Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 11
Encontrada(s) 1 página(s)

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

An indirect training approach for implicit constitutive modelling using recurrent neural networks and the virtual fields method

  • Lourenço, Rúben
  • Georgieva, Petia
  • Cueto, Elías
  • Andrade-Campos, A.
Accurate material description is crucial to achieve high-quality results in computational analysis software. Phenomenological constitutive laws generalize the material behaviour observed in simple mechanical tests. The resulting empirical expressions contain parameters that need to be calibrated through an inverse optimization process. Advancements in Digital Image Correlation (DIC) techniques have enabled the extraction of non-uniform multi-axial displacement fields, facilitating the development of heterogeneous mechanical specimens and unlocking access to richer material behaviour data. Despite these advancements, constitutive models are still limited by mathematical formulations and biases due to simplifying assumptions. Artificial Neural Networks (ANNs), as universal function approximators, could potentially drive a paradigm shift in the field. ANNs do not require explicit pre-formulations, hence avoiding the need for identifying unknown parameters. Moreover, ANNs are able to implicitly learn patterns purely from data, allowing to circumvent the biases induced by the simplifying assumptions of first-order principles. However, the training of ANNs for implicit constitutive modelling is not straightforward, given the impossibility of obtaining direct measurements of stress. This paper explores the integration of Recurrent Neural Networks (RNNs) with the Virtual Fields Method (VFM) for material modelling. This approach uses global force and displacement data to indirectly train the neural network, with the equilibrium being evaluated globally through the VFM loss function. The proposed method is (i) computationally efficient by not requiring Finite Element Analysis (FEA), (ii) compatible with both full-field measurements and numerically generated data, and (iii) able to handle experimental boundary conditions.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Thermodynamics-informed super-resolution of scarce temporal dynamics data

  • Bermejo-Barbanoj, Carlos
  • Moya, Beatriz
  • Badías, Alberto
  • Chinesta, Francisco
  • Cueto, Elías
We present a method to increase the resolution of measurements of a physical system and subsequently predict its time evolution using thermodynamics-aware neural networks. Our method uses adversarial autoencoders, which reduce the dimensionality of the full order model to a set of latent variables that are enforced to match a prior, for example a normal distribution. Adversarial autoencoders are seen as generative models, and they can be trained to generate high-resolution samples from low-resolution inputs, meaning they can address the so-called super-resolution problem. Then, a second neural network is trained to learn the physical structure of the latent variables and predict their temporal evolution. This neural network is known as a structure-preserving neural network. It learns the metriplectic-structure of the system and applies a physical bias to ensure that the first and second principles of thermodynamics are fulfilled. The integrated trajectories are decoded to their original dimensionality, as well as to the higher dimensionality space produced by the adversarial autoencoder and they are compared to the ground truth solution. The method is tested with two examples of flow over a cylinder, where the fluid properties are varied between both examples.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

A comparison of single and double generator formalisms for thermodynamics-informed neural networks

  • Urdeitx, Pau
  • Alfaro, Icíar
  • González, David
  • Chinesta, Francisco
  • Cueto, Elías
The development of inductive biases has been shown to be a very effective way to increase the accuracy and robustness of neural networks, particularly when they are used to predict physical phenomena. These biases significantly increase the certainty of predictions, decrease the error made and allow considerably smaller datasets to be used. There are a multitude of methods in the literature to develop these biases. One of the most effective ways, when dealing with physical phenomena, is to introduce physical principles of recognised validity into the network architecture. The problem becomes more complex without knowledge of the physical principles governing the phenomena under study. A very interesting possibility then is to turn to the principles of thermodynamics, which are universally valid, regardless of the level of abstraction of the description sought for the phenomenon under study. To ensure compliance with the principles of thermodynamics, there are formulations that have a long tradition in many branches of science. In the field of rheology, for example, two main types of formalisms are used to ensure compliance with these principles: one-generator and two-generator formalisms. In this paper we study the advantages and disadvantages of each, using classical problems with known solutions and synthetic data.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Graph neural networks informed locally by thermodynamics

  • Tierz, Alicia
  • Alfaro, Icíar
  • González, David
  • Chinesta, Francisco
  • Cueto, Elías
Thermodynamics-informed neural networks employ inductive biases for the enforcement of the first and second principles of thermodynamics. To construct these biases, a metriplectic evolution of the physical system under study is assumed. This provides excellent results, when compared to uninformed, black box networks. While the degree of accuracy can be increased in one or two orders of magnitude, in the case of graph networks, this requires assembling global Poisson and dissipation matrices, which breaks the local structure of such networks. In order to avoid this drawback, a local version of the metriplectic biases has been developed in this work, which avoids the aforementioned matrix assembly, thus preserving the node-by-node structure of the graph networks. We apply this framework for examples in the fields of solid and fluid mechanics. Our approach demonstrates significant computational efficiency and strong generalization capabilities, accurately making inferences on examples significantly different from those encountered during training.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

On the Feasibility of Foundational Models for the Simulation of Physical Phenomena

  • Tierz, Alicia
  • Martínez Iparraguirre, Mikel
  • Alfaro, Icíar
  • González Ibáñez, David
  • Chinesta, Francisco
  • Cueto, Elías
ABSTRACTWe explore the feasibility of foundation models for the simulation of physical phenomena, with emphasis on continuum (solid and fluid) mechanics. Although so‐called “learned simulators” have shown some success when applied to specific tasks, it remains to be studied to what extent they can undergo severe changes in domain shape, boundary conditions, and/or constitutive laws and still provide robust (i.e., hallucination‐free) and accurate results. In this paper, we perform an exhaustive study of these features, put ourselves in the worst‐case scenario, and study their resistance to such strong changes in their domain of application.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

A neural network architecture for physically-consistent haptic rendering

  • Hernández, Quercus
  • Lopes de Sousa Martins, Pedro
  • Tesan, Lucas
  • Alfaro, Icíar
  • González, David
  • Chinesta, Francisco
  • Cueto, Elías
We develop a method for realistic haptic rendering of generalized solids that employs graph neural networks. In order to give these neural networks the required realism, biases of different types are used. In the case of solids with (hyper)elastic, hence reversible, behaviour, we use Hamiltonian neural networks. These networks employ a cognitive bias that ensures energy conservation during the simulation. In the case of dissipative solids (particularly visco-hyperelastic solids) we extend the cognitive bias so that it takes into account entropy production. In this way, regardless of the type of solid considered, the developed neural networks are able to provide the haptic devices with the necessary information for a realistic and, above all, physically consistent rendering. The results are tested using a set of haptic gloves and virtual reality glasses, which have shown excellent realism, but not without certain limitations due to the state of the art in the development of hardware of this type.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Meshgraphnets informed locally by thermodynamics for the simulation of flows around arbitrarily shaped objects

  • Bermejo, Carlos
  • Badías, Alberto
  • González, David
  • Cueto, Elías
We present a thermodynamics-informed graph neural network framework for learning the time evolution of complex physical systems, incorporating thermodynamic structure via a nodal port-metriplectic formulation. Built upon the MeshGraphNet architecture, our method replaces the standard decoder with multiple specialized decoders that predict local energy and entropy gradients, along with Poisson and dissipative operators. These components are assembled at each graph node according to the GENERIC formalism, enforcing the first and second laws of thermodynamics. The framework is evaluated on two examples involving incompressible fluid flow past obstacles: one with varying cylindrical obstacles and another with obstacles of different types, not seen during training. The proposed model shows accurate long-term predictions, robust generalization to unseen geometries, and substantial speedups compared to traditional numerical solvers.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

On the under-reaching phenomenon in message passing neural PDE solvers: Revisiting the CFL condition

  • Tesán, Lucas
  • Martínez Iparraguirre, Mikel
  • González, David
  • Lopes De Sousa Martins, Pedro
  • Cueto, Elías
This paper proposes sharp lower bounds for the number of message passing iterations required in graph neural networks (GNNs) when solving partial differential equations (PDE). This significantly reduces the need for exhaustive hyperparameter tuning. Bounds are derived for the three fundamental classes of PDEs (hyperbolic, parabolic and elliptic) by relating the physical characteristics of the problem in question to the message-passing requirement of GNNs. In particular, we investigate the relationship between the physical constants of the equations governing the problem, the spatial and temporal discretisation and the message passing mechanisms in GNNs. When the number of message passing iterations is below these proposed limits, information does not propagate efficiently through the network, resulting in poor solutions, even for deep GNN architectures. In contrast, when the suