Papers
Machine learning for physics, as it is published.
New work from arXiv on CFD surrogates, neural operators and physics-informed learning, refreshed every hour. Search it, filter by topic, or start with the papers we keep coming back to.
Papers
New work from arXiv on CFD surrogates, neural operators and physics-informed learning, refreshed every hour. Search it, filter by topic, or start with the papers we keep coming back to.
564 papers in Turbulence · newest first
physics.flu-dyn
We propose the Artificial Intelligence Velocimetry-Thermometry (AIVT) method to infer hidden temperature fields from experimental turbulent velocity data. This physics-informed machine learning method enables us to infer continuous temperature fields using only sparse velocity data, hence eliminating the need for direct temperature measurements. Specifically, AIVT is based on physics-informed Kolmogorov-Arnold Networks (not neural networks) and is trained by optimizing a combined loss function that minimizes the residuals of the velocity data, boundary conditions, and the governing equations. We apply AIVT to a unique set of experimental volumetric and simultaneous temperature and velocity data of Rayleigh-Bénard convection (RBC) that we acquired by combining Particle Image Thermometry and Lagrangian Particle Tracking. This allows us to compare AIVT predictions and measurements directly. We demonstrate that we can reconstruct and infer continuous and instantaneous velocity and temperature fields from sparse experimental data at a fidelity comparable to direct numerical simulations (DNS) of turbulence. This, in turn, enables us to compute important quantities for quantifying turbulence, such as fluctuations, viscous and thermal dissipation, and QR distribution. This paradigm shift in processing experimental data using AIVT to infer turbulent fields at DNS-level fidelity is a promising avenue in breaking the current deadlock of quantitative understanding of turbulence at high Reynolds numbers, where DNS is computationally infeasible.
physics.flu-dyn
This study proposes a novel super-resolution (or SR) framework for generating high-resolution turbulent boundary layer (TBL) flow from low-resolution inputs. The framework combines a super-resolution generative adversarial neural network (SRGAN) with down-sampling modules (DMs), integrating the residual of the continuity equation into the loss function. DMs selectively filter out components with excessive energy dissipation in low-resolution fields prior to the super-resolution process. The framework iteratively applies the SRGAN and DM procedure to fully capture the energy cascade of multi-scale flow structures, collectively termed the SRGAN-based energy cascade framework (EC-SRGAN). Despite being trained solely on turbulent channel flow data (via "zero-shot transfer"), EC-SRGAN exhibits remarkable generalization in predicting TBL small-scale velocity fields, accurately reproducing wavenumber spectra compared to DNS results. Furthermore, a super-resolution core is trained at a specific super-resolution ratio. By leveraging this pre-trained super-resolution core, EC-SRGAN efficiently reconstructs TBL fields at multiple super-resolution ratios from various levels of low-resolution inputs, showcasing strong flexibility. By learning turbulent scale invariance, EC-SRGAN demonstrates robustness across different TBL datasets. These results underscore EC-SRGAN potential for generating and predicting wall turbulence with high flexibility, offering promising applications in addressing diverse TBL-related challenges.
math.DS
We introduce a simple, stochastic, a-posteriori, turbulence closure model based on a reduced subgrid scale term. This subgrid scale term is tailor-made to capture the statistics of a small set of spatially-integrate quantities of interest (QoIs), with only one unresolved scalar time series per QoI. In contrast to other data-driven surrogates the dimension of the "learning problem" is reduced from an evolving field to one scalar time series per QoI. We use an a-posteriori, nudging approach to find the distribution of the scalar series over time. This approach has the advantage of taking the interaction between the solver and the surrogate into account. A stochastic surrogate parametrization is obtained by random sampling from the found distribution for the scalar time series. Compared to an a-priori trained convolutional neural network, evaluating the new method is computationally much cheaper and gives similar long-term statistics.
