arXiv subjects

Petros Koumoutsakos

Publications and source records attributed to Petros Koumoutsakos.

At least 19 recordsLinked to original sources

arXiv2026

Data-driven impeller model for efficient large eddy simulations of metastable von Kármán flows

The von Kármán turbulent swirling flow exhibits intriguing large-scale metastable dynamics, including low-frequency state switching. The study of state switching demands long-duration high-fidelity simulations at high Reynolds numbers that capture the flow generated by the impellers. Blade-resolved Large Eddy Simulations (LES) are computationally prohibitive, limiting access to these slow dynamics. Here, we develop a model for the action of the impellers on the flow using experimental data from Particle Image Velocimetry (PIV) and torque measurements of the von Kármán flow. The impeller-region velocity is parametrized via B-splines and coupled to the LES through momentum forcing. An initial set of B-spline coefficients is inferred using the Optimizing a DIscrete Loss (ODIL) framework constrained by the Reynolds-Averaged Navier--Stokes (RANS) equations, PIV measurements in the optically accessible portion of the device, and impeller torque measurements. The coefficients are then refined by the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), which minimizes the discrepancy between the LES time-averaged velocity and torque and their experimental counterparts. Using the data-driven impeller model, we perform long-duration LES of the von Kármán flow. We find that the simulation reproduces the mean flow in the bulk and displays metastable state-switching dynamics. We further show that these metastable states are not axisymmetric and consist of an alternating four-cell flow pattern that slowly rotates around the axis of the cylindrical vessel. The proposed approach provides a practical and computationally efficient route to investigating large-scale dynamics in impeller-driven turbulent flows.

physics.flu-dyn↗

arXiv2026

A hybrid proper orthogonal decomposition and diffusion framework for reduced-order forecasting of turbulent flow dynamics

Forecasting turbulent flow dynamics requires a balance between predictive fidelity and computational efficiency. Diffusion-based generative models can represent complex spatiotemporal dynamics, but their application to high-dimensional turbulent flows remains computationally expensive. In contrast, proper orthogonal decomposition (POD) provides compact, physically interpretable reduced-order representations, although aggressive modal truncation can remove relevant flow structures. This work introduces a hybrid reduced-order generative forecasting framework that combines POD with Generative Learning of Effective Dynamics (G-LED). The method performs temporal prediction in a physics-based modal space and uses diffusion-based reconstruction to recover physically meaningful flow-field representations. It is assessed using experimental measurements of the turbulent wake behind a circular cylinder. Three configurations are compared: full-field G-LED, global POD-G-LED, and localized POD-G-LED. Full-field G-LED provides the highest fidelity, preserving richer vorticity fluctuations and more consistent turbulent kinetic energy distributions, but requires approximately 17 h for diffusion-model training, 7 h for Transformer training, and 3 min to predict 100 future snapshots. By transferring prediction to a reduced POD space, global POD-G-LED reduces these costs to approximately 8 h, 2 h, and 50 s, respectively, while retaining dominant wake organization and coherent energetic structures. A localized POD-G-LED formulation assigns different modal resolutions to distinct wake regions and improves vorticity statistics and energetic distributions relative to the global reduced-order configuration. These results show that coupling physics-based modal representations with diffusion-based generative reconstruction offers an effective route to efficient turbulent-flow forecasting.

physics.flu-dyn↗

arXiv2026

A geometry-aligned multi-fidelity framework for uncertainty quantification of wildfire spread

Forward propagation of input uncertainties in physics-based wildfire models is computationally prohibitive, limiting the use of high-fidelity simulators in risk assessment workflows. This work introduces a geometry-aligned bi-fidelity surrogate framework that addresses the convection-dominated nature of wildfire spread by mapping low- and high-fidelity solution snapshots onto a common reference domain prior to basis selection and reconstruction. Unlike conventional bi-fidelity schemes, which combine spatially shifted snapshots and thus suffer from oscillations and excess basis requirements near sharp fronts, the proposed mapping aligns the dominant front geometry through per-variable shift/stretch transforms in 1D and an activity indicator-based affine alignment in 2D, so that reduced bases compare physically corresponding structures rather than displaced ones. Building on the ADfiRe physics-based simulator, we demonstrate the method on 1D and 2D test cases in which low- and high-fidelity models differ in mesh resolution and physical completeness. Across both settings, the geometry-aligned surrogate reproduces full-field temperature and fuel composition with substantially lower error than its unmapped counterpart, eliminates Gibbs-type oscillations near steep gradients, and recovers high-fidelity probability density functions for key quantities of interest (e.g., maximum temperature, evaporated moisture, and burned area). After offline training, online predictions are roughly three orders of magnitude cheaper than direct high-fidelity evaluation, making the framework a practical building block for many-query uncertainty quantification once the offline cost is amortized over enough queries. We discuss the conditions under which the geometric alignment is most effective, its limitations for non-convex or topologically complex fronts, and the path toward validation against real data.

