Learning Stiffness Dependent Fluid Structure Dynamics from Coarse Flow Representations

Abstract

This paper develops a data-driven framework for long-term prediction of fluid--structure interaction (FSI) dynamics, focusing on the flow-induced vibration (FIV) of a flexible plate. A stiffness-conditioned neural evolution operator jointly represents the Eulerian flow field and Lagrangian structural state. The plate is represented by 101 ordered structural tokens carrying nodal coordinates and velocities, with nondimensional bending stiffness as a global conditioning variable. Bidirectional cross-attention couples fluid and structural representations within a hybrid CNN-Transformer architecture. Trained with staged multi-step autoregressive rollouts and symmetry-reflected trajectories, a single operator captures three stiffness-dependent response regimes: deflected--flapping, deflected, and flapping. The predicted trajectories preserve the principal flow structures, structural oscillations, and dominant frequencies, while blind 1000-step rollouts remain bounded. The operator also interpolates to stiffness values excluded from training. To reduce sensitivity to under-resolved near-wall gradients in force reconstruction, we develop a differentiable aerodynamic-force module based on the derivative-moment transformation (DMT). Conventional wall-stress surface integrals are replaced by an enclosed 2D curve integral around the core vortex region, enabling accurate reconstruction of lift and drag. A Signed Distance Function (SDF) and smoothed Dirac-delta formulation make the integration fully differentiable while preserving gradient flow. The proposed framework provides an accurate and differentiable surrogate for stiffness-dependent FSI dynamics, enabling efficient parameter studies and future stiffness optimization for flow-energy-harvesting applications.

Quantum field theory with scale-dependent weighted measures: nonlinear momentum representations, vacuum fluctuations, and causal structure
S. González-Salud, R. Cartas-Fuentevilla
arXiv29 Oct 2026PreprintCC BY 4.0Full textHigh-energy & particle physics Quantum physics & quantum informationarXiv
Learning Memory-Dependent Neural Network Correlation Potentials for Accurate Electron Dynamics
Harish S. Bhat, Christine M. Isborn
arXiv3 Sep 2026PreprintarXiv licenceFull textPhysical chemistry Optimisation & controlarXiv
Stiffness-dependent Dielectric Relaxation in Thermoreversible Microgels: Effects of Temperature and Strain
Sayantan Chanda, Sonali Vasant Kawale, Ranjini Bandyopadhyay
arXiv29 Sep 2026PreprintCC BY 4.0Full textPhysical chemistry Condensed matterarXiv
Machine-learning screening and first-principles lattice dynamics resolve the most stable structure of NaCa(BH₄)₃ from a data-driven candidate search
Sanaa Ismail, Ricardo Amaral, Attia A. Gadallah +2
arXiv25 Sep 2026PreprintarXiv licenceFull textCondensed matterarXiv
From Analytic Structure to Quantum Complexity: Walsh-Pauli Representations of Continuum Operators
Raphael De Sousa, Partha Nandi, Francesco Petruccione
arXiv10 Sep 2026PreprintarXiv licenceFull textMathematical physics High-energy & particle physicsarXiv

Cite

BibTeX RIS

BibTeX

@misc{pu2026learning,
  title = {{Learning Stiffness Dependent Fluid Structure Dynamics from Coarse Flow Representations}},
  author = {Chun-Jun Pu and Li-Wei Chen and Hai-Bo Huang},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.26816},
  eprint = {2609.26816},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.26816},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Learning Stiffness Dependent Fluid Structure Dynamics from Coarse Flow Representations
AU  - Chun-Jun Pu
AU  - Li-Wei Chen
AU  - Hai-Bo Huang
PY  - 2026
DA  - 2026/09/19
PB  - arXiv
DO  - 10.48550/arxiv.2609.26816
UR  - https://arxiv.org/abs/2609.26816
AB  - This paper develops a data-driven framework for long-term prediction of fluid--structure interaction (FSI) dynamics, focusing on the flow-induced vibration (FIV) of a flexible plate. A stiffness-conditioned neural evolution operator jointly represents the Eulerian flow field and Lagrangian structural state. The plate is represented by 101 ordered structural tokens carrying nodal coordinates and velocities, with nondimensional bending stiffness as a global conditioning variable. Bidirectional cross-attention couples fluid and structural representations within a hybrid CNN-Transformer architecture. Trained with staged multi-step autoregressive rollouts and symmetry-reflected trajectories, a single operator captures three stiffness-dependent response regimes: deflected--flapping, deflected, and flapping. The predicted trajectories preserve the principal flow structures, structural oscillations, and dominant frequencies, while blind 1000-step rollouts remain bounded. The operator also interpolates to stiffness values excluded from training. To reduce sensitivity to under-resolved near-wall gradients in force reconstruction, we develop a differentiable aerodynamic-force module based on the derivative-moment transformation (DMT). Conventional wall-stress surface integrals are replaced by an enclosed 2D curve integral around the core vortex region, enabling accurate reconstruction of lift and drag. A Signed Distance Function (SDF) and smoothed Dirac-delta formulation make the integration fully differentiable while preserving gradient flow. The proposed framework provides an accurate and differentiable surrogate for stiffness-dependent FSI dynamics, enabling efficient parameter studies and future stiffness optimization for flow-energy-harvesting applications.
ER  -

Abstract

Related

Cite