arXiv subjects

Zelong Yuan

Publications and source records attributed to Zelong Yuan.

16 recordsLinked to original sources

arXiv2026

Physics informed wavelet Fourier representation for multiscale fluid dynamics

Multiscale fluid flows often contain localized flow structures, such as viscous shock layers, wet-dry fronts, steady viscous wakes, decaying vortical structures, and vortex-shedding patterns, whose accurate prediction requires the simultaneous preservation of global conservation trends and small-scale gradients. This study examines these flow-physics requirements through a physics-informed wavelet-Fourier (PIWF) representation for multiscale fluid dynamics. Instead of relying on a single monolithic neural approximator, the formulation separates two complementary components of the flow field within a physics-informed neural representation: long-range coherent modes through a Fourier-basis branch and localized steep-gradient or vortical features through a compactly supported wavelet branch. The outputs are fused with a residual multilayer perceptron using channel attention, and the governing equations, initial conditions, and boundary conditions are imposed directly through the physics-informed loss. The model is assessed on five canonical fluid-dynamics problems: Burgers' equation, the shallow water equations, Kovasznay flow, Taylor--Green vortex flow, and two-dimensional cylinder wake flow. The results show that PIWF improves the resolution of shock-like gradients, wet--dry interfaces, steady wake fields, decaying vortical structures, vorticity extrema, and broadband wake spectra relative to standard physics-informed neural networks and physics-informed Kolmogorov--Arnold networks. These findings indicate that a wavelet-Fourier physics-informed representation can provide a useful route for analyzing multiscale flow phenomena when high-fidelity interior reference data are limited or unavailable.

physics.flu-dyn↗

arXiv2026

Self-scaling tensor basis neural network for Reynolds stress modeling of wall-bounded turbulence

Recent advances in data-driven turbulence modeling have established tensor basis neural networks (TBNN) as a physically grounded framework for Reynolds-stress closure in Reynolds-averaged Navier-Stokes (RANS) simulations. However, their robustness in wall-bounded turbulent flows remains limited across Reynolds numbers and geometries due to the lack of an intrinsic scaling mechanism. In this work, we propose a self-scaling tensor basis neural network (STBNN) for Reynolds-stress modeling of wall-bounded turbulence. The model incorporates an invariant velocity-gradient normalization derived from the first two invariants of the velocity-gradient tensor, providing an intrinsic and geometry-independent scale that balances strain and rotation effects without relying on empirical coefficients or wall-distance inputs. Owing to its frame-indifferent formulation, the approach preserves Galilean and rotational invariance while maintaining a physically interpretable representation of Reynolds-stress anisotropy. STBNN is evaluated through a priori and a posteriori studies using direct numerical simulation (DNS) data of canonical wall-bounded flows, including plane channel and periodic hill flows. In a priori tests, the model accurately reproduces Reynolds-stress anisotropy, with correlation coefficients exceeding 99% and relative errors below 10%, while capturing near-wall scaling and logarithmic-layer behavior. In a posteriori RANS simulations, STBNN predicts mean velocity profiles in close agreement with DNS and improves prediction of separation and reattachment compared with linear and quadratic eddy-viscosity models and the baseline TBNN. Notably, a model trained at low Reynolds numbers generalizes to higher Reynolds numbers and unseen geometries. These results demonstrate the effectiveness of the proposed framework for data-driven Reynolds-stress modeling in wall-bounded turbulent flows.

physics.flu-dyn↗

arXiv2026

Machine-learning-based simulation of turbulent flows over periodic hills using a hybrid U-Net and Fourier neural operator framework

Simulating massively separated turbulent flows over bodies is one of the major applications for large-eddy simulation (LES). In the current work, we propose a machine-learning-based LES framework for the rapid simulation of turbulent flows over periodic hills using a hybrid U-Net and Fourier neural operator (HUFNO) framework. The newly proposed HUFNO model integrates the strengths of both the convolutional neural network (CNN) and Fourier neural operator (FNO) in a novel way that the FNO is applied in the periodic directions of the flow field while the non-periodicity is handled by the CNN-based U-Net framework. In the numerical tests, compared to the original FNO and the U-Net framework, the HUFNO model shows a higher accuracy in the predictions of the velocity field and Reynolds stresses. Further numerical experiments in the LES show that the HUFNO framework outperforms the traditional Smagorinsky (SMAG) model and the wall-adapted local eddy-viscosity (WALE) model in the predictions of the turbulence statistics, the energy spectrum, the invariant characteristics of velocity gradients, the wall stresses and the flow separation structures, with much lower computational cost. Importantly, the accuracy and efficiency are transferable to unseen initial conditions, Reynolds number and hill shapes, underscoring its great potentials for the fast prediction of strongly separated turbulent flows over curved boundaries.

