A single physics-informed CNN, trained on two load steps of one notched specimen, can replace the phase-field equation inside a FEM loop and reproduce crack propagation across several unseen geometries and load steps.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CE 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Integrated Finite Element Neural Network (IFENN) for Phase-Field Fracture with Minimal Input and Generalized Geometry-Load Handling
A single physics-informed CNN, trained on two load steps of one notched specimen, can replace the phase-field equation inside a FEM loop and reproduce crack propagation across several unseen geometries and load steps.