Neural inference of fluid-structure interactions from sparse off-body measurements

Abstract

We report a novel physics-informed neural framework for reconstructing unsteady fluid-structure interactions (FSI) from sparse, single-phase observations of the flow. Our approach combines a modal surface model with coordinate neural representations of the fluid and solid states, constrained by the fluid's governing equations and interface conditions. Using only off-body Lagrangian particle tracks and a moving-wall boundary condition, the method infers both flow fields and structural motion. It does not require a constitutive model for the solid or measurements of surface position, although including these can improve performance. We demonstrate the approach numerically on two canonical FSI benchmarks: vortex-induced oscillations of a 2D flapping plate and pulse-wave propagation in a 3D flexible pipe. We also demonstrate it on flow around a swimming fish. In all cases, the framework achieves accurate reconstructions of flow states and structural deformations despite acute data sparsity near the moving interface. A key result is that reconstructions remain robust to over-parameterization. This work extends physics-informed neural networks to coupled fluid-structure dynamics learned from single-phase observations, and it provides a pathway toward quantitative FSI analysis when flow measurements are sparse and structural measurements are asynchronous or unavailable.

References in corpus (9)

• Deep Learning of Vortex Induced Vibrations
• Machine Learning Holography for 3D Particle Field Imaging
• Variational Physics Informed Neural Networks: the role of quadratures and test functions
• A unified continuum and variational multiscale formulation for fluids, solids, and fluid-structure interaction
• Optimizing a DIscrete Loss (ODIL) to solve forward and inverse problems for partial differential equations using machine learning tools
• Regularized Inverse Holographic Volume Reconstruction for 3D Particle Tracking
• Physics-informed neural networks for unsteady incompressible flows with time-dependent moving boundaries
• Physics-informed neural networks modeling for systems with moving immersed boundaries: application to an unsteady flow past a plunging foil
• Joint reconstruction and segmentation of noisy velocity images as an inverse Navier-Stokes problem