- Physical Review Journals
All JournalsPhysics Magazine
Access Provided by Xinjiang University

Physical Review Fluids

Export citation

Export citation

Choose format for download:

Download Citation
  • Access by Xinjiang University

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

Rui Tang1, Ke Zhou1, Jifu Tan2, and Samuel J. Grauer1,*

  • *Contact author: sgrauer@psu.edu
PDF

Phys. Rev. Fluids 11, 074901 – Published 13 July, 2026

DOI: https://doi.org/10.1103/k48m-pmjz

Abstract

We report a 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 two-dimensional (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 overparameterization. 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.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (66)

  1. T. Nakata and H. Liu, A fluid–structure interaction model of insect flight with flexible wings, J. Comput. Phys. 231, 1822 (2012).
  2. H. Zhu, Q. Sun, X. Liu, J. Liu, H. Sun, W. Wu, P. Tan, and Z. Chen, Fluid–structure interaction-based aerodynamic modeling for flight dynamics simulation of parafoil system, Nonlin. Dynam. 104, 3445 (2021).
  3. A. Calderer, X. Guo, L. Shen, and F. Sotiropoulos, Fluid–structure interaction simulation of floating structures interacting with complex, large-scale ocean waves and atmospheric turbulence with application to floating offshore wind turbines, J. Comput. Phys. 355, 144 (2018).
  4. A. Korobenko, M.-C. Hsu, I. Akkerman, J. Tippmann, and Y. Bazilevs, Structural mechanics modeling and FSI simulation of wind turbines, Math. Models Methods Appl. Sci. 23, 249 (2013).
  5. H. Suito, K. Takizawa, V. Q. Huynh, D. Sze, and T. Ueda, FSI analysis of the blood flow and geometrical characteristics in the thoracic aorta, Comput. Mech. 54, 1035 (2014).
  6. A. H. Lee, R. L. Campbell, B. A. Craven, and S. A. Hambric, Fluid–structure interaction simulation of vortex-induced vibration of a flexible hydrofoil, J. Vib. Acoust. 139, 041001 (2017).
  7. J. Donea, S. Giuliani, and J.-P. Halleux, An arbitrary Lagrangian-Eulerian finite element method for transient dynamic fluid-structure interactions, Comput. Methods Appl. Mech. Eng. 33, 689 (1982).
  8. T. J. Hughes, W. K. Liu, and T. K. Zimmermann, Lagrangian-Eulerian finite element formulation for incompressible viscous flows, Comput. Methods Appl. Mech. Eng. 29, 329 (1981).
  9. A. Legay, J. Chessa, and T. Belytschko, An Eulerian–Lagrangian method for fluid–structure interaction based on level sets, Comput. Methods Appl. Mech. Eng. 195, 2070 (2006).
  10. N. Jenkins and K. Maute, Level set topology optimization of stationary fluid-structure interaction problems, Struct. Multidiscip. Optim. 52, 179 (2015).
  11. Y. Yu, H. Baek, and G. E. Karniadakis, Generalized fictitious methods for fluid–structure interactions: Analysis and simulations, J. Comput. Phys. 245, 317 (2013).
  12. A. Pathak and M. Raessi, A 3D, fully Eulerian, VOF-based solver to study the interaction between two fluids and moving rigid bodies using the fictitious domain method, J. Comput. Phys. 311, 87 (2016).
  13. C. S. Peskin, The immersed boundary method, Acta Numer, 11, 479 (2002).
  14. F. Sotiropoulos and X. Yang, Immersed boundary methods for simulating fluid–structure interaction, Prog. Aerosp. Sci. 65, 1 (2014).
  15. G. Hou, J. Wang, and A. Layton, Numerical methods for fluid-structure interaction—A review, Commun. Comput. Phys. 12, 337 (2012).
  16. S. Haeri and J. Shrimpton, On the application of immersed boundary, fictitious domain and body-conformal mesh methods to many particle multiphase flows, Int. J. Multiphase Flow 40, 38 (2012).
  17. L. Chatellier, S. Jarny, F. Gibouin, and L. David, A parametric PIV/DIC method for the measurement of free surface flows, Exp. Fluids 54, 1488 (2013).
  18. R. Bleischwitz, R. De Kat, and B. Ganapathisubramani, On the fluid-structure interaction of flexible membrane wings for MAVs in and out of ground-effect, J. Fluids Struct. 70, 214 (2017).
  19. P. Zhang, S. D. Peterson, and M. Porfiri, Combined particle image velocimetry/digital image correlation for load estimation, Exp. Therm Fluid Sci. 100, 207 (2019).
  20. R. Hortensius, J. C. Dutton, and G. S. Elliott, Simultaneous planar PIV and sDIC measurements of an axisymmetric jet flowing across a compliant surface, in 55th AIAA Aerospace Sciences Meeting (AIAA, Reston, VA, 2017), p. 1886.
