arXiv subjects

Soledad Le Clainche

Publications and source records attributed to Soledad Le Clainche.

At least 37 records · Page 2Linked to original sources

arXiv2025

Generative artificial intelligence and hybrid models to accelerate LES in reactive flows: Application to hydrogen/methane combustion

With increasing emphasis on carbon neutrality, accurate and efficient combustion prediction has become essential for the design and optimization of new generation combustion systems. This study established a computational framework by combining large eddy simulation (LES) with a generative machine learning approach which integrates modal decomposition and neural network, enabling fast prediction of hydrogen-methane combustion. A canonical jet-in-hot-coflow burner was selected as the benchmark configuration. LES was performed using eddy dissipation concept model in conjunction with a 17-species and 58-step skeletal mechanism. Reasonable agreement between LES results and experimental data was obtained for temperature and species mass fraction, confirming the accuracy of the present LES results. Flow characteristics and flame structures were analyzed, providing a reference for choosing parameters in prediction. Proper orthogonal decomposition (POD) was used to extract dominant flow features, and a hybrid autoregressive model, which combines modal decomposition with a deep learning (POD-DL) was constructed to forecast the temporal evolution of the combustion field. Comparison between the predicted results and LES data, including instantaneous contours, radial distributions, histogram and relative root mean square error, demonstrated a reasonable agreement. The main complexity lies in capturing the chaotic and fine-scale structures inherent to turbulent combustion. To the authors' knowledge, this is the first application of such a hybrid generative model to reactive flow prediction, representing an important step toward using data-driven surrogates to accelerate CFD simulations in combustion research. The proposed approach achieves speed-up ratios of 121 and 845 relative to LES for two tested cases. The implementation will be integrated into the upcoming release of the ModelFLOWs-app.

physics.flu-dyn↗

arXiv2025

Ensemble Kalman Filter for Data Assimilation coupled with low-resolution computations techniques applied in Fluid Dynamics

This paper presents an innovative Reduced-Order Model (ROM) for merging experimental and simulation data using Data Assimilation (DA) to estimate the "True" state of a fluid dynamics system, leading to more accurate predictions. Our methodology introduces a novel approach implementing the Ensemble Kalman Filter (EnKF) within a reduced-dimensional framework, grounded in a robust theoretical foundation and applied to fluid dynamics. To address the substantial computational demands of DA, the proposed ROM employs low-resolution (LR) techniques to drastically reduce computational costs. This approach involves downsampling datasets for DA computations, followed by an advanced reconstruction technique based on low-cost Singular Value Decomposition (lcSVD). The lcSVD method, a key innovation in this paper, has never been applied to DA before and offers a highly efficient way to enhance resolution with minimal computational resources. Our results demonstrate significant reductions in both computation time and RAM usage through the LR techniques without compromising the accuracy of the estimations. For instance, in a turbulent test case, the LR approach with a compression rate of 15.9 can achieve a speed-up of 13.7 and a RAM compression of 90.9% while maintaining a low Relative Root Mean Square Error (RRMSE) of 2.6%, compared to 0.8% in the high-resolution (HR) reference. Furthermore, we highlight the effectiveness of the EnKF in estimating and predicting the state of fluid flow systems based on limited observations and low-fidelity numerical data. This paper highlights the potential of the proposed DA method in fluid dynamics applications, particularly for improving computational efficiency in CFD and related fields. Its ability to balance accuracy with low computational and memory costs makes it suitable for large-scale and real-time applications, such as environmental monitoring or aerospace.

cs.CE↗

arXiv2025

An Adaptive Framework for Autoregressive Forecasting in CFD Using Hybrid Modal Decomposition and Deep Learning

This work presents, to the best of the authors' knowledge, the first generalizable and fully data-driven adaptive framework designed to stabilize deep learning (DL) autoregressive forecasting models over long time horizons, with the goal of reducing the computational cost required in computational fluid dynamics (CFD) simulations.The proposed methodology alternates between two phases: (i) predicting the evolution of the flow field over a selected time interval using a trained DL model, and (ii) updating the model with newly generated CFD data when stability degrades, thus maintaining accurate long-term forecasting. This adaptive retraining strategy ensures robustness while avoiding the accumulation of predictive errors typical in autoregressive models. The framework is validated across three increasingly complex flow regimes, from laminar to turbulent, demonstrating from 30 \% to 95 \% reduction in computational cost without compromising physical consistency or accuracy. Its entirely data-driven nature makes it easily adaptable to a wide range of time-dependent simulation problems. The code implementing this methodology is available as open-source and it will be integrated into the upcoming release of the ModelFLOWs-app.

physics.flu-dyn↗

arXiv2025

HOSVD-SR: A Physics-Based Deep Learning Framework for Super-Resolution in Fluid Dynamics

In this work we present a novel methodology that combines Higher Order Singular Value Decomposition (HOSVD) with Deep Learning (DL) techniques for super-resolution in computational fluid dynamics (CFD) and sparse experimental datasets. This approach, referred to as HOSVD-SR 1, integrates modal decomposition techniques with Machine Learning (ML), creating a hybrid model grounded in the underlying physics of the studied phenomena and capable of enhancing data dimensionality. The proposed methodology leverages HOSVD, a robust variant of SVD, ideal for high dimensional data, which extracts the singular values and modes (spatial and temporal) associated with each dimension of the database in tensor form, reducing noise and addressing challenges related to turbulent flows. HOSVD is employed to capture the key physical patterns from a under-resolved fluid mechanics database. Each spatial mode matrix serves as input for a decoder/autoencoder type neural network trained to increase the dimensionality of the tensor and accurately reconstruct experimental and CFD-like databases. The HOSVD-SR methodology has been tested on both two- and three-dimensional numerical databases of flow past a circular cylinder, as well as an experimental database of circular cylinder wake flows under a turbulent flow regime. HOSVD-SR has successfully addressed both laminar and turbulent cases, outperforming a previously proposed SVD-based methodology in terms of accuracy. HOSVD-SR is physics-based, making it a robust and highly generalizable method that can also be implemented for other data generation approaches in fluid mechanics problems.

physics.flu-dyn↗

arXiv2025

Hybrid machine learning models based on physical patterns to accelerate CFD simulations: a short guide on autoregressive models

Accurate modeling of the complex dynamics of fluid flows is a fundamental challenge in computational physics and engineering. This study presents an innovative integration of High-Order Singular Value Decomposition (HOSVD) with Long Short-Term Memory (LSTM) architectures to address the complexities of reduced-order modeling (ROM) in fluid dynamics. HOSVD improves the dimensionality reduction process by preserving multidimensional structures, surpassing the limitations of Singular Value Decomposition (SVD). The methodology is tested across numerical and experimental data sets, including two- and three-dimensional (2D and 3D) cylinder wake flows, spanning both laminar and turbulent regimes. The emphasis is also on exploring how the depth and complexity of LSTM architectures contribute to improving predictive performance. Simpler architectures with a single dense layer effectively capture the periodic dynamics, demonstrating the network's ability to model non-linearities and chaotic dynamics. The addition of extra layers provides higher accuracy at minimal computational cost. These additional layers enable the network to expand its representational capacity, improving the prediction accuracy and reliability. The results demonstrate that HOSVD outperforms SVD in all tested scenarios, as evidenced by using different error metrics. Efficient mode truncation by HOSVD-based models enables the capture of complex temporal patterns, offering reliable predictions even in challenging, noise-influenced data sets. The findings underscore the adaptability and robustness of HOSVD-LSTM architectures, offering a scalable framework for modeling fluid dynamics.

physics.flu-dyn↗

arXiv2025

A low cost singular value decomposition based data assimilation technique for analysis of heterogeneous combustion data

