arXiv subjects

Gianluca Iaccarino

Publications and source records attributed to Gianluca Iaccarino.

At least 19 recordsLinked to original sources

arXiv2026

Physically Organized Latent Spaces in Unsupervised Autoencoders: Evidence from Aerodynamic Databases

Using aerodynamic databases of progressively increasing complexity - from linear analytical theory to nonlinear separated RANS flows - we demonstrate that an autoencoder, trained solely by reconstruction, can spontaneously organize the latent representation according to physically meaningful variables and aerodynamic laws. Although the latent coordinates themselves vary under random initialization, the underlying physical organization is reproducible, even quantitatively. In addition, we present a three-dimensional latent space representation of the airfoil flow obtained by an unsupervised autoencoder with sequential training in which the inviscid field is decoupled from the boundary layer effects with the third latent variable strongly correlated with the Reynolds number and aerodynamic drag. Cross-validated regression and affine-alignment measures show that physical parameters and responses are strongly and linearly encoded in the learned coordinates, consistently across independent trainings.

physics.flu-dyn↗

arXiv2026

An LES model with finite-rate phase change and subgrid spray based on a thermodynamically consistent four-equation multiphase model

In this work, an LES model with finite-rate phase change and subgrid spray based on a high-resolution numerical scheme for multiphase multi-component simulations which satisfies interface equilibrium and phase immiscibility conditions is proposed. The multiphase model is based on a robust implementation of the four-equation multiphase model which assumes a strict subgrid equilibrium of pressure, temperature, and velocity. Critically, the equilibrium assumptions of the four-equation model provide large computational savings compared to modeling the full non-equilibrium multiphase system. To obtain predictive capabilities with these restrictive equilibrium assumptions, a new phase-confined form of the Eulerian $Σ$ spray model is proposed to predict subgrid interfacial surface area while avoiding unphysical leakage across interfaces. Additionally, an improved finite rate phase change model which is thermodynamically bounded by the equilibration of the Gibbs-free energy is coupled with the $Σ$ equation to model complex phase change regimes. The full modeling framework is validated using the Engine Combustion Network (ECN) Spray A case in non-evaporating and evaporating conditions and shows excellent agreement with experimental measurements.

physics.flu-dyn↗

arXiv2026

Generative prediction of laser-induced rocket ignition with dynamic latent space representations

Accurate and predictive scale-resolving simulations of laser-ignited rocket engines are highly time-consuming because the problem includes turbulent fuel-oxidizer mixing dynamics, laser-induced energy deposition, and high-speed flame growth. This is conflated with the large design space primarily corresponding to the laser operating conditions and target location. To enable rapid exploration and uncertainty quantification, we propose a data-driven surrogate modeling approach that combines convolutional autoencoders (cAEs) with neural ordinary differential equations (neural ODEs). The present target application of an ML-based surrogate model to leading-edge multi-physics turbulence simulation is part of a paradigm shift in the deployment of surrogate models towards increasing real-world complexity. Sequentially, the cAE spatially compresses high-dimensional flow fields into a low-dimensional latent space, wherein the system's temporal dynamics are learned via neural ODEs. Once trained, the model generates fast spatiotemporal predictions from initial conditions and specified operating inputs. By learning a surrogate to replace the entirety of the time-evolving simulation, the cost of predicting an ignition trial is reduced by several orders of magnitude, allowing efficient exploration of the input parameter space. Further, as the current framework yields a spatiotemporal field prediction, appraisal of the model output's physical grounding is more tractable. This approach marks a significant step toward real-time digital twins for laser-ignited rocket combustors and represents surrogate modeling in a complex system context.

cs.LG↗

arXiv2025

Learning Aerodynamics for the Control of Flying Humanoid Robots

Robots with multi-modal locomotion are an active research field due to their versatility in diverse environments. In this context, additional actuation can provide humanoid robots with aerial capabilities. Flying humanoid robots face challenges in modeling and control, particularly with aerodynamic forces. This paper addresses these challenges from a technological and scientific standpoint. The technological contribution includes the mechanical design of iRonCub-Mk1, a jet-powered humanoid robot, optimized for jet engine integration, and hardware modifications for wind tunnel experiments on humanoid robots for precise aerodynamic forces and surface pressure measurements. The scientific contribution offers a comprehensive approach to model and control aerodynamic forces using classical and learning techniques. Computational Fluid Dynamics (CFD) simulations calculate aerodynamic forces, validated through wind tunnel experiments on iRonCub-Mk1. An automated CFD framework expands the aerodynamic dataset, enabling the training of a Deep Neural Network and a linear regression model. These models are integrated into a simulator for designing aerodynamic-aware controllers, validated through flight simulations and balancing experiments on the iRonCub-Mk1 physical prototype.