physics.flu-dyn↗

arXiv2025

Physics-informed Kolmogorov-Arnold Network with Chebyshev Polynomials for Fluid Mechanics

Solving partial differential equations (PDEs) is essential in scientific forecasting and fluid dynamics. Traditional approaches often incur expensive computational costs and trade-offs in efficiency and accuracy. Recent deep neural networks have improved the accuracy but require high-quality training data. Physics-informed neural networks (PINNs) effectively integrate physical laws to reduce the data reliance in limited sample scenarios. A novel machine-learning framework, Chebyshev physics-informed Kolmogorov--Arnold network (ChebPIKAN), is proposed to integrate the robust architectures of Kolmogorov--Arnold networks (KAN) with physical constraints to enhance the calculation accuracy of PDEs for fluid mechanics. We study the fundamentals of KAN, take advantage of the orthogonality of Chebyshev polynomial basis functions in spline fitting, and integrate physics-informed loss functions that are tailored to specific PDEs in fluid dynamics, including Allen--Cahn equation, nonlinear Burgers equation, Helmholtz equations, Kovasznay flow, cylinder wake flow, and cavity flow. Extensive experiments demonstrate that the proposed ChebPIKAN model significantly outperforms the standard KAN architecture in solving various PDEs by effectively embedding essential physical information. These results indicate that augmenting KAN with physical constraints can alleviate the overfitting issues of KAN and improve the extrapolation performance. Consequently, this study highlights the potential of ChebPIKAN as a powerful tool in computational fluid dynamics and propose a path toward fast and reliable predictions in fluid mechanics and beyond.

physics.flu-dyn↗

arXiv2025

Physics-informed Fourier Basis Neural Network for Fluid Mechanics

Solving partial differential equations (PDEs) is an important yet challenging task in fluid mechanics. In this study, we embed an improved Fourier series into neural networks and propose a physics-informed Fourier basis neural network (FBNN) by incorporating physical information to solve canonical PDEs in fluid mechanics. The results demonstrated that the proposed framework exhibits a strong nonlinear fitting capability and exceptional periodic modeling performance. In particular, our model shows significant advantages for the Burgers equation with discontinuous solutions and Helmholtz equation with strong periodicity. By directly introducing sparse distributed data to reconstruct the entire flow field, we further intuitively validated the direct superiority of FBNN over conventional artificial neural networks (ANN) as well as the benefits of incorporating physical information into the network. By adjusting the activation functions of networks and comparing with an ANN and conventional physics-informed neural network, we proved that performance of the proposed FBNN architecture is not highly sensitive to the choice of activation functions. The nonlinear fitting capability of FBNN avoids excessive reliance on activation functions, thereby mitigating the risk of suboptimal outcomes or training failures stemming from unsuitable activation function choices.hese results highlightthe potential of PIFBNN as a powerful tool in computational fluid dynamics.

physics.flu-dyn↗

arXiv2024

Prediction of turbulent channel flow using Fourier neural operator-based machine-learning strategy

Fast and accurate predictions of turbulent flows are of great importance in the science and engineering field. In this paper, we investigate the implicit U-Net enhanced Fourier neural operator (IUFNO) in the stable prediction of long-time dynamics of three-dimensional (3D) turbulent channel flows. The trained IUFNO models are tested in the large-eddy simulations (LES) at coarse grids for three friction Reynolds numbers: $Re_τ\approx180$, $395$ and $590$. The adopted near-wall mesh grids are tangibly coarser than the general requirements for wall-resolved LES. Compared to the original Fourier neural operator (FNO), the implicit FNO (IFNO) and U-Net enhanced FNO (UFNO), the IUFNO model has a much better long-term predictive ability. The numerical experiments show that the IUFNO framework outperforms the traditional dynamic Smagorinsky model (DSM) and the wall-adapted local eddy-viscosity (WALE) model in the predictions of a variety of flow statistics and structures, including the mean and fluctuating velocities, the probability density functions (PDFs) and joint PDF of velocity fluctuations, the Reynolds stress profile, the kinetic energy spectrum, and the Q-criterion (vortex structures). Meanwhile, the trained IUFNO models are computationally much faster than the traditional LES models. Thus, the IUFNO model is a promising approach for the fast prediction of wall-bounded turbulent flow.

physics.flu-dyn↗

arXiv2024

Fourier neural operator for large eddy simulation of compressible Rayleigh-Taylor turbulence

The Fourier neural operator (FNO) framework is applied to the large eddy simulation (LES) of three-dimensional compressible Rayleigh-Taylor (RT) turbulence with miscible fluids at Atwood number $A_t=0.5$, stratification parameter $Sr=1.0$, and Reynolds numbers $Re=10000$ and 30000. The FNO model is first used for predicting three-dimensional compressible turbulence. The different magnitudes of physical fields are normalized using root mean square values for an easier training of FNO models. In the \emph{a posteriori} tests, the FNO model outperforms the velocity gradient model (VGM), the dynamic Smagorinsky model (DSM), and implicit large eddy simulation (ILES) in predicting various statistical quantities and instantaneous structures, and is particularly superior to traditional LES methods in predicting temperature fields and velocity divergence. Moreover, the computational efficiency of the FNO model is much higher than that of traditional LES methods. FNO models trained with short-time, low Reynolds number data exhibit a good generalization performance on longer-time predictions and higher Reynolds numbers in the \emph{a posteriori} tests.

physics.flu-dyn↗

arXiv2024

A transformer-based neural operator for large-eddy simulation of turbulence

Predicting the large-scale dynamics of three-dimensional (3D) turbulence is challenging for machine learning approaches. This paper introduces a transformer-based neural operator (TNO) to achieve precise and efficient predictions in the large-eddy simulation (LES) of 3D turbulence. The performance of the proposed TNO model is systematically tested and compared with LES using classical sub-grid scale (SGS) models, including the dynamic Smagorinsky model (DSM) and the dynamic mixed model (DMM), as well as the original Fourier neural operator (FNO) model, in homogeneous isotropic turbulence (HIT) and free-shear turbulent mixing layer. The numerical simulations comprehensively evaluate the performance of these models on a variety of flow statistics, including the velocity spectrum, the probability density functions (PDFs) of vorticity, the PDFs of velocity increments, the evolution of turbulent kinetic energy, and the iso-surface of the Q-criterion. The results indicate that the accuracy of the TNO model is comparable to the LES with DSM model, and outperforms the FNO model and LES using DMM in HIT. In the free-shear turbulence, the TNO model exhibits superior accuracy compared to other models. Moreover, the TNO model has fewer parameters than the FNO model and enables long-term stable predictions, which the FNO model cannot achieve. The well-trained TNO model is significantly faster than traditional LES with DSM and DMM models, and can be generalized to higher Taylor-Reynolds number cases, indicating its strong potential for 3D nonlinear engineering applications.

physics.flu-dyn↗

arXiv2024

The discrete direct deconvolution model in the large eddy simulation of turbulence

The discrete direct deconvolution model (D3M) is developed for the large-eddy simulation (LES) of turbulence. The D3M is a discrete approximation of previous direct deconvolution model studied by Chang et al. ["The effect of sub-filter scale dynamics in large eddy simulation of turbulence," Phys. Fluids 34, 095104 (2022)]. For the first type model D3M-1, the original Gaussian filter is approximated by local discrete formulation of different orders, and direct inverse of the discrete filter is applied to reconstruct the unfiltered flow field. The inverse of original Gaussian filter can be also approximated by local discrete formulation, leading to a fully local model D3M-2. Compared to traditional models including the dynamic Smagorinsky model (DSM) and the dynamic mixed model (DMM), the D3M-1 and D3M-2 exhibit much larger correlation coefficients and smaller relative errors in the a priori studies. In the a posteriori validations, both D3M-1 and D3M-2 can accurately predict turbulence statistics, including velocity spectra, probability density functions (PDFs) of sub-filter scale (SFS) stresses and SFS energy flux, as well as time-evolving kinetic energy spectra, momentum thickness, and Reynolds stresses in turbulent mixing layer. Moreover, the proposed model can also well capture spatial structures of the Q-criterion iso surfaces. Thus, the D3M holds potential as an effective SFS modeling approach in turbulence simulations.

physics.flu-dyn↗

arXiv2023

Ensemble data assimilation-based mixed subgrid-scale model for large-eddy simulations