This article applies low-cost singular value decomposition (lcSVD) for the first time, to the authors knowledge, on combustion reactive flow databases. The lcSVD algorithm is a novel approach to SVD, suitable for calculating high-resolution 2D or 3D proper orthogonal decomposition (POD) modes and temporal coefficients using data from sensors. Consequently, the computational cost associated with this technique is much lower compared to standard SVD. Additionally, for the analysis of full n-dimensional datasets, the method reduces data dimensionality by selecting a strategically reduced number of points from the original dataset through optimal sensor placement or uniform sampling before performing SVD. Moreover, the properties of data assimilation of heterogeneous databases of this method are illustrated using two distinct reactive flow test cases: a numerical database modeling an axisymmetric, time-varying laminar coflow flame with a fuel mixture of 65% methane and 35% nitrogen, using air as the oxidizer, and experimental data generated from a turbulent bluff-body-stabilized hydrogen flame. The computational speed-up and memory gains associated with the lcSVD algorithm compared to SVD can reach values larger than 10, with compression factors greater than 2000. Applying lcSVD for data assimilation to reconstruct the flow dynamics combining data from sensors with simulation measurements, we found errors smaller than 1% in the most relevant species modelling the flow.

physics.flu-dyn↗

arXiv2025

Transformer-Based Fault-Tolerant Control for Fixed-Wing UAVs Using Knowledge Distillation and In-Context Adaptation

This study presents a transformer-based approach for fault-tolerant control in fixed-wing Unmanned Aerial Vehicles (UAVs), designed to adapt in real time to dynamic changes caused by structural damage or actuator failures. Unlike traditional Flight Control Systems (FCSs) that rely on classical control theory and struggle under severe alterations in dynamics, our method directly maps outer-loop reference values -- altitude, heading, and airspeed -- into control commands using the in-context learning and attention mechanisms of transformers, thus bypassing inner-loop controllers and fault-detection layers. Employing a teacher-student knowledge distillation framework, the proposed approach trains a student agent with partial observations by transferring knowledge from a privileged expert agent with full observability, enabling robust performance across diverse failure scenarios. Experimental results demonstrate that our transformer-based controller outperforms industry-standard FCS and state-of-the-art reinforcement learning (RL) methods, maintaining high tracking accuracy and stability in nominal conditions and extreme failure cases, highlighting its potential for enhancing UAV operational safety and reliability.

cs.RO↗

arXiv2025

Generalization capabilities and robustness of hybrid models grounded in physics compared to purely deep learning models

This study investigates the generalization capabilities and robustness of purely deep learning (DL) models and hybrid models based on physical principles in fluid dynamics applications, specifically focusing on iteratively forecasting the temporal evolution of flow dynamics. Three autoregressive models were compared: a hybrid model (POD-DL) that combines proper orthogonal decomposition (POD) with a long-short term memory (LSTM) layer, a convolutional autoencoder combined with a convolutional LSTM (ConvLSTM) layer and a variational autoencoder (VAE) combined with a ConvLSTM layer. These models were tested on two high-dimensional, nonlinear datasets representing the velocity field of flow past a circular cylinder in both laminar and turbulent regimes. The study used latent dimension methods, enabling a bijective reduction of high-dimensional dynamics into a lower-order space to facilitate future predictions. While the VAE and ConvLSTM models accurately predicted laminar flow, the hybrid POD-DL model outperformed the others across both laminar and turbulent flow regimes. This success is attributed to the model's ability to incorporate modal decomposition, reducing the dimensionality of the data, by a non-parametric method, and simplifying the forecasting component. By leveraging POD, the model not only gained insight into the underlying physics, improving prediction accuracy with less training data, but also reduce the number of trainable parameters as POD is non-parametric. The findings emphasize the potential of hybrid models, particularly those integrating modal decomposition and deep learning, in predicting complex flow dynamics.

physics.flu-dyn↗

arXiv2024

A unified view of elastic and elasto-inertial turbulence in channel flows at low and moderate Reynolds numbers