cs.CE↗

arXiv2026

Bayesian Inference for PDE-based Inverse Problems using the Optimization of a Discrete Loss

Inverse problems are crucial for many applications in science, engineering and medicine that involve data assimilation, design, and imaging. Their solution infers the parameters or latent states of a complex system from noisy data and partially observable processes. When measurements are an incomplete or indirect view of the system, additional knowledge is required to accurately solve the inverse problem. Adopting a physical model of the system in the form of partial differential equations (PDEs) is a potent method to close this gap. In particular, the method of optimizing a discrete loss (ODIL) has shown great potential in terms of robustness and computational cost. In this work, we introduce B-ODIL, a Bayesian extension of ODIL, that integrates the PDE loss of ODIL as prior knowledge and combines it with a likelihood describing the data. B-ODIL employs a Bayesian formulation of PDE-based inverse problems to infer solutions with quantified uncertainties. We demonstrate the capabilities of B-ODIL in a series of synthetic benchmarks involving PDEs in one, two, and three dimensions. We showcase the application of B-ODIL in estimating tumor concentration and its uncertainty in a patient's brain from MRI scans using a three-dimensional tumor growth model.

stat.ME↗

arXiv2026

Structure tensor Reynolds-averaged Navier-Stokes turbulence models with equivariant neural networks

Accurate and generalizable Reynolds-averaged Navier-Stokes (RANS) models for turbulent flows rely on effective closures, but currently available closures are notoriously unreliable. Kassinos et al. (J. Fluid Mechanics, 428, pp. 213-248, 2001) hypothesized that this unreliability of RANS models was due to an insufficient description of the statistical state of the turbulence and proposed a set of structure tensors as a candidate for a sufficiently rich description. To test this hypothesis for the rapid pressure-strain term, we introduce tensor-based, symmetry aware closures in terms of the structure tensors using equivariant neural networks (ENNs), and present an algorithm for enforcing algebraic contraction relations among tensor components. Using data from rapid distortion theory, experiments show that such ENNs can effectively learn relationships involving high-order tensors. The resulting ENN structure tensor models are orders of magnitude more accurate than existing models for the rapid pressure-strain correlation, effectively validating the Kassinos et al. hypothesis for this term. Results show that ENNs provide a physically consistent alternative to classical tensor basis models, enabling end-to-end learning of unclosed terms in RANS and other tensor modeling domains, and rapid exploration of model dependencies.

physics.flu-dyn↗

arXiv2026

Prediction of Extreme Events in Multiscale Simulations of Geophysical Turbulence using Reinforcement Learning

Accurate subgrid-scale closures are essential for weather/climate models, where predicting extreme events is critical. Traditional closures have structural errors, e.g., producing excessive diffusion that dampens extremes. Artificial intelligence has gained attention for closure modeling, but the prediction of extreme events remains challenging. Supervised offline learning needs abundant high-fidelity training data and can lead to instabilities. Online learning algorithms are emerging as an alternative, but reliance on differentiable numerical solvers or scalable optimizers hinders broad use. Here, we introduce SMARL to develop closures for canonical prototypes of atmospheric/oceanic turbulence, using only the enstrophy spectrum, estimated from a few high-fidelity samples, as reward. This reward ensures that the model captures the cascades of scales in these simulations. These online-learned closures enable stable simulations, with up to five orders of magnitude fewer degrees of freedom, that reproduce high-fidelity simulation statistics and capture in particular extremes. We interpret the closures by analyzing the SMARL policy and demonstrate generalization to other flows. The results highlight SMARL as a potent tool for developing closures capable of capturing extremes in atmospheric/oceanic flows, opening new capabilities for effective climate modeling.

physics.geo-ph↗

arXiv2025

A Critical Assessment of Pattern Comparisons Between POD and Autoencoders in Intraventricular Flows

Understanding intraventricular hemodynamics requires compact and physically interpretable representations of the underlying flow structures, as characteristic flow patterns are closely associated with cardiovascular conditions and can support early detection of cardiac deterioration. Conventional visualization of velocity or pressure fields, however, provides limited insight into the coherent mechanisms driving these dynamics. Reduced-order modeling techniques, like Proper Orthogonal Decomposition (POD) and Autoencoder (AE) architectures, offer powerful alternatives to extract dominant flow features from complex datasets. This study systematically compares POD with several AE variants (Linear, Nonlinear, Convolutional, and Variational) using left ventricular flow fields obtained from computational fluid dynamics simulations. We show that, for a suitably chosen latent dimension, AEs produce modes that become nearly orthogonal and qualitatively resemble POD modes that capture a given percentage of kinetic energy. As the number of latent modes increases, AE modes progressively lose orthogonality, leading to linear dependence, spatial redundancy, and the appearance of repeated modes with substantial high-frequency content. This degradation reduces interpretability and introduces noise-like components into AE-based reduced-order models, potentially complicating their integration with physics-based formulations or neural-network surrogates. The extent of interpretability loss varies across the AEs, with nonlinear, convolutional, and variational models exhibiting distinct behaviors in orthogonality preservation and feature localization. Overall, the results indicate that AEs can reproduce POD-like coherent structures under specific latent-space configurations, while highlighting the need for careful mode selection to ensure physically meaningful representations of cardiac flow dynamics.

physics.flu-dyn↗

arXiv2025

Generative Urban Flow Modeling: From Geometry to Airflow with Graph Diffusion

Urban wind flow modeling and simulation play an important role in air quality assessment and sustainable city planning. A key challenge for modeling and simulation is handling the complex geometries of the urban landscape. Low order models are limited in capturing the effects of geometry, while high-fidelity Computational Fluid Dynamics (CFD) simulations are prohibitively expensive, especially across multiple geometries or wind conditions. Here, we propose a generative diffusion framework for synthesizing steady-state urban wind fields over unstructured meshes that requires only geometry information. The framework combines a hierarchical graph neural network with score-based diffusion modeling to generate accurate and diverse velocity fields without requiring temporal rollouts or dense measurements. Trained across multiple mesh slices and wind angles, the model generalizes to unseen geometries, recovers key flow structures such as wakes and recirculation zones, and offers uncertainty-aware predictions. Ablation studies confirm robustness to mesh variation and performance under different inference regimes. This work develops is the first step towards foundation models for the built environment that can help urban planners rapidly evaluate design decisions under densification and climate uncertainty.

cs.LG↗

arXiv2025

Scalable, Cloud-Based Simulations of Blood Flow and Targeted Drug Delivery in Retinal Capillaries

We investigate the capabilities of cloud computing for large-scale,tightly-coupled simulations of biological fluids in complex geometries, traditionally performed in supercomputing centers. We demonstrate scalable and efficient simulations in the public cloud. We perform meso-scale simulations of blood flow in image-reconstructed capillaries, and examine targeted drug delivery by artificial bacterial flagella (ABFs). The simulations deploy dissipative particle dynamics (DPD) with two software frameworks, Mirheo (developed by our team) and LAMMPS. Mirheo exhibits remarkable weak scalability for up to 512 GPUs. Similarly, LAMMPS demonstrated excellent weak scalability for pure solvent as well as for blood suspensions and ABFs in reconstructed retinal capillaries. In particular, LAMMPS maintained weak scaling above 90% on the cloud for up to 2,000 cores. Our findings demonstrate that cloud computing can support tightly coupled, large-scale scientific simulations with competitive performance.

physics.flu-dyn↗

arXiv2025

Individualizing Glioma Radiotherapy Planning by Optimization of Data and Physics-Informed Discrete Loss

Brain tumor growth is unique to each glioma patient and extends beyond what is visible in imaging scans, infiltrating surrounding brain tissue. Understanding these hidden patient-specific progressions is essential for effective therapies. Current treatment plans for brain tumors, such as radiotherapy, typically involve delineating a uniform margin around the visible tumor on pre-treatment scans to target this invisible tumor growth. This "one size fits all" approach is derived from population studies and often fails to account for the nuances of individual patient conditions. We present the GliODIL framework, which infers the full spatial distribution of tumor cell concentration from available multi-modal imaging, leveraging a Fisher-Kolmogorov type physics model to describe tumor growth. This is achieved through the newly introduced method of Optimizing the Discrete Loss, where both data and physics-based constraints are softly assimilated into the solution. Our test dataset comprises 152 glioblastoma patients with pre-treatment imaging and post-treatment follow-ups for tumor recurrence monitoring. By blending data-driven techniques with physics-based constraints, GliODIL enhances recurrence prediction in radiotherapy planning, challenging traditional uniform margins and strict adherence to the Fisher-Kolmogorov partial differential equation model, which is adapted for complex cases.

