gas path estimation

2026-08-22

0

PRE

AI

DOI：10.1016/j.ast.2026.113604

Original

Original request for help

Save

Abstract

En 中文

• Customized architecture tailored for aero-engine characteristics. • Directed topology graphs encode physical information explicitly. • Explicit memory units preserve critical temporal information. • Different flight phases are processed through separate channels. • Efficient parameter updates mitigate performance degradation.

An adaptive physics-informed graph neural network-based method for aero-engine gas path estimation

Abstract

Journal

Organization