Learning Physics-Informed Surrogate Model of Linear Elastic Displacement Fields from Geometry
2026-07-10 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a quick and accurate model to help monitor cracks in elastic materials in real time. Their model, based on a physics-informed neural network called DeepONet, predicts how the material moves when forces and crack shapes change, without needing lots of simulation data. They carefully handle conditions on the crack surface using a special penalty method. They tested their approach on one type of crack shape and showed it works, aiming to expand it to many crack types in the future.
structural health monitoringfractured elastic domainsurrogate modelDeepONetphysics-informed neural networkdisplacement fieldboundary conditionstraction-free conditionpenalty methodfinite element method
Authors
Rodolphe Barlogis, Ferhat Tamssaouet, Quentin Falcoz, Stéphane Grieu
Abstract
This work aims to develop a fast and physically consistent surrogate model for real-time structural health monitoring of fractured elastic domains. We propose a physics-informed DeepONet framework that predicts displacement fields from both boundary conditions and fracture geometry, using a dedicated encoding strategy for the latter and without relying on finite-element-generated training data. The traction-free condition on the fracture boundary is imposed weakly through a localized penalty term. The presented numerical example focuses on one representative fracture geometry, demonstrating the feasibility of the formulation and laying the groundwork for extensions to surrogate modeling across diverse fracture geometries.