physics.flu-dyn
Turbulent flows play an important role in many scientific and technological design problems. Both Sub-Grid Scale (SGS) models in Large Eddy Simulations (LES) and Reynolds Averaged Navier Stokes (RANS) based modeling will require turbulence models for computational research of turbulent flows in the future. Turbulence model-based simulations suffer from a multitude of causes of forecast uncertainty. For example, the simplifications and assumptions employed to make these turbulence models computationally tractable and economical lead to predictive uncertainty. For safety-critical engineering design applications, we need reliable estimates of this uncertainty. This article focuses on Uncertainty Quantification (UQ) for Computational Fluid Dynamics (CFD) simulations. We review recent advances in the estimate of many types of uncertainty components, including numerical, aleatoric, and epistemic. We go into further depth on the possible use of Machine Learning (ML) methods to quantify these uncertainties. Above all, we elaborate further on significant limitations in these techniques. These range from realizability constraints on the Eigenspace Perturbation Method (EPM) to the requirement for Monte Carlo (MC) approaches for mixed uncertainty. Based on this study, we pinpoint important problems that need to be addressed and offer focused solutions to move beyond these obstacles.
physics.flu-dyn
We apply two data assimilation (DA) methods, a smoother and a filter, and a model-free machine learning (ML) shallow network to forecast two weakly turbulent systems. We analyse the effect of the spatial sparsity of observations on accuracy of the predictions obtained from these data-driven methods. Based on the results, we divide the spatial sparsity levels in three zones. First is the good-predictions zone in which both DA and ML methods work. We find that in the good-predictions zone the observations remain dense enough to accurately capture the fractal manifold of the system's dynamics, which is measured using the correlation dimension. The accuracy of the DA methods in this zone remains almost as good as for full-resolution observations. Second is the reasonable-predictions zone in which the DA methods still work but at reduced prediction accuracy. Third is the bad-predictions zone in which even the DA methods fail. We find that the sparsity level up to which the DA methods work is almost the same up to which chaos synchronisation of these systems can be achieved. The main implications of these results are that they (i) firmly establish the spatial resolution up to which the data-driven methods can be utilised, (ii) provide measures to determine if adding more sensors will improve the predictions, and (iii) quantify the advantage (in terms of the required measurement resolution) of using the governing equations within data-driven methods. We also discuss the applicability of these results to fully developed turbulence.
physics.flu-dyn
Simulating spatiotemporal turbulence with high fidelity remains a cornerstone challenge in computational fluid dynamics (CFD) due to its intricate multiscale nature and prohibitive computational demands. Traditional approaches typically employ closure models, which attempt to represent small-scale features in an unresolved manner. However, these methods often sacrifice accuracy and lose high-frequency/wavenumber information, especially in scenarios involving complex flow physics. In this paper, we introduce an innovative neural differentiable modeling framework designed to enhance the predictability and efficiency of spatiotemporal turbulence simulations. Our approach features differentiable hybrid modeling techniques that seamlessly integrate deep neural networks with numerical PDE solvers within a differentiable programming framework, synergizing deep learning with physics-based CFD modeling. Specifically, a hybrid differentiable neural solver is constructed on a coarser grid to capture large-scale turbulent phenomena, followed by the application of a Bayesian conditional diffusion model that generates small-scale turbulence conditioned on large-scale flow predictions. Two innovative hybrid architecture designs are studied, and their performance is evaluated through comparative analysis against conventional large eddy simulation techniques with physics-based subgrid-scale closures and purely data-driven neural solvers. The findings underscore the potential of the neural differentiable modeling framework to significantly enhance the accuracy and computational efficiency of turbulence simulations. This study not only demonstrates the efficacy of merging deep learning with physics-based numerical solvers but also sets a new precedent for advanced CFD modeling techniques, highlighting the transformative impact of differentiable programming in scientific computing.
physics.flu-dyn
Wall-cooling effect in hypersonic boundary layers can significantly alter the near-wall turbulence behavior, which is not accurately modeled by traditional RANS turbulence models. To address this shortcoming, this paper presents a turbulence modeling approach for hypersonic flows with cold-wall conditions using an iterative ensemble Kalman method. Specifically, a neural-network-based turbulence model is used to provide closure mapping from mean flow quantities to Reynolds stress as well as a variable turbulent Prandtl number. Sparse observation data of velocity and temperature are used to train the turbulence model. This approach is analyzed using direct numerical simulation database for zero-pressure gradient (ZPG) boundary layer flows over a flat plate with a Mach number between 6 and 14 and wall-to-recovery temperature ratios ranging from 0.18 to 0.76. Two training cases are conducted: 1) a single training case with observation data from one flow case, 2) a joint training case where data from two flow cases are simultaneously used for training. Trained models are also tested for generalizability on the remaining flow cases in each of the training cases. The results are also analyzed for insights to inform the future work towards enhancing the generalizability of the learned turbulence model.