physics.med-ph↗

arXiv2025

Energy Matching: Unifying Flow Matching and Energy-Based Models for Generative Modeling

Current state-of-the-art generative models map noise to data distributions by matching flows or scores. A key limitation of these models is their inability to readily integrate available partial observations and additional priors. In contrast, energy-based models (EBMs) address this by incorporating corresponding scalar energy terms. Here, we propose Energy Matching, a framework that endows flow-based approaches with the flexibility of EBMs. Far from the data manifold, samples move from noise to data along irrotational, optimal transport paths. As they approach the data manifold, an entropic energy term guides the system into a Boltzmann equilibrium distribution, explicitly capturing the underlying likelihood structure of the data. We parameterize these dynamics with a single time-independent scalar field, which serves as both a powerful generator and a flexible prior for effective regularization of inverse problems. The present method substantially outperforms existing EBMs on CIFAR-10 and ImageNet generation in terms of fidelity, while retaining simulation-free training of transport-based approaches away from the data manifold. Furthermore, we leverage the flexibility of the method to introduce an interaction energy that supports the exploration of diverse modes, which we demonstrate in a controlled protein generation setting. This approach learns a scalar potential energy, without time conditioning, auxiliary generators, or additional networks, marking a significant departure from recent EBM methods. We believe this simplified yet rigorous formulation significantly advances EBMs capabilities and paves the way for their wider adoption in generative modeling in diverse domains.

cs.LG↗

arXiv2025

Ising energy model for the stochastic prediction of tumor islets

A major challenge in diagnosing and treating cancer is the infiltrative growth of tumors into surrounding tissues. This microscopic spread of the disease is invisible on most diagnostic imaging modalities and can often only be detected histologically in biopsies. The purpose of this paper is to develop a physically based model of tumor spread that captures the histologically observed behavior in terms of seeding small tumor islets in prostate cancer. The model is based on three elementary events: a tumor cell can move, duplicate, or die. The propensity of each event is given by an Ising-like Hamiltonian that captures correlations between neighboring cells. The model parameters were fitted to clinical data obtained from surgical specimens taken from 23 prostate cancer patients. The results demonstrate that this straightforward physical model effectively describes the distribution of the size and the number of tumor islets in prostate cancer. The simulated tumor islets exhibit a regular, approximately spherical shape, correctly mimicking the shapes observed in histology. This is due to the Ising interaction term between neighboring cells acting as a surface tension that gives rise to regularly shaped islets. The model addresses the important clinical need of calculating the probability of tumor involvement in specific sub-volumes of the prostate, which is required for radiation treatment planning and other applications.

physics.med-ph↗

arXiv2025

Data-Driven Discovery of Interpretable Kalman Filter Variants through Large Language Models and Genetic Programming

Algorithmic discovery has traditionally relied on human ingenuity and extensive experimentation. Here we investigate whether a prominent scientific computing algorithm, the Kalman Filter, can be discovered through an automated, data-driven, evolutionary process that relies on Cartesian Genetic Programming (CGP) and Large Language Models (LLM). We evaluate the contributions of both modalities (CGP and LLM) in discovering the Kalman filter under varying conditions. Our results demonstrate that our framework of CGP and LLM-assisted evolution converges to near-optimal solutions when Kalman optimality assumptions hold. When these assumptions are violated, our framework evolves interpretable alternatives that outperform the Kalman filter. These results demonstrate that combining evolutionary algorithms and generative models for interpretable, data-driven synthesis of simple computational modules is a potent approach for algorithmic discovery in scientific computing.

cs.NE↗

arXiv2025

Controlling Topological Defects in Polar Fluids via Reinforcement Learning

Topological defects in active polar fluids exhibit complex dynamics driven by internally generated stresses, reflecting the deep interplay between topology, flow, and non-equilibrium hydrodynamics. Feedback control offers a powerful means to guide such systems, enabling transitions between dynamic states. We investigated closed-loop steering of integer-charged defects in a confined active fluid by modulating the spatial profile of activity. Using a continuum hydrodynamic model, we show that localized control of active stress induces flow fields that can reposition and direct defects along prescribed trajectories by exploiting non-linear couplings in the system. A reinforcement learning framework is used to discover effective control strategies that produce robust defect transport across both trained and novel trajectories. The results highlight how AI agents can learn the underlying dynamics and spatially structure activity to manipulate topological excitations, offering insights into the controllability of active matter and the design of adaptive, self-organized materials.

