Online Learning of a Spectral Eddy Viscosity Closure for Large Eddy Simulation Using Ensemble Kalman Inversion

Katerina Kostova; Yifei Guan

Home/Archive/arXiv

arXiv · 2610.12170

Online Learning of a Spectral Eddy Viscosity Closure for Large Eddy Simulation Using Ensemble Kalman Inversion

Katerina Kostova·Yifei Guan

View original source ↗Read source document ↗Export metadata ↓

Abstract

Large Eddy Simulation (LES) resolves only the large, energy-containing structures of turbulent flows while modeling the effects of unresolved subgrid scales. As a result, the ac- curacy of LES strongly depends on the subgrid-scale closure. This work investigates the online calibration of a spectral viscosity closure for forced two-dimensional turbulence using ensemble Kalman Inversion (EKI). The closure consists of a wavenumber-dependent viscosity function, which is parameterized by a 16-dimensional vector. An ensemble of potential parameter vec- tors is passed through coarse resolution LES simulations, and the resulting kinetic energy spec- tra are compared with a DNS reference spectrum up to the LES-resolved wavenumber range. Then, ensemble statistics iteratively update the spectral viscosity parameters in order to reduce the model-data discrepancy. The convergence of the spectral mismatch is used to assess the accuracy of the calibrated closure by comparing LES with the filtered DNS data in spectral space. The results demonstrate the potential of EKI-based online calibration as a derivative- free framework for constructing data-driven turbulence closures with multiple parameters.

Explore related subjects

physics.flu-dyn

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Katerina Kostova, Yifei Guan. 2026-10-08. Online Learning of a Spectral Eddy Viscosity Closure for Large Eddy Simulation Using Ensemble Kalman Inversion. https://arxiv.org/abs/2610.12170

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Differential geometry of particle motion in Stokesian regime

We present a differential geometric framework for the motion of a non-Brownian particle in the presence of fixed obstacles in a quiescent fluid, in the deterministic Stokesian regime. While the Helmholtz Minimum Dissipation Theorem suggests that the hydrodynamic resistance tensor $R_{ij}$ acts as the natural Riemannian metric of the fluid domain, we demonstrate that particle trajectories driven by constant external forces are \emph{not} geodesics of this pure resistance metric. Instead, they experience a geometric drift perpendicular to the geodesic path due to the manifold's curvature. To reconcile this, we introduce a unified geometric formalism, proving that physical trajectories are geodesics of a conformally scaled metric, $\tilde{g}_{ij} = \mathcal{D}(\mathbf{x})R_{ij}$, where $\mathcal{D}$ is the local power dissipation. This framework establishes that the affine parameter along the trajectory corresponds to the cumulative energy dissipated. We apply this theory to the scattering of a spherical particle by a fixed obstacle, showing that the previously derived trajectory of the particle is recovered as a direct consequence of the curvature of this dissipation-scaled manifold.

physics.flu-dyn↗

arXiv2026

Accurate simulation of pulled and pushed fronts in the nonautonomous Fisher-Kolomogorov-Piskunov-Petrovsky equation

We introduce a novel numerical method for direct simulation of front propagation in the Fisher-KPP equation with a time-dependent parameter on an infinite domain. The method computes a time-dependent boundary condition that accurately captures the leading-edge dynamics by coupling the nonlinear simulation region to a linear approximation region in which the dynamics can be solved exactly via the Green's function of the linearized equation. This approach enables precise front velocity measurements on relatively small computational domains for a variety of nonautonomous regimes and initial conditions for which existing numerical methods break down. We apply the method to pulled and pushed fronts in the Fisher-KPP equation with quadratic and quadratic-cubic nonlinearities, finding that it improves the accuracy of the simulated front velocity even for constant parameters and a fixed domain size. For pulled fronts with a diffusion coefficient that increases algebraically in time, our results reveal a deviation from the natural asymptotic velocity predicted by linear theory, whose explanation requires nonlinear theory. For pushed fronts with constant parameters, the method reproduces the exponential convergence to the theoretical asymptotic front speed and profile with improved precision. For a slowly time-varying linear growth parameter, we find that the pushed front velocity follows the changing parameter adiabatically if the asymptotic pushed velocity remains faster than the natural asymptotic pulled velocity. As the growth parameter moves toward the pushed--pulled transition point, the competition between the pushed and pulled fronts can result in both delayed and even premature onset of the pushed--pulled transition, depending on the form of parameter growth. The numerical method presented here proves to be an effective tool for analyzing front propagation in nonautonomous systems.

physics.flu-dyn↗

arXiv2026

Harmonic Balance Unified Gas-Kinetic Scheme for Multiscale Periodic Flows

A time-domain harmonic balance unified gas-kinetic scheme (HB-UGKS) is developed for simulating periodic non-equilibrium flows across all Knudsen regimes. By applying a time-spectral operator, the unsteady periodic problem is reformulated into a block-coupled, quasi-steady system. To preserve the intrinsic multiscale transportcollision flux coupling of the UGKS, a source-separated formulation incorporates the harmonic balance coupling strictly at the cell-residual level. This allows the periodic limit cycle to be resolved directly via pseudo-time marching with local time-stepping, advancing all temporal collocation points simultaneously and completely bypassing physical startup transients. The framework is validated against three complementary benchmarks: two small-amplitude sheardriven oscillatory flows and a thermally driven cavity under finite-amplitude excitation. The scheme accurately captures intricate non-equilibrium kinetic phenomena, including aspectratio anti-resonance scaling, dynamic shear traction, nonlinear waveform distortions, and acoustic streaming. Across all cases, the HB-UGKS preserves the fidelity of standard timeaccurate simulations, while achieving order-of-magnitude speedups in high-frequency regimes where long transient timescales dominate.

physics.flu-dyn↗