cs.CE
This work presents a methodology to predict a near-optimal spacing function, which defines the element sizes, suitable to perform steady RANS turbulent viscous flow simulations. The strategy aims at utilising existing high fidelity simulations to compute a target spacing function and train an artificial neural network (ANN) to predict the spacing function for new simulations, either unseen operating conditions or unseen geometric configurations. Several challenges induced by the use of highly stretched elements are addressed. The final goal is to substantially reduce the time and human expertise that is nowadays required to produce suitable meshes for simulations. Numerical examples involving turbulent compressible flows in two dimensions are used to demonstrate the ability of the trained ANN to predict a suitable spacing function. The influence of the NN architecture and the size of the training dataset are discussed. Finally, the suitability of the predicted meshes to perform simulations is investigated.
quant-ph
Atmospheric turbulence is the main barrier to large-scale free-space quantum communication networks. Aberrations distort optical information carriers, thus limiting or preventing the possibility of establishing a secure link between two parties. For this reason, forecasting the turbulence strength within an optical channel is highly desirable, as it allows for knowing the optimal timing to establish a secure link in advance. Here, we train a Recurrent Neural Network, TAROCCO, to predict the turbulence strength within a free-space channel. The training is based on weather and turbulence data collected over 9 months for a 5.4 km intra-city free-space link across the City of Ottawa. The implications of accurate predictions from our network are demonstrated in a simulated high-dimensional Quantum Key Distribution protocol based on orbital angular momentum states of light across different turbulence regimes. TAROCCO will be crucial in validating a free-space channel to optimally route the key exchange for secure communications in real experimental scenarios.
physics.flu-dyn
Within the context of machine learning-based closure mappings for RANS turbulence modelling, physical realizability is often enforced using ad-hoc postprocessing of the predicted anisotropy tensor. In this study, we address the realizability issue via a new physics-based loss function that penalizes non-realizable results during training, thereby embedding a preference for realizable predictions into the model. Additionally, we propose a new framework for data-driven turbulence modelling which retains the stability and conditioning of optimal eddy viscosity-based approaches while embedding equivariance. Several modifications to the tensor basis neural network to enhance training and testing stability are proposed. We demonstrate the conditioning, stability, and generalization of the new framework and model architecture on three flows: flow over a flat plate, flow over periodic hills, and flow through a square duct. The realizability-informed loss function is demonstrated to significantly increase the number of realizable predictions made by the model when generalizing to a new flow configuration. Altogether, the proposed framework enables the training of stable and equivariant anisotropy mappings, with more physically realizable predictions on new data. We make our code available for use and modification by others. Moreover, as part of this study, we explore the applicability of Kolmogorov-Arnold Networks (KAN) to turbulence modeling, assessing its potential to address non-linear mappings in the anisotropy tensor predictions and demonstrating promising results for the flat plate case.
physics.flu-dyn
With the development of high performance computer and experimental technology, the study of turbulence has accumulated a large number of high fidelity data. However, few general turbulence knowledge has been found from the data. So we use the symbolic regression (SR) method to find a new mixing length formula which is generally valid in wall-bounded turbulence, and this formula has physical interpretation that it has correct asymptotic relationships in viscous sublayer,buffer layer, log-law region and outer region. Coupled with Reynolds averaged Navier-Stokes (RANS) solver, we test several classic cases. The prediction results fully demonstrate the accuracy and generalization of the formula. So far, we have found that SR method can help us find general laws from complex turbulent systems, and it is expected that through this 'white box' machine learning method, more turbulence knowledge with physical interpretation can be found in the future.
physics.flu-dyn
Heavy and light particles are commonly found in many natural phenomena and industrial processes, such as suspensions of bubbles, dust, and droplets in incompressible turbulent flows. Based on a recent machine learning approach using a diffusion model that successfully generated single tracer trajectories in three-dimensional turbulence and passed most statistical benchmarks across time scales, we extend this model to include heavy and light particles. Given the particle type - tracer, light, or heavy - the model can generate synthetic, realistic trajectories with correct fat-tail distributions for acceleration, anomalous power laws, and scale dependent local slope properties. This work paves the way for future exploration of the use of diffusion models to produce high-quality synthetic datasets for different flow configurations, potentially allowing interpolation between different setups and adaptation to new conditions.
