arXiv subjects

Adrián Corrochano

Publications and source records attributed to Adrián Corrochano.

11 recordsLinked to original sources

arXiv2026

Reduced-order modeling of a viscoelastic turbulent jet with hybrid machine learning models

Adding flexible polymers to a Newtonian solvent confers complex properties to the resulting solution. The additional complexity substantially increases the computational cost of numerical simulations, which often makes them prohibitively expensive. Here, we propose hybrid reduced-order models to accelerate simulations of viscoelastic turbulent jets. The model combines modal decompositions with deep networks: we use proper orthogonal decomposition to obtain a compact representation of the data, and a neural network is trained to predict the mode coefficients in the low-dimensional space. Results show that the hybrid model effectively captures the long-term behavior of the viscoelastic jet, that we demonstrate by computing relevant statistics of the jet. While small models are capable of predicting large-scale dynamics more than one-step at a time, thus facilitating greater accelerations, larger models are mandatory for forecasting smaller-scale dynamics, with skip connections the most effective strategy for deeper and generalizable models. The proposed methodology underpins the potential of hybrid approaches for compact and robust reduced-order models of viscoelastic turbulent jets.

physics.flu-dyn↗

arXiv2026

Coherent structures in Newtonian and viscoelastic turbulent planar jets

The addition of a small amount of long-chain polymers confers viscoelastic properties to Newtonian flows. The resulting non-Newtonian solution now exhibits different dynamics, such as enhanced mixing at low Reynolds, where elastic instabilities can trigger elastic turbulence even though inertial turbulence is absent. Here, we study this phenomenon in viscoelastic planar jets and, in particular, we do it from the perspective of coherent structures to understand how elastic turbulence is triggered and sustained, which remain barely explored in this setup. We introduce the spatio-temporal Koopman decomposition for extracting the dominant flow patterns, and we compare them with those from Newtonian planar jets at high Reynolds number. Global flow structures are similar between jets, with low-frequency streaks and high-frequency wave packets dominating the turbulent dynamics. However, structures are strikingly different in the near field, where elasticity-driven streaks affect the dynamics in the potential core of the viscoelastic planar jet, modifying the bulk flow and interacting with the flow instability. The analysis of the polymer field reveals stretched polymer filaments and centre-mode structures, which support the implication of the near-field streaks on sustaining elastic turbulence in three-dimensional viscoelastic planar jets.

physics.flu-dyn↗

arXiv2025

The coherent structures of EVP Fluid Flow Past a Circular Cylinder

This study investigates the impact of elasticity and plasticity on two-dimensional flow past a circular cylinder at Reynolds number $Re = 100$. Ten direct numerical simulations were performed using the Saramito-Herschel-Bulkley model to represent viscoelastic and elastoviscoplastic (EVP) fluids. The flow evolves from a periodic von Kármán vortex street to chaotic-like regimes. Proper Orthogonal Decomposition (POD) and Higher Order Dynamic Mode Decomposition (HODMD) are applied to extract dominant flow structures and their temporal dynamics. For viscoelastic fluids, increasing the Weissenberg number $Wi$ elongates the recirculation bubble and shifts it downstream, resulting in more intricate but still periodic behavior. In EVP fluids, seven cases explore variations in Bingham number Bn, solvent viscosity ratio $β_s$, and power law index n, aiming to qualitatively assess their influence rather than determine critical thresholds. Results indicate that stronger plastic effects, especially with n>1, lead to increased flow complexity. Three dynamic regimes are identified: (i) periodic; (ii) transitional, with elongated recirculation and disrupted periodicity; and (iii) fully complex, with breakdown of recirculation. Overall, the study highlights the interplay between inertia, elasticity, and yield stress in non-Newtonian flows past obstacles and identifies key parameters driving the transition from periodic to complex regimes.

physics.flu-dyn↗

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↗

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↗

arXiv2023

Deep Learning combined with singular value decomposition to reconstruct databases in fluid dynamics

Fluid Dynamics problems are characterized by being multidimensional and nonlinear. Therefore, experiments and numerical simulations are complex and time-consuming. Motivated by this, the need arises to find new techniques to obtain data in a simpler way and in less time. In this article, we present a novel methodology based on physical principles to reconstruct three-, four- and five-dimensional databases from a strongly sparse sensors as input. The methodology consists of combining Single Value Decomposition (SVD) with neural networks. The neural network used is characterized by a simple architecture based on combining two autoencoders that work in parallel and are joined in the last layer. This new algorithm has been proved with three databases with different dimensions and complexities: in an Atmospheric Boundary Layer (ABL) with a turbulence model and in the flow past a two- and a three-dimensional cylinder. Summarizing, this work proposes a new hybrid physics-based machine learning model with a simple, robust and generalizable architecture, which allows reconstructing databases from very few sensors and with a very low computational cost.

physics.flu-dyn↗

arXiv2023

Hierarchical Higher-Order Dynamic Mode Decomposition for Clustering and Feature Selection