An ensemble Kalman filter (EnKF)-based mixed model (EnKF-MM) is proposed for the subgrid-scale (SGS) closure in the large-eddy simulation (LES) of turbulence. The model coefficients are determined through the EnKF-based data assimilation technique. The direct numerical simulation (DNS) results are filtered to obtain the benchmark data for LES. Reconstructing the correct kinetic energy spectrum of the filtered DNS (fDNS) data has been adopted as the target for the EnKF to optimize the coefficient of the functional part in the mixed model. The proposed EnKF-MM framework is subsequently tested in the LES of both the incompressible homogeneous isotropic turbulence (HIT) and turbulent mixing layer (TML). The performance of LES is comprehensively examined through the predictions of the flow statistics including the velocity spectrum, the probability density functions (PDFs) of the SGS stress, the PDF of the strain rate and the PDF of the SGS energy flux. The structure functions, the evolution of turbulent kinetic energy, the mean flow and the Reynolds stress profile, and the iso-surface of the Q-criterion are also examined to evaluate the spatial-temporal predictions by different SGS models. The results of the EnKF-MM framework are consistently more satisfying compared to the traditional SGS models, including the dynamic Smagorinsky model (DSM), the dynamic mixed model (DMM) and the velocity gradient model (VGM), demonstrating its great potential in the optimization of SGS models for LES of turbulence.

physics.flu-dyn↗

arXiv2023

Long-term predictions of turbulence by implicit U-Net enhanced Fourier neural operator

Long-term predictions of nonlinear dynamics of three-dimensional (3D) turbulence are very challenging for machine learning approaches. In this paper, we propose an implicit U-Net enhanced Fourier neural operator (IU-FNO) for stable and efficient predictions on the long-term large-scale dynamics of turbulence. The IU-FNO model employs implicit recurrent Fourier layers for deeper network extension and incorporates the U-net network for the accurate prediction on small-scale flow structures. The model is systematically tested in large-eddy simulations of three types of 3D turbulence, including forced homogeneous isotropic turbulence (HIT), temporally evolving turbulent mixing layer, and decaying homogeneous isotropic turbulence. The numerical simulations demonstrate that the IU-FNO model is more accurate than other FNO-based models including vanilla FNO, implicit FNO (IFNO) and U-Net enhanced FNO (U-FNO), and dynamic Smagorinsky model (DSM) in predicting a variety of statistics including the velocity spectrum, probability density functions (PDFs) of vorticity and velocity increments, and instantaneous spatial structures of flow field. Moreover, IU-FNO improves long-term stable predictions, which has not been achieved by the previous versions of FNO. Besides, the proposed model is much faster than traditional LES with DSM model, and can be well generalized to the situations of higher Taylor-Reynolds numbers and unseen flow regime of decaying turbulence.

physics.flu-dyn↗

arXiv2023

Adjoint-based variational optimal mixed models for large-eddy simulation of turbulence

An adjoint-based variational optimal mixed model (VOMM) is proposed for subgrid-scale (SGS) closure in large-eddy simulation (LES) of turbulence. The stabilized adjoint LES equations are formulated by introducing a minimal regularization to address the numerical instabilities of the long-term gradient evaluations in chaotic turbulent flows. The VOMM model parameters are optimized by minimizing the discrepancy of energy dissipation spectra between LES calculations and a priori knowledge of direct numerical simulation (DNS) using the gradient-based optimization. The a posteriori performance of the VOMM model is comprehensively examined in LES of three turbulent flows, including the forced homogeneous isotropic turbulence, decaying homogenous isotropic turbulence, and temporally evolving turbulent mixing layer. The VOMM model outperforms the dynamic Smagorinsky model (DSM), dynamic mixed model (DMM) and approximate deconvolution model (ADM) in predictions of various turbulence statistics, including the velocity spectrum, structure functions, statistics of velocity increments and vorticity, temporal evolutions of the turbulent kinetic energy, dissipation rate, momentum thickness and Reynolds stress, as well as the instantaneous vortex structures at different grid resolutions and times. In addition, the VOMM model only takes up 30% time of the DMM model for all flow scenarios. These results demonstrate that the proposed VOMM model improves the numerical stability of LES and has high a posteriori accuracy and computational efficiency by incorporating the a priori information of turbulence statistics, highlighting that the VOMM model has a great potential to develop advanced SGS models in the LES of turbulence.

physics.flu-dyn↗

arXiv2022

Linear attention coupled Fourier neural operator for simulation of three-dimensional turbulence