math.NA
In this paper, we propose an equation-based parametric Reduced Order Model (ROM), whose accuracy is improved with data-driven terms added into the reduced equations. These additions have the aim of reintroducing contributions that in standard ROMs are not taken into account. In particular, in this work we consider two types of contributions: the turbulence modeling, added through a reduced-order approximation of the eddy viscosity field, and the correction model, aimed to re-introduce the contribution of the discarded modes. Both approaches have been investigated in previous works and the goal of this paper is to extend the model to a parametric setting making use of ad-hoc machine learning procedures. More in detail, we investigate different neural networks' architectures, from simple dense feed-forward to Long-Short Term Memory neural networks, in order to find the most suitable model for the re-introduced contributions. We tested the methods on two test cases with different behaviors: the periodic turbulent flow past a circular cylinder and the unsteady turbulent flow in a channel-driven cavity. In both cases, the parameter considered is the Reynolds number and the machine learning-enhanced ROM considerably improved the pressure and velocity accuracy with respect to the standard ROM.
physics.flu-dyn
Driven by fundamental thermodynamic efficiency considerations, an emerging trend in the energy and propulsion systems is that the working fluid operates at a pressure above the critical pressure. Energy transport is thus accompanied by dramatic and strongly nonlinear variations of thermophysical properties, which cause abnormal heat transfer behavior and non-ideal gas effects. This situation raises a crucial challenge for the heat exchanger and turbomachinery design, overall energy and exergy efficiency. We aim to provide a multi-scale overview of the flow and thermal behavior of fluid above the critical point: microscopic physics and macroscopic transport. Microscopic physics, i.e. near-critical thermophysical properties, phase transition and fluid dynamics, are introduced. A particular focus will be on the supercritical pseudo boiling process, which is a generalization of classical liquid-vapor phase transitions to non-equilibrium supercritical states. These new perspectives lead to a revised view of the state space. Further, recent results demonstrated the possibility of stable supercritical fluid interfaces without surface tension. On the macroscale, recent progress on the modeling of turbulent flow and convective heat transfer of supercritical fluids are summarized. Direct numerical simulation is able to offer insights into the physics of thermal fluids. We start with a description of fundamental fluid mechanics problems related to supercritical fluids. In addition, the heat transfer deterioration in supercritical fluids is found to be closely connected to the flow relaminarization by the non-uniform body-force. Finally, various modeling approaches such as recently developed advanced Reynolds-averaged turbulence modeling as well as machine-learning methods are summarized.
eess.SY
Turbulent diffusion causes particles placed in proximity to separate. We investigate the required swimming efforts to maintain an active particle close to its passively advected counterpart. We explore optimally balancing these efforts by developing a novel physics-informed reinforcement learning strategy and comparing it with prescribed control and physics-agnostic reinforcement learning strategies. Our scheme, coined the actor-physicist, is an adaptation of the actor-critic algorithm in which the neural network parameterized critic is replaced with an analytically derived physical heuristic function, the physicist. We validate the proposed physics-informed reinforcement learning approach through extensive numerical experiments in both synthetic BK and more realistic Arnold-Beltrami-Childress flow environments, demonstrating its superiority in controlling particle dynamics when compared to standard reinforcement learning methods.
physics.flu-dyn
The precise simulation of turbulent flows is of immense importance in a variety of scientific and engineering fields, including climate science, freshwater science, and the development of energy-efficient manufacturing processes. Within the realm of turbulent flow simulation, direct numerical simulation (DNS) is widely considered to be the most reliable approach, but it is prohibitively expensive for long-term simulation at fine spatial scales. Given the pressing need for efficient simulation, there is an increasing interest in building machine learning models for turbulence, either by reconstructing DNS from alternative low-fidelity simulations or by predicting DNS based on the patterns learned from historical data. However, standard machine learning techniques remain limited in capturing complex spatio-temporal characteristics of turbulent flows, resulting in limited performance and generalizability. This paper presents a novel physics-enhanced neural operator (PENO) that incorporates physical knowledge of partial differential equations (PDEs) to accurately model flow dynamics. The model is further refined by a self-augmentation mechanism to reduce the accumulated error in long-term simulations. The proposed method is evaluated through its performance on two distinct sets of 3D turbulent flow data, showcasing the model's capability to reconstruct high-resolution DNS data, maintain the inherent physical properties of flow transport, and generate flow simulations across various resolutions. Additionally, experimental results on multiple 2D vorticity flow series, generated by different PDEs, highlight the transferability and generalizability of the proposed method. This confirms its applicability to a wide range of real-world scenarios in which extensive simulations are needed under diverse settings.