cond-mat.soft↗

arXiv2025

Contactless Precision Steering of Particles in a Fluid inside a Cube with Rotating Walls

Contactless manipulation of small objects is essential for biomedical and chemical applications, such as cell analysis, assisted fertilisation, and precision chemistry. Established methods, including optical, acoustic, and magnetic tweezers, are now complemented by flow control techniques that use flow-induced motion to enable precise and versatile manipulation. However, trapping multiple particles in fluid remains a challenge. This study introduces a novel control algorithm capable of steering multiple particles in flow. The system uses rotating disks to generate flow fields that transport particles to precise locations. Disk rotations are governed by a feedback control policy based on the Optimising a Discrete Loss (ODIL) framework, which combines fluid dynamics equations with path objectives into a single loss function. Our experiments, conducted in both simulations and with the physical device, demonstrate the capability of the approach to transport two beads simultaneously to predefined locations, advancing robust contactless particle manipulation for biomedical applications.

physics.flu-dyn↗

arXiv2025

Data-driven shape inference in three-dimensional steady state supersonic flows using ODIL and JAX-Fluids

We present a novel data- and first-principles-driven method for inferring the shape of a solid obstacle and its flow field in three-dimensional steady-state supersonic flows. The method combines the Optimizing a Discrete Loss (ODIL) technique with the automatically differentiable JAX-Fluids CFD solver to jointly reconstruct flow fields and obstacle shapes. ODIL minimizes the discrete residual of the governing PDE via gradient descent-based algorithms and inherits the consistency and stability of the chosen numerical discretization. Discrete residuals and their gradients are computed using JAX-Fluids, which features nonlinear shock-capturing schemes and level-set-based immersed solid boundaries. We validate our method on synthetic data for challenging inverse problems, including shape inference of solid obstacles in 3D steady-state supersonic flows. In particular, we study flow around a cylinder, sphere, and ellipse. Two shape representations are investigated: (1) parametric, where the shape is described by a small set of parameters (e.g., radius of the cylinder or sphere) optimized jointly with the flow field, and (2) free-form, where the level-set function is optimized pointwise over the mesh without predefined shapes. For the parametric case, we provide a detailed comparison with Physics-Informed Neural Networks. We demonstrate that the combination of nonlinear shock-capturing discretization and level-set-based interface representation enables accurate inference of obstacle shapes and flow fields via the ODIL method. This approach opens new avenues for solving complex inverse problems in supersonic aerodynamics.

physics.flu-dyn↗

arXiv2025

Optimal Navigation in Microfluidics via the Optimization of a Discrete Loss

Optimal path planning and control of microscopic devices navigating in fluid environments is essential for applications ranging from targeted drug delivery to environmental monitoring. These tasks are challenging due to the complexity of microdevice-flow interactions. We introduce a closed-loop control method that optimizes a discrete loss (ODIL) in terms of dynamics and path objectives. In comparison with reinforcement learning, ODIL is more robust, up to three orders faster, and excels in high-dimensional action/state spaces, making it a powerful tool for navigating complex flow environments.

physics.comp-ph↗

arXiv2025

Reinforcement Learning Closures for Underresolved Partial Differential Equations using Synthetic Data

Partial Differential Equations (PDEs) describe phenomena ranging from turbulence and epidemics to quantum mechanics and financial markets. Despite recent advances in computational science, solving such PDEs for real-world applications remains prohibitively expensive because of the necessity of resolving a broad range of spatiotemporal scales. In turn, practitioners often rely on coarse-grained approximations of the original PDEs, trading off accuracy for reduced computational resources. To mitigate the loss of detail inherent in such approximations, closure models are employed to represent unresolved spatiotemporal interactions. We present a framework for developing closure models for PDEs using synthetic data acquired through the method of manufactured solutions. These data are used in conjunction with reinforcement learning to provide closures for coarse-grained PDEs. We illustrate the efficacy of our method using the one-dimensional and two-dimensional Burgers' equations and the two-dimensional advection equation. Moreover, we demonstrate that closure models trained for inhomogeneous PDEs can be effectively generalized to homogeneous PDEs. The results demonstrate the potential for developing accurate and computationally efficient closure models for systems with scarce data.

cs.LG↗