In this work, a new algorithm based on the application of higher-order dynamic mode decomposition (HODMD) is proposed for feature selection and variables clustering in reacting flow simulations. The hierarchical HODMD (h-HODMD) performs a reduction of the model order, followed by the iterative selection of the best reconstructed variables thus creating clusters of features which can eventually be associated with distinct dynamical phenomena. Firstly, h-HODMD is combined with different data pre-processing techniques to assess their influence on the algorithm in terms of reconstruction error. Afterwards, the algorithm is applied to analyze three different databases obtained from numerical simulations of a non-premixed co-flow methane flame, and its performance are compared with the standard HODMD in terms of the achievable degree of reduction as well as in terms of reconstruction error. Results show that h-HODMD improves the reconstruction for all the variables when compared to the standard HODMD algorithm. This condition is achieved thanks to the iterative variables' clustering: finding dedicated modes for a specific group of features does in fact lead to a better reconstruction of the dynamics with respect to the case when the same (global) modes are used to reconstruct the entire set of variables. Finally, the clusters of variables found by means of h-HODMD are analyzed, and it is observed that the algorithm can group chemical species whose behavior is also consistent from a kinetic point of view. In fact, it allows for the possibility to formulate inexpensive reduced dynamical models for predicting flames liftoff, as well as for identifying the formation of local extinction and blowout conditions, to formulate accurate reduced models to describe the formation of pollutants in aviation, and for control purposes.

physics.flu-dyn↗

arXiv2023

A predictive physics-aware hybrid reduced order model for reacting flows

In this work, a new hybrid predictive Reduced Order Model (ROM) is proposed to solve reacting flow problems. This algorithm is based on a dimensionality reduction using Proper Orthogonal Decomposition (POD) combined with deep learning architectures. The number of degrees of freedom is reduced from thousands of temporal points to a few POD modes with their corresponding temporal coefficients. Two different deep learning architectures have been tested to predict the temporal coefficients, based on recursive (RNN) and convolutional (CNN) neural networks. From each architecture, different models have been created to understand the behavior of each parameter of the neural network. Results show that these architectures are able to predict the temporal coefficients of the POD modes, as well as the whole snapshots. The RNN shows lower prediction error for all the variables analyzed. The model was also found capable of predicting more complex simulations showing transfer learning capabilities.

cs.LG↗

arXiv2023

Mode selection in concentric jets. The steady-steady 1:2 resonant mode interaction with O(2) symmetry

In this article, a thorough characterization of the configuration composed by two concentric jets at a low Reynolds number is presented. The analysis comprises a layout with a wide range for the velocity ratio between the inner and outer jets, defined within the interval [0, 2], and also details the influence of the distance between jets, where the wall thicknesses separating the two jets is [0.5, 4]. Global linear stability analysis identifies the most significant modes driving the changes in the flow dynamics. The neutral lines revealing the critical Reynolds number connected to the presence of the main (steady and unsteady) flow bifurcations, which are presented by global azimuthal modes, show the high complexity of the problem under study, where hysteresis and other types of complex cycles are pointed out. Finally, the mode interaction is analysed, highlighting the presence of travelling waves emerging from the interaction of steady states, and the existence of robust heteroclinic cycles that are asymptotically stable. The high level of detail in the results presented, makes this work as a reference for future research development in the field of concentric jets.

physics.flu-dyn↗

arXiv2022

Higher order dynamic mode decomposition to model reacting flows

In this work, the application of the multi-dimensional higher order dynamic mode decomposition (HODMD) is proposed for the first time to analyse combustion databases. In particular, HODMD has been adapted and combined with other pre-processing techniques (generally used in machine learning), in light of the multivariate nature of the data. A truncation step separate the main dynamics driving the flow from less relevant non-linear dynamics. The method is applied to analyse a database obtained from a Computational Fluid Dynamics (CFD) simulation of an axisymmetric, time varying, non-premixed, co-flow methane flame carried out by means of a detailed kinetic mechanism. Results show that HODMD can reconstruct the main jet dynamics with a reduced number of relevant modes, able to reproduce the system dynamics. These modes are found to be representative for the main flow physics with two main advantages: (i) they provide for the possibility to achieve a strong simplification with respect to the high-dimensional input data, and at the same time (ii) a small reconstruction error with respect to the original dataset is observed. In addition, the method was also validated considering a reduced matrix obtained using Principal Component Analysis (PCA) based feature selection and the Varimax rotation. This validation also reveals that it is not important to have all the variables in the dataset, just a group of them is necessary to obtain the main dynamics of the system. This has an impact on feature selection and on the cost these methodologies for very massive data.

physics.flu-dyn↗

arXiv2022

On the generation and destruction mechanisms of arch vortices in urban fluid flows

Studying and interpreting the different flow patterns present in urban areas is becoming essential since they help develop new approaches to fight climate change through an improved understanding of the dynamics of the pollutants in urban environments. This study uses higher order dynamic mode decomposition (HODMD) to analyze a high-fidelity database of the turbulent flow in various simplified urban environments. The geometry simulated consists of two buildings separated by a certain distance. Three different cases have been studied, corresponding to the three different regimes identified in the bibliography. We recognize the characteristics of the well-known arch vortex forming on the leeward side of the first building and document possible generation and destruction mechanisms of this vortex based on the resulting temporal modes. These so-called vortex-generating and vortex-breaking modes are further analyzed via proper-orthogonal decomposition. We show that the arch vortex plays a prominent role in the dispersion of pollutants in urban environments, where its generation leads to an increase in their concentration; therefore, the reported mechanisms are of extreme importance in the context of urban sustainability.

physics.flu-dyn↗