cs.LG
Recent advancements in operator-type neural networks have shown promising results in approximating the solutions of spatiotemporal Partial Differential Equations (PDEs). However, these neural networks often entail considerable training expenses, and may not always achieve the desired accuracy required in many scientific and engineering disciplines. In this paper, we propose a new learning framework to address these issues. A new spatiotemporal adaptation is proposed to generalize any Fourier Neural Operator (FNO) variant to learn maps between Bochner spaces, which can perform an arbitrary-length temporal super-resolution for the first time. To better exploit this capacity, a new paradigm is proposed to refine the commonly adopted end-to-end neural operator training and evaluations with the help from the wisdom from traditional numerical PDE theory and techniques. Specifically, in the learning problems for the turbulent flow modeled by the Navier-Stokes Equations (NSE), the proposed paradigm trains an FNO only for a few epochs. Then, only the newly proposed spatiotemporal spectral convolution layer is fine-tuned without the frequency truncation. The spectral fine-tuning loss function uses a negative Sobolev norm for the first time in operator learning, defined through a reliable functional-type a posteriori error estimator whose evaluation is exact thanks to the Parseval identity. Moreover, unlike the difficult nonconvex optimization problems in the end-to-end training, this fine-tuning loss is convex. Numerical experiments on commonly used NSE benchmarks demonstrate significant improvements in both computational efficiency and accuracy, compared to end-to-end evaluation and traditional numerical PDE solvers under certain conditions. The source code is publicly available at https://github.com/scaomath/torch-cfd.
eess.SY
Channel turbulence is a formidable obstacle for free-space optical (FSO) communication. Anticipation of turbulence levels is highly important for mitigating disruptions but has not been demonstrated without dedicated, auxiliary hardware. We show that machine learning (ML) can be applied to raw FSO data streams to rapidly predict channel turbulence levels with no additional sensing hardware. FSO was conducted through a controlled channel in the lab under six distinct turbulence levels, and the efficacy of using ML to classify turbulence levels was examined. ML-based turbulence level classification was found to be >98% accurate with multiple ML training parameters. Classification effectiveness was found to depend on the timescale of changes between turbulence levels but converges when turbulence stabilizes over about a one minute timescale.
physics.flu-dyn
Engineering design and scientific analysis rely upon computer simulations of turbulent fluid flows using turbulence models. These turbulence models are empirical and approximate, leading to large uncertainties in their predictions that hamper scientific and engineering advances. We outline a Physics Constrained Deep Learning framework to estimate turbulence model uncertainties using physics based Eigenspace Perturbations along with Deep Learning based guidance. The Deep Learning based modulation controls the spatial variation in perturbation magnitude to improve the calibration of uncertainty estimates over the state of the art physics based methods.
physics.plasm-ph
Generative artificial intelligence methods are employed for the first time to construct a surrogate model for plasma turbulence that enables long time transport simulations. The proposed GAIT (Generative Artificial Intelligence Turbulence) model is based on the coupling of a convolutional variational auto-encoder, that encodes precomputed turbulence data into a reduced latent space, and a recurrent neural network and decoder that generates new turbulence states 400 times faster than the direct numerical integration. The model is applied to the Hasegawa-Wakatani (HW) plasma turbulence model, that is closely related to the quasigeostrophic model used in geophysical fluid dynamics. Very good agreement is found between the GAIT and the HW models in the spatio-temporal Fourier and Proper Orthogonal Decomposition spectra, and the flow topology characterized by the Okubo-Weiss decomposition. The GAIT model also reproduces Lagrangian transport including the probability distribution function of particle displacements and the effective turbulent diffusivity.