  21. A. D'Aguanno, P. Quesada Allerhand, F. F. J. Schrijer, and B. W. van Oudheusden, Characterization of shock-induced panel flutter with simultaneous use of DIC and PIV, Exp. Fluids 64, 15 (2023).
  22. W. I. Kösters and S. Hoerner, Simultaneous flow measurement and deformation tracking for passive flow control experiments involving fluid–structure interactions, J. Fluids Struct. 121, 103956 (2023).
  23. F. M. Mitrotta, J. Sodja, and A. Sciacchitano, On the combined flow and structural measurements via robotic volumetric PTV, Meas. Sci. Technol. 33, 045201 (2022).
  24. A. Schröder and D. Schanz, 3D Lagrangian particle tracking in fluid mechanics, Annu. Rev. Fluid Mech. 55, 511 (2023).
  25. C. J. Elkins and M. T. Alley, Magnetic resonance velocimetry: Applications of magnetic resonance imaging in the measurement of fluid motion, Exp. Fluids 43, 823 (2007).
  26. A. Kontogiannis, S. V. Elgersma, A. J. Sederman, and M. P. Juniper, Joint reconstruction and segmentation of noisy velocity images as an inverse Navier–Stokes problem, J. Fluid Mech. 944, A40 (2022).
  27. P. Karnakov, S. Litvinov, and P. Koumoutsakos, Solving inverse problems in physics by optimizing a discrete loss: Fast and accurate learning without neural networks, PNAS Nexus 3, pgae005 (2023).
  28. A. B. Buhendwa, D. A. Bezgin, P. Karnakov, N. A. Adams, and P. Koumoutsakos, Data-driven shape inference in three-dimensional steady-state supersonic flows Optimizing a discrete loss with JAX-Fluids, Phys. Rev. Fluids 10, 084902 (2025).
  29. M. Raissi, P. Perdikaris, and G. E. Karniadakis, Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, J. Comput. Phys. 378, 686 (2019).
  30. M. Raissi, Z. Wang, M. S. Triantafyllou, and G. E. Karniadakis, Deep learning of vortex-induced vibrations, J. Fluid Mech. 861, 119 (2019).
  31. E. Kharazmi, D. Fan, Z. Wang, and M. S. Triantafyllou, Inferring vortex induced vibrations of flexible cylinders using physics-informed neural networks, J. Fluids Struct. 107, 103367 (2021).
  32. H. Tang, Y. Liao, H. Yang, and L. Xie, A transfer learning-physics informed neural network (TL-PINN) for vortex-induced vibration, Ocean Eng. 266, 113101 (2022).
  33. Y. Zhu, W. Kong, J. Deng, and X. Bian, Physics-informed neural networks for incompressible flows with moving boundaries, Phys. Fluids 36, 013617 (2024).
  34. R. Sundar, D. Majumdar, D. Lucor, and S. Sarkar, Physics-informed neural networks modelling for systems with moving immersed boundaries: Application to an unsteady flow past a plunging foil, J. Fluids Struct. 125, 104066 (2024).
  35. M. A. Calicchia, R. Mittal, J.-H. Seo, and R. Ni, Reconstructing the pressure field around swimming fish using a physics-informed neural network, J. Exp. Biol. 226, jeb244983 (2023).
  36. H. Wang, F. Wu, Y. Liu, X. He, S. Feng, and S. Wang, Machine-learning-based pressure reconstruction with moving boundaries, J. Fluid Mech. 1008, A21 (2025).
  37. Y. Zhu, W. Chen, J. Deng, and X. Bian, Physics-informed neural networks for hidden boundary detection and flow field reconstruction, Acta Mech. Sin. 42, 725273 (2026).
  38. K. Zhou and S. J. Grauer, On the joint estimation of flow fields and particle properties from Lagrangian data, arXiv:2510.00479 [physics.flu-dyn].
  39. K. Zhou, R. Tang, G. Ke, and S. J. Grauer, Neural-implicit particle advection for flow reconstruction from Lagrangian tracks, in 16th International Symposium on Particle Image Velocimetry (ISPIV 2025 Organizing Committee, Tokyo, Japan, 2025), p. 24.
  40. S. Wang, Y. Teng, and P. Perdikaris, Understanding and mitigating gradient flow pathologies in physics-informed neural networks, SIAM J. Sci. Comput. 43, A3055 (2021).
  41. M. Tancik, P. P. Srinivasan, B. Mildenhall, S. Fridovich-Keil, N. Raghavan, U. Singhal, R. Ramamoorthi, J. T. Barron, and R. Ng, Fourier features let networks learn high frequency functions in low dimensional domains, in Proceedings of the 34th International Conference on Neural Information Processing Systems, edited by H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, and H. Lin (Curran Associates, Inc., Red Hook, NY, 2020), pp. 7537–7547.
  42. J. M. Lee, Smooth Manifolds (Springer, Berlin, 2003).
  43. Y. Ma and Y. Fu, Manifold Learning Theory and Applications (CRC Press, Boca Raton, FL, 2012).
  44. J. M. Melenk and I. Babuška, The partition of unity finite element method: Basic theory and applications, Comput. Methods Appl. Mech. Eng. 139, 289 (1996).