cs.RO↗

arXiv2024

Effect of span size in Scale Resolving Simulations of airfoil stall and post-stall

In aerodynamic design, the use of Computational Fluid Dynamics (CFD) tools has long relied on Reynolds-Averaged Navier-Stokes (RANS) and Unsteady RANS (URANS) turbulence models. However, these methods face limitations in predicting massively separated flow regions. As a result, there is growing emphasis on scale-resolving methods such as Large Eddy Simulation (LES), which have become more feasible due to advancements in computational resources. This study focuses on the LES of airfoil flows, specifically analyzing the influence of the spanwise extent of the computational domain on aerodynamic forces. We examine the NREL S826 airfoil at a chord Reynolds number of $Re_c = 10^5$ under different angles of attack, with simulations performed using the PyFR solver. Our results highlight the importance of adequate spanwise resolution, particularly in post-stall flow conditions, where larger span sizes are necessary to accurately capture three-dimensional flow structures. The findings suggest that for certain flow regimes, traditional two-dimensional statistical representations may not be sufficient, necessitating further investigation into 3D flow phenomena.

physics.flu-dyn↗

arXiv2024

Neural network-based closure models for large-eddy simulations with explicit filtering

Data from direct numerical simulations of turbulent flows are commonly used to train neural network-based models as subgrid closures for large-eddy simulations; however, models with low a priori accuracy have been observed to fortuitously provide better a posteriori results than models with high a priori accuracy. This anomaly can be traced to a dataset shift in the learning problem, arising from inconsistent filtering in the training and testing stages. We propose a resolution to this issue that uses explicit filtering of the nonlinear advection term in the large-eddy simulation momentum equations to control aliasing errors. Within the context of explicitly-filtered large-eddy simulations, we develop neural network-based models for which a priori accuracy is a good predictor of a posteriori performance. We evaluate the proposed method in a large-eddy simulation of a turbulent flow in a plane channel at $Re_τ = 180$. Our findings show that an explicitly-filtered large-eddy simulation with a filter-to-grid ratio of 2 sufficiently controls the numerical errors so as to allow for accurate and stable simulations.

physics.flu-dyn↗

arXiv2024

A systematic dataset generation technique applied to data-driven automotive aerodynamics

A novel strategy for generating datasets is developed within the context of drag prediction for automotive geometries using neural networks. A primary challenge in this space is constructing a training databse of sufficient size and diversity. Our method relies on a small number of starting data points, and provides a recipe to interpolate systematically between them, generating an arbitrary number of samples at the desired quality. We test this strategy using a realistic automotive geometry, and demonstrate that convolutional neural networks perform exceedingly well at predicting drag coefficients and surface pressures. Promising results are obtained in testing extrapolation performance. Our method can be applied to other problems of aerodynamic shape optimization.

cs.LG↗

arXiv2024

Compositional Generative Inverse Design

Inverse design, where we seek to design input variables in order to optimize an underlying objective function, is an important problem that arises across fields such as mechanical engineering to aerospace engineering. Inverse design is typically formulated as an optimization problem, with recent works leveraging optimization across learned dynamics models. However, as models are optimized they tend to fall into adversarial modes, preventing effective sampling. We illustrate that by instead optimizing over the learned energy function captured by the diffusion model, we can avoid such adversarial examples and significantly improve design performance. We further illustrate how such a design system is compositional, enabling us to combine multiple different diffusion models representing subcomponents of our desired system to design systems with every specified component. In an N-body interaction task and a challenging 2D multi-airfoil design task, we demonstrate that by composing the learned diffusion model at test time, our method allows us to design initial states and boundary shapes that are more complex than those in the training data. Our method generalizes to more objects for N-body dataset and discovers formation flying to minimize drag in the multi-airfoil design task. Project website and code can be found at https://github.com/AI4Science-WestlakeU/cindm.

cs.LG↗

arXiv2023

Simultaneous Identification and Denoising of Dynamical Systems

In recent years there has been a push to discover the governing equations dynamical systems directly from measurements of the state, often motivated by systems that are too complex to directly model. Although there has been substantial work put into such a discovery, doing so in the case of large noise has proved challenging. Here we develop an algorithm for Simultaneous Identification and Denoising of a Dynamical System (SIDDS). We infer the noise in the state measurements by requiring that the denoised data satisfies the dynamical system with an equality constraint. This is unlike existing work where the mismatch in the dynamical system is treated as a penalty in the objective. We assume the dynamics is represented in a pre-defined basis and develop a sequential quadratic programming approach to solve the SIDDS problem featuring a direct solution of KKT system with a specialized preconditioner. In addition, we show how we can include sparsity promoting regularization using an iteratively reweighted least squares approach. The resulting algorithm leads to estimates of the dynamical system that approximately achieve the Cramér-Rao lower bound and, with sparsity promotion, can correctly identify the sparsity structure for higher levels of noise than existing techniques. Moreover, because SIDDS decouples the data from the evolution of the dynamical system, we show how to modify the problem to accurately identify systems from low sample rate measurements. The inverse problem approach and solution framework used by SIDDS has the potential to be expanded to related problems identifying governing equations from noisy data.

math.OC↗

arXiv2022

Machine Learning to Predict Aerodynamic Stall

A convolutional autoencoder is trained using a database of airfoil aerodynamic simulations and assessed in terms of overall accuracy and interpretability. The goal is to predict the stall and to investigate the ability of the autoencoder to distinguish between the linear and non-linear response of the airfoil pressure distribution to changes in the angle of attack. After a sensitivity analysis on the learning infrastructure, we investigate the latent space identified by the autoencoder targeting extreme compression rates, i.e. very low-dimensional reconstructions. We also propose a strategy to use the decoder to generate new synthetic airfoil geometries and aerodynamic solutions by interpolation and extrapolation in the latent representation learned by the autoencoder.

physics.flu-dyn↗

arXiv2022

Spinning-enabled Wireless Amphibious Origami Millirobot

Wireless millimeter-scale origami robots that can locomote in narrow spaces and morph their shapes have recently been explored with great potential for biomedical applications. Existing millimeter-scale origami devices usually require separate geometrical components for locomotion and functions, which increases the complexity of the robotic systems and their operation upon limited locomotion modes. Additionally, none of them can achieve both on-ground and in-water locomotion. Here we report a magnetically actuated amphibious origami millirobot that integrates capabilities of spinning-enabled multimodal locomotion, controlled delivery of liquid medicine, and cargo transportation with wireless operation. This millirobot takes full advantage of the geometrical features and folding/unfolding capability of Kresling origami, a triangulated hollow cylinder, to fulfill multifunction: its geometrical features are exploited for generating omnidirectional locomotion in various working environments, including on unstructured ground, in liquids, and at air-liquid interfaces through rolling, flipping, and spinning-induced propulsion; the folding/unfolding is utilized as a pumping mechanism for integrated multifunctionality such as controlled delivery of liquid medicine; furthermore, the spinning motion provides a sucking mechanism for targeted solid cargo transportation. This origami millirobot breaks the conventional way of utilizing origami folding only for shape reconfiguration and integrates multiple functions in one simple body. We anticipate the reported magnetic amphibious origami millirobots have the potential to serve as minimally invasive devices for biomedical diagnoses and treatments.

cs.RO↗

arXiv2022

An integrated heterogeneous computing framework for ensemble simulations of laser-induced ignition

An integrated computational framework is introduced to study complex engineering systems through physics-based ensemble simulations on heterogeneous supercomputers. The framework is primarily designed for the quantitative assessment of laser-induced ignition in rocket engines. We develop and combine an implicit programming system, a compressible reacting flow solver, and a data generation/management strategy on a robust and portable platform. We systematically present this framework using test problems on a hybrid CPU/GPU machine. Efficiency, scalability, and accuracy of the solver are comprehensively assessed with canonical unit problems. Ensemble data management and autoencoding are demonstrated using a canonical diffusion flame case. Sensitivity analysis of the ignition of a turbulent, gaseous fuel jet is performed using a simplified, three-dimensional model combustor. Our approach unifies computer science, physics and engineering, and data science to realize a cross-disciplinary workflow. The framework is exascale-oriented and can be considered a benchmark for future computational science studies of real-world systems.

cs.CE↗

arXiv2021

Task-parallel in-situ temporal compression of large-scale computational fluid dynamics data

Present day computational fluid dynamics simulations generate extremely large amounts of data, sometimes on the order of TB/s. Often, a significant fraction of this data is discarded because current storage systems are unable to keep pace. To address this, data compression algorithms can be applied to data arrays containing flow quantities of interest to reduce the overall amount of storage. Compression methods either exactly reconstruct the original dataset (lossless compression) or provide an approximate representation of the original dataset (lossy compression). The matrix column interpolative decomposition (ID) can be implemented as a type of lossy compression for data matrices that factors the original data matrix into a product of two smaller factor matrices. One of these matrices consists of a subset of the columns of the original data matrix, while the other is a coefficient matrix which approximates the columns of the original data matrix as linear combinations of the selected columns. Motivating this work is the observation that the structure of ID algorithms makes them a natural fit for the asynchronous nature of task-based parallelism; they are able to operate independently on sub-domains of the system of interest and, as a result, provide varied levels of compression. Using the task-based Legion programming model, a single-pass ID algorithm (SPID) for CFD applications is implemented. Performance studies, scalability, and the accuracy of the compression algorithms are presented for an analytical Taylor-Green vortex problem, followed by a large-scale implementation of a compressible Taylor-Green vortex using a high-order Navier-Stokes solver. In both cases, compression factors exceeding 100 are achieved with relative errors at or below 10e-3. Moreover, strong and weak scaling results demonstrate that introducing SPID to solvers leads to negligible increases in runtime.

cs.CE↗

arXiv2020

Pass-efficient methods for compression of high-dimensional turbulent flow data

The future of high-performance computing, specifically on future Exascale computers, will presumably see memory capacity and bandwidth fail to keep pace with data generated, for instance, from massively parallel partial differential equation (PDE) systems. Current strategies proposed to address this bottleneck entail the omission of large fractions of data, as well as the incorporation of $\textit{in situ}$ compression algorithms to avoid overuse of memory. To ensure that post-processing operations are successful, this must be done in a way that a sufficiently accurate representation of the solution is stored. Moreover, in situations where the input/output system becomes a bottleneck in analysis, visualization, etc., or the execution of the PDE solver is expensive, the the number of passes made over the data must be minimized. In the interest of addressing this problem, this work focuses on the utility of pass-efficient, parallelizable, low-rank, matrix decomposition methods in compressing high-dimensional simulation data from turbulent flows. A particular emphasis is placed on using coarse representation of the data -- compatible with the PDE discretization grid -- to accelerate the construction of the low-rank factorization. This includes the presentation of a novel single-pass matrix decomposition algorithm for computing the so-called interpolative decomposition. The methods are described extensively and numerical experiments on two turbulent channel flow data are performed. In the first (unladen) channel flow case, compression factors exceeding $400$ are achieved while maintaining accuracy with respect to first- and second-order flow statistics. In the particle-laden case, compression factors of 100 are achieved and the compressed data is used to recover particle velocities.

physics.comp-ph↗

arXiv2019

Estimated aleatoric uncertainty from initial and inlet conditions for variable density mixing

Variable density flows occur in a variety of different systems with a wide range of scales, from astrophysics to atmospheric flows to inertial confinement fusion or reacting flows. Given the inherent limitations of RANS simulations, it is important to find ways to quantify the uncertainty in the predictions. The aleatoric uncertainty of two different variable density flows to inlet and initial conditions, respectively, is studied. The two cases are the turbulent mixing of a jet in a co-flow at small Atwood number $A$, and the Rayleigh-Taylor mixing in a tilted rocket rig at medium $A$. Uncertainty estimates are made for simulations using the BHR-1 turbulence model and compared to reference data. Estimated distributions of measurement uncertainties from an experimental report are used to do Monte-Carlo sampling for some inlet parameters of the jet in a co-flow case. The results show only little sensitivity to all inlet parameter perturbations, and even with increased perturbations the estimated uncertainty band does not fully match the experimental results. Therefore, the model predictions are very robust with respect to the inlet parameter variability. For the tilted rocket rig case, a perturbation to the initial interface location is varied over an expected range of parameter values. The simulation results are significantly more sensitive to the initial conditions than the jet in co-flow simulation results are to the inlet conditions. The estimated uncertainty band of the tilted rocket rig matches quite well the experimental data at nominal perturbation amplitude. The predictions of this Rayleigh-Taylor instability case hence are less robust to the initial interface perturbations.