The chaotic flow of elastic fluids at low Reynolds number (Re) is typically distinguished into elasto-inertial and elastic turbulence (EIT/ET). However, the clear separation among these two turbulent regimes in parallel flows with a gradual Re decrease remains elusive. Here, spanning Re over four orders of magnitude, we investigate the statistical and structural nature of the flows computed by fully resolved 3D numerical simulations, revealing that as inertia becomes sub-dominant all flows share similar dynamical properties. Supported by the results, we thus propose a unified view of fully developed ET and EIT in parallel flows.

physics.flu-dyn↗

arXiv2024

LC-SVD-DLinear: A low-cost physics-based hybrid machine learning model for data forecasting using sparse measurements

This article introduces a novel methodology that integrates singular value decomposition (SVD) with a shallow linear neural network for forecasting high resolution fluid mechanics data. The method, termed LC-SVD-DLinear, combines a low-cost variant of singular value decomposition (LC-SVD) with the DLinear architecture, which decomposes the input features-specifically, the temporal coefficients-into trend and seasonality components, enabling a shallow neural network to capture the non-linear dynamics of the temporal data. This methodology uses under-resolved data, which can either be input directly into the hybrid model or downsampled from high resolution using two distinct techniques provided by the methodology. Working with under-resolved cases helps reduce the overall computational cost. Additionally, we present a variant of the method, LC-HOSVD-DLinear, which combines a low-cost version of the high-order singular value decomposition (LC-HOSVD) algorithm with the DLinear network, designed for high-order data. These approaches have been validated using two datasets: first, a numerical simulation of three-dimensional flow past a circular cylinder at $Re = 220$; and second, an experimental dataset of turbulent flow passing a circular cylinder at $Re = 2600$. The combination of these datasets demonstrates the robustness of the method. The forecasting and reconstruction results are evaluated through various error metrics, including uncertainty quantification. The work developed in this article will be included in the next release of ModelFLOWs-app

physics.flu-dyn↗

arXiv2024

EigenHearts: Cardiac Diseases Classification Using EigenFaces Approach

In the realm of cardiovascular medicine, medical imaging plays a crucial role in accurately classifying cardiac diseases and making precise diagnoses. However, the field faces significant challenges when integrating data science techniques, as a significant volume of images is required for these techniques. As a consequence, it is necessary to investigate different avenues to overcome this challenge. In this contribution, we offer an innovative tool to conquer this limitation. In particular, we delve into the application of a well recognized method known as the EigenFaces approach to classify cardiac diseases. This approach was originally motivated for efficiently representing pictures of faces using principal component analysis, which provides a set of eigenvectors (aka eigenfaces), explaining the variation between face images. As this approach proven to be efficient for face recognition, it motivated us to explore its efficiency on more complicated data bases. In particular, we integrate this approach, with convolutional neural networks (CNNs) to classify echocardiography images taken from mice in five distinct cardiac conditions (healthy, diabetic cardiomyopathy, myocardial infarction, obesity and TAC hypertension). Performing a preprocessing step inspired from the eigenfaces approach on the echocardiography datasets, yields sets of pod modes, which we will call eigenhearts. To demonstrate the proposed approach, we compare two testcases: (i) supplying the CNN with the original images directly, (ii) supplying the CNN with images projected into the obtained pod modes. The results show a substantial and noteworthy enhancement when employing SVD for pre-processing, with classification accuracy increasing by approximately 50%.

eess.IV↗

arXiv2024

Modeling heart flow dynamics using numerical simulations to identify the vortex ring: a practical guide

In this study, we present a comprehensive numerical analysis of blood flow within human left ventricle models, with particular emphasis on optimizing simulation conditions to enhance the realism and computational efficiency of heart flow dynamics. The objective is to determine the most effective mesh configurations, flow conditions, and boundary settings necessary for accurately capturing the formation and behavior of the vortex ring, a pivotal element in ventricular flow dynamics. Utilizing a computational fluid mechanics approach, we review the influence of both idealized and patient specific geometries on simulation outcomes. It is imperative to consider the necessity of dynamic wall motion and the precise calibration of inlet and outlet boundary conditions, which must be designed to mimic physiological conditions as accurately as possible. These factors are of paramount importance in achieving a balance between computational resource demands and the fidelity of the simulations, thereby providing valuable insights for future cardiovascular modeling efforts.

physics.flu-dyn↗

arXiv2024

A Novel Data Augmentation Tool for Enhancing Machine Learning Classification: A New Application of the Higher Order Dynamic Mode Decomposition for Improved Cardiac Disease Identification

In this work, a data-driven, modal decomposition method, the higher order dynamic mode decomposition (HODMD), is combined with a convolutional neural network (CNN) in order to improve the classification accuracy of several cardiac diseases using echocardiography images. The HODMD algorithm is used first as feature extraction technique for the echocardiography datasets, taken from both healthy mice and mice afflicted by different cardiac diseases (Diabetic Cardiomyopathy, Obesity, TAC Hypertrophy and Myocardial Infarction). A total number of 130 echocardiography datasets are used in this work. The dominant features related to each cardiac disease were identified and represented by the HODMD algorithm as a set of DMD modes, which then are used as the input to the CNN. In a way, the database dimension was augmented, hence HODMD has been used, for the first time to the authors knowledge, for data augmentation in the machine learning framework. Six sets of the original echocardiography databases were hold out to be used as unseen data to test the performance of the CNN. In order to demonstrate the efficiency of the HODMD technique, two testcases are studied: the CNN is first trained using the original echocardiography images only, and second training the CNN using a combination of the original images and the DMD modes. The classification performance of the designed trained CNN shows that combining the original images with the DMD modes improves the results in all the testcases, as it improves the accuracy by up to 22%. These results show the great potential of using the HODMD algorithm as a data augmentation technique.

eess.IV↗

arXiv2024

Review on vortex dynamics in the left ventricle as an early diagnosis marker for heart diseases and its treatment outcomes

The heart is the central part of the cardiovascular network. Its role is to pump blood to various body organs. Many cardiovascular diseases occur due to an abnormal functioning of the heart. A diseased heart leads to severe complications and in some cases death of an individual. The medical community believes that early diagnosis and treatment of heart diseases can be controlled by referring to numerical simulations of image-based heart models. Computational Fluid Dynamics (CFD) is a commonly used tool for patient-specific simulations in the cardiac flows, and it can be equipped to allow a better understanding of flow patterns. In this paper, we review the progress of CFD tools to understand the flow patterns in healthy and dilated cardiomyopathic (DCM) left ventricles (LV). The formation of an asymmetric vortex in a healthy LV shows an efficient way of blood transport. The vortex pattern changes before any change in the geometry of LV is noticeable. This flow change can be used as a marker of DCM progression. We can conclude that understanding vortex dynamics in LV using various vortex indexes coupled with data-driven approaches can be used as an early diagnosis tool and improvement in DCM treatment.

physics.med-ph↗

arXiv2024

Intercepting Unauthorized Aerial Robots in Controlled Airspace Using Reinforcement Learning

The proliferation of unmanned aerial vehicles (UAVs) in controlled airspace presents significant risks, including potential collisions, disruptions to air traffic, and security threats. Ensuring the safe and efficient operation of airspace, particularly in urban environments and near critical infrastructure, necessitates effective methods to intercept unauthorized or non-cooperative UAVs. This work addresses the critical need for robust, adaptive systems capable of managing such threats through the use of Reinforcement Learning (RL). We present a novel approach utilizing RL to train fixed-wing UAV pursuer agents for intercepting dynamic evader targets. Our methodology explores both model-based and model-free RL algorithms, specifically DreamerV3, Truncated Quantile Critics (TQC), and Soft Actor-Critic (SAC). The training and evaluation of these algorithms were conducted under diverse scenarios, including unseen evasion strategies and environmental perturbations. Our approach leverages high-fidelity flight dynamics simulations to create realistic training environments. This research underscores the importance of developing intelligent, adaptive control systems for UAV interception, significantly contributing to the advancement of secure and efficient airspace management. It demonstrates the potential of RL to train systems capable of autonomously achieving these critical tasks.