  45. G. Berkooz, P. Holmes, and J. L. Lumley, The proper orthogonal decomposition in the analysis of turbulent flows, Annu. Rev. Fluid Mech. 25, 539 (1993).
  46. K. Taira, S. L. Brunton, S. T. Dawson, C. W. Rowley, T. Colonius, B. J. McKeon, O. T. Schmidt, S. Gordeyev, V. Theofilis, and L. S. Ukeiley, Modal analysis of fluid flows: An overview, AIAA J. 55, 4013 (2017).
  47. C. Leake, H. Johnson, and D. Mortari, The Theory of Functional Connections: A Functional Interpolation Framework with Applications (Lulu Press, Inc., Morrisville, NC, 2022).
  48. S. Berrone, C. Canuto, and M. Pintore, Variational physics informed neural networks: The role of quadratures and test functions, J. Sci. Comput. 92, 100 (2022).
  49. Z. Mao and X. Meng, Physics-informed neural networks with residual/gradient-based adaptive sampling methods for solving partial differential equations with sharp solutions, Appl. Math. Mech. 44, 1069 (2023).
  50. X. Wan, T. Zhou, and Y. Zhou, Adaptive importance sampling for deep Ritz, Commun. Appl Math. Comput. 7, 929 (2025).
  51. J. M. Taylor and D. Pardo, Stochastic quadrature rules for solving PDEs using neural networks, Comput. Math. Appl. 214, 1 (2026).
  52. J. P. Molnar and S. J. Grauer, Algorithm for time-resolved background-oriented Schlieren tomography applied to high-speed flows, in AIAA SciTech 2025 Forum (AIAA, Reston, VA, 2025), p. 1060.
  53. K. Hormann and N. Sukumar, Generalized Barycentric Coordinates in Computer Graphics and Computational Mechanics (CRC Press, Boca Raton, FL, 2017).
  54. S. Turek, J. Hron, M. Madlik, M. Razzaq, H. Wobker, and J. F. Acker, Numerical simulation and benchmarking of a monolithic multigrid solver for fluid-structure interaction problems with application to hemodynamics, in Fluid Structure Interaction II: Modelling, Simulation, Optimization, edited by H.-J. Bungartz, M. Mehl, and M. Schäfer (Springer, Berlin, 2010), pp. 193–220.
  55. J. Liu and A. L. Marsden, A unified continuum and variational multiscale formulation for fluids, solids, and fluid–structure interaction, Comput. Methods Appl. Mech. Eng. 337, 549 (2018).
  56. A. Updegrove, N. M. Wilson, J. Merkow, H. Lan, A. L. Marsden, and S. C. Shadden, SimVascular: An open source pipeline for cardiovascular simulation, Ann. Biomed Eng. 45, 525 (2017).
  57. Y. Pan and G. V. Lauder, Combining computational fluid dynamics and experimental data to understand fish schooling behavior, Integr. Comp. Biol. 64, 753 (2024).
  58. A. Menzer, Y. Pan, G. V. Lauder, and H. Dong, Fish schools in a vertical diamond formation: Effect of vertical spacing on hydrodynamic interactions, Phys. Rev. Fluids 10, 043104 (2025).
  59. D. Schanz, S. Gesemann, and A. Schröder, Shake-The-Box: Lagrangian particle tracking at high particle image densities, Exp. Fluids 57, 70 (2016).
  60. Particle-structure collisions in these simulations arise from numerical integration error (e.g., when the advection step is not sufficiently small near boundary layers) or from under-resolved velocity gradients near solid surfaces. We treat these events as a practical source of track error, analogous to the loss of visibility or tracking failures in experiments.
  61. This sampling rate is higher than in many laboratory systems; it is used here to resolve the short time window of pulse-wave propagation in this benchmark. The impact of the sampling rate is not explored in this study.
  62. K. Mallery and J. Hong, Regularized inverse holographic volume reconstruction for 3D particle tracking, Opt. Express 27, 18069 (2019).
  63. S. Shao, K. Mallery, S. S. Kumar, and J. Hong, Machine learning holography for 3D particle field imaging, Opt. Express 28, 2987 (2020).
  64. M. Moaven, A. Gururaj, V. Raghav, and B. Thurow, Improving depth uncertainty in plenoptic camera-based velocimetry, Exp. Fluids 65, 49 (2024).
  65. In principle, one could also infer the σi within a Bayesian formulation; we leave this for future work.
  66. J. P. Molnar and S. J. Grauer, Flow field tomography with uncertainty quantification using a Bayesian physics-informed neural network, Meas. Sci. Technol. 33, 065305 (2022).

Outline

Information

Phys. Rev. Fluids 11, 074901– Published 13 July, 2026
  • Received 29 June 2025
  • Accepted 9 June 2026
Article part of CHORUS
Crossmark - Check for updates button

©2026 American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation