High-fidelity Three-Dimensional Flow Field Reconstruction Driven by Turbulent Fluid Dynamics
Abstract
High-fidelity three-dimensional turbulent flow fields are essential for scientific and engineering analyses, yet obtaining them through direct numerical simulation or high-resolution measurements remains expensive. Super-resolution reconstruction from coarse-grained data therefore offers a promising alternative, but remains fundamentally challenging because fine-scale structures are intrinsically three-dimensional and dynamically coupled across scales. Existing data-driven approaches still struggle to recover multiscale volumetric structures and their coherent temporal evolution. To address these limitations, we propose a turbulence-aware framework for time-resolved three-dimensional flow reconstruction. The framework couples a multiscale neural operator, designed to recover volumetric structures, with a state space model to promote temporal consistency in the reconstructed dynamics. In addition, Navier-Stokes residuals are incorporated as soft constraints to regularize the reconstruction toward physically plausible dynamics. A mixture-of-experts mechanism is also introduced to support unsupervised adaptation to unknown coarse-graining. Across evaluations on canonical turbulent systems and atmospheric reanalysis data, the proposed framework reduces reconstruction errors relative to learning-based and classical baselines, while better preserving multiscale spatial structure and coherent temporal evolution. These results validate the effectiveness of the proposed framework in reconstructing high-fidelity three-dimensional flow fields from coarse-grained data.
Related
- Hurricane Laura-induced loads on wind turbines -- Part I: High-fidelity modeling and sensitivity to turbulent inflow conditions
- Data-driven stochastic reconstruction of dengue epidemic dynamics: state-dependent stochasticity and temporal organization
- Disorder-driven loss of crystalline coherence in two-dimensional electron solids: Quantum Hartree-Fock and classical molecular dynamics
- Electrically driven Rabi dynamics of magnetic-field-induced corner states in a two-dimensional topological insulator
- Ferroelastic domain wall driven glassy dynamics and photoresponse in a SrTiO3 (111) two-dimensional electron gas
BibTeX
@misc{kong2026highfidelity,
title = {{High-fidelity Three-Dimensional Flow Field Reconstruction Driven by Turbulent Fluid Dynamics}},
author = {Hanlin Kong and Jianan Wei and Yuntian Chen and Yujia Wang and Wenguan Wang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.18045},
eprint = {2610.18045},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.18045},
}
RIS
TY - UNPB
TI - High-fidelity Three-Dimensional Flow Field Reconstruction Driven by Turbulent Fluid Dynamics
AU - Hanlin Kong
AU - Jianan Wei
AU - Yuntian Chen
AU - Yujia Wang
AU - Wenguan Wang
PY - 2026
DA - 2026/10/13
PB - arXiv
DO - 10.48550/arxiv.2610.18045
UR - https://arxiv.org/abs/2610.18045
AB - High-fidelity three-dimensional turbulent flow fields are essential for scientific and engineering analyses, yet obtaining them through direct numerical simulation or high-resolution measurements remains expensive. Super-resolution reconstruction from coarse-grained data therefore offers a promising alternative, but remains fundamentally challenging because fine-scale structures are intrinsically three-dimensional and dynamically coupled across scales. Existing data-driven approaches still struggle to recover multiscale volumetric structures and their coherent temporal evolution. To address these limitations, we propose a turbulence-aware framework for time-resolved three-dimensional flow reconstruction. The framework couples a multiscale neural operator, designed to recover volumetric structures, with a state space model to promote temporal consistency in the reconstructed dynamics. In addition, Navier-Stokes residuals are incorporated as soft constraints to regularize the reconstruction toward physically plausible dynamics. A mixture-of-experts mechanism is also introduced to support unsupervised adaptation to unknown coarse-graining. Across evaluations on canonical turbulent systems and atmospheric reanalysis data, the proposed framework reduces reconstruction errors relative to learning-based and classical baselines, while better preserving multiscale spatial structure and coherent temporal evolution. These results validate the effectiveness of the proposed framework in reconstructing high-fidelity three-dimensional flow fields from coarse-grained data.
ER -