physics.flu-dyn↗

arXiv2019

Immersed-finite-element method for deformable particle suspensions in viscous and viscoelastic media

Deformable elastic bodies in viscous and viscoelastic media constitute a large portion of synthetic and biological complex fluids. We present a parallelized 3D-simulation methodology which fully resolves the momentum balance in the solid and fluid domains. An immersed boundary algorithm is exploited known as the immersed finite element method (IFEM) which accurately determines the internal forces in the solid domain. The scheme utilized has the advantages of requiring no costly re-meshing, handling finite Reynold's number, as well as incorporating non-linear viscoelasticity in the fluid domain. Our algorithm is designed for computationally efficient simulation of multi-particle suspensions with mixed structure types. The internal force calculation in the solid domain in the IFEM is coupled with a finite volume based incompressible fluid solver, both of which are massively parallelized for distributed memory architectures. We performed extensive case studies to ensure the fidelity of our algorithm. Namely, a series of single particle simulations for capsules, red blood cells, and elastic solid deformable particles were conducted in viscous and viscoelastic media. All of our results are in excellent quantitative agreement with the corresponding reported data in the literature which are based on different simulation platforms. Furthermore, we assess the accuracy of multi-particle simulation of blood suspensions (red blood cells in plasma) with and without platelets. Finally, we present the results of a novel simulation of multiple solid deformable objects in a viscoelastic medium.

physics.comp-ph↗

arXiv2018

A Comprehensive Physics-Informed Machine Learning Framework for Predictive Turbulence Modeling

Although an increased availability of computational resources has enabled high-fidelity simulations of turbulent flows, the RANS models are still the dominant tools for industrial applications. However, the predictive capabilities of RANS models are limited by potential inaccuracy driven by hypotheses in the Reynolds stress closure. Recently, a Physics-Informed Machine Learning (PIML) approach has been proposed to learn the functional form of Reynolds stress discrepancy in RANS simulations based on available data. It has been demonstrated that the learned discrepancy function can be used to improve Reynolds stresses in different flows where data are not available. However, owing to a number of challenges, the improvements have been demonstrated only in the Reynolds stress prediction but not in the corresponding propagated quantities of interest. In this work, we introduce the procedures toward a complete PIML framework for predictive turbulence modeling, including learning Reynolds stress discrepancy function, predicting Reynolds stresses in different flows, and propagating to mean flow fields. The process of Reynolds stress propagation and predictive accuracy of the propagated velocity field are investigated. To improve the learning-prediction performance, the input features are enriched based on an integrity basis of invariants. The fully developed turbulent flow in a square duct is used as the test case. The discrepancy model is trained on flow fields obtained from several Reynolds numbers and evaluated on a duct flow at a Reynolds number higher than any of the training cases. The predicted Reynolds stresses are propagated to velocity field through RANS equations. Numerical results show excellent predictive performances in both Reynolds stresses and their propagated velocities, demonstrating the merits of the PIML approach in predictive turbulence modeling.

physics.flu-dyn↗

arXiv2018

Cutting the Double Loop: Theory and Algorithms for Reliability-Based Design Optimization with Statistical Uncertainty

Statistical uncertainties complicate engineering design -- confounding regulated design approaches, and degrading the performance of reliability efforts. The simplest means to tackle this uncertainty is double loop simulation; a nested Monte Carlo method that, for practical problems, is intractable. In this work, we introduce a flexible, general approximation technique that obviates the double loop. This approximation is constructed in the context of a novel theory of reliability design under statistical uncertainty: We introduce metrics for measuring the efficacy of RBDO strategies (effective margin and effective reliability), minimal conditions for controlling uncertain reliability (precision margin), and stricter conditions that guarantee the desired reliability at a designed confidence level. We provide a number of examples with open-source code to demonstrate our approaches in a reproducible fashion.

stat.ME↗