cs.RO↗

arXiv2024

Automatic Cardiac Pathology Recognition in Echocardiography Images Using Higher Order Dynamic Mode Decomposition and a Vision Transformer for Small Datasets

Heart diseases are the main international cause of human defunction. According to the WHO, nearly 18 million people decease each year because of heart diseases. Also considering the increase of medical data, much pressure is put on the health industry to develop systems for early and accurate heart disease recognition. In this work, an automatic cardiac pathology recognition system based on a novel deep learning framework is proposed, which analyses in real-time echocardiography video sequences. The system works in two stages. The first one transforms the data included in a database of echocardiography sequences into a machine-learning-compatible collection of annotated images which can be used in the training stage of any kind of machine learning-based framework, and more specifically with deep learning. This includes the use of the Higher Order Dynamic Mode Decomposition (HODMD) algorithm, for the first time to the authors' knowledge, for both data augmentation and feature extraction in the medical field. The second stage is focused on building and training a Vision Transformer (ViT), barely explored in the related literature. The ViT is adapted for an effective training from scratch, even with small datasets. The designed neural network analyses images from an echocardiography sequence to predict the heart state. The results obtained show the superiority of the proposed system and the efficacy of the HODMD algorithm, even outperforming pretrained Convolutional Neural Networks (CNNs), which are so far the method of choice in the literature.

eess.IV↗

arXiv2023

Low-cost singular value decomposition with optimal sensor placement

This paper presents a new method capable of reconstructing datasets with great precision and very low computational cost using a novel variant of the singular value decomposition (SVD) algorithm that has been named low-cost SVD (lcSVD). This algorithm allows to reconstruct a dataset from a minimum amount of points, that can be selected randomly, equidistantly or can be calculated using the optimal sensor placement functionality that is also presented in this paper, which finds minimizing the reconstruction error to validate the calculated sensor positions. This method also allows to find the optimal number of sensors, aiding users in optimizing experimental data recollection. The method is tested in a series of datasets, which vary between experimental and numerical simulations, two- and three-dimensional data and laminar and turbulent flow, which have been used to demonstrate the capacity of this method based on its high reconstruction accuracy, robustness, and computational resource optimization. Maximum speed-up factors of 630 and memory reduction of 37% are found when compared to the application of standard SVD to the dataset. This method will be incorporated into ModelFLOWs-app's next version release.

cs.CE↗

arXiv2023

Data-driven assessment of arch vortices in simplified urban flows

Understanding flow structures in urban areas is widely recognized as a challenging concern due to its effect on urban development, air quality, and pollutant dispersion. In this study, state-of-the-art data-driven methods for modal analysis of simplified urban flows are used to study the dominant flow processes in these environments. Higher order dynamic mode decomposition (HODMD), a highly-efficient method to analyze turbulent flows, is used together with traditional techniques such as proper-orthogonal decomposition (POD) to analyze high-fidelity simulation data of a simplified urban environment. Furthermore, the spatio-temporal Koopman decomposition (STKD) will be applied to the temporal modes obtained with HODMD to perform spatial analysis. The flow interaction within the canopy influences the flow structures, particularly the arch vortex. The latter is a vortical structure generally found downstream of wall-mounted obstacles, which is generated as a consequence of flow separation. Therefore, the main objective of the present study is to characterize the mechanisms that promote these phenomena in urban areas with different geometries. Remarkably, among all the vortical structures identified by the HODMD algorithm, low- and high-frequency modes are classified according to their relation with the arch vortex. They are referred to as vortex-generator and vortex-breaker modes, respectively. This classification implies that one of the processes driving the formation and destruction of major vortical structures in between the buildings is the interaction between low- and high-frequency structures. The high energy revealed by the POD for the vortex-breaker modes points to this destruction process as the mechanism driving the flow dynamics.

physics.flu-dyn↗