Modeling three-dimensional (3D) turbulence by neural networks is difficult because 3D turbulence is highly-nonlinear with high degrees of freedom and the corresponding simulation is memory-intensive. Recently, the attention mechanism has been shown as a promising approach to boost the performance of neural networks on turbulence simulation. However, the standard self-attention mechanism uses $O(n^2)$ time and space with respect to input dimension $n$, and such quadratic complexity has become the main bottleneck for attention to be applied on 3D turbulence simulation. In this work, we resolve this issue with the concept of linear attention network. The linear attention approximates the standard attention by adding two linear projections, reducing the overall self-attention complexity from $O(n^2)$ to $O(n)$ in both time and space. The linear attention coupled Fourier neural operator (LAFNO) is developed for the simulation of 3D turbulence. Numerical simulations show that the linear attention mechanism provides 40\% error reduction at the same level of computational cost, and LAFNO can accurately reconstruct a variety of statistics and instantaneous spatial structures of 3D turbulence. The linear attention method would be helpful for the improvement of neural network models of 3D nonlinear problems involving high-dimensional data in other scientific domains.

physics.flu-dyn↗

arXiv2022

Temporally sparse data assimilation for the small-scale reconstruction of turbulence

Previous works have shown that the small-scale information of incompressible homogeneous isotropic turbulence (HIT) is fully recoverable as long as sufficient large-scale structures are continuously enforced through temporally continuous data assimilation (TCDA). In the current work, we show that the assimilation time step can be relaxed to values about 1 $\sim$ 2 orders larger than that for TCDA, using a temporally sparse data assimilation (TSDA) strategy, while the accuracy is still maintained or even slightly better in the presence of non-negligible large-scale errors. The one-step data assimilation (ODA) is examined to unravel the mechanism of TSDA. It is shown that the relaxation effect for errors above the assimilation wavenumber $k_a$ is responsible for the error decay in ODA. Meanwhile, The errors contained in the large scales can propagate into small scales and make the high-wavenumber ($k>k_a$) error noise decay slower with TCDA than TSDA. This mechanism is further confirmed by incorporating different levels of errors in the large scales of the reference flow field. The advantage of TSDA is found to grow with the magnitude of the incorporated errors. Thus, it is potentially more beneficial to adopt TSDA if the reference data contains non-negligible errors. Finally, an outstanding issue raised in previous works regarding the possibility of recovering the dynamics of sub-Kolmogorov scales using direct numerical simulation (DNS) data at Kolmogorov scale resolution is also discussed.

physics.flu-dyn↗

arXiv2022

Attention-Enhanced Neural Network Models for Turbulence Simulation

Deep neural network models have shown a great potential in accelerating the simulation of fluid dynamic systems. Once trained, these models can make inference within seconds, thus can be extremely efficient. However, they suffer from a generalization problem when the flow becomes chaotic and turbulent. One of the most important reasons is that, existing models lack the mechanism to handle the unique characteristic of turbulent flow: multi-scale flow structures are non-uniformly distributed and strongly nonequilibrium. In this work, we address this issue with the concept of visual attention: intuitively, we expect the attention module to capture the nonequilibrium of turbulence by automatically adjusting weights on different regions. We benchmark the performance improvement with a state of the art neural network model, the Fourier Neural Operator (FNO), on two-dimensional (2D) turbulence prediction task. Numerical experiments show that the attention-enhanced neural network model can generalize well on higher Reynolds numbers flow, and can accurately reconstruct a variety of statistics and instantaneous spatial structures of turbulence. The attention mechanism provides 40% error reduction with 1% increase of parameters, at the same level of computational cost.

physics.flu-dyn↗

arXiv2020

Deconvolutional artificial neural network models for large eddy simulation of turbulence

Deconvolutional artificial neural network (DANN) models are developed for subgrid-scale (SGS) stress in large eddy simulation (LES) of turbulence. The filtered velocities at different spatial points are used as input features of the DANN models to reconstruct the unfiltered velocity. The grid width of the DANN models is chosen to be smaller than the filter width, in order to accurately model the effects of SGS dynamics. The DANN models can predict the SGS stress more accurately than the conventional approximate deconvolution method (ADM) and velocity gradient model (VGM) in a prior study: the correlation coefficients can be made larger than 99\% and the relative errors can be made less than 15\% for the DANN model. In an a posteriori study, a comprehensive comparison of the DANN model, the implicit large eddy simulation (ILES), the dynamic Smagorinsky model (DSM), and the dynamic mixed model (DMM) shows that: the DANN model is superior to the ILES, DSM, and DMM models in the prediction of the velocity spectrum, various statistics of velocity and the instantaneous coherent structures without increasing the considerable computational cost. Besides, the trained DANN models without any fine-tuning can predict the velocity statistics well for different filter widths. These results indicate that the DANN framework with consideration of SGS spatial features is a promising approach to develop advanced SGS models in the LES of turbulence.

physics.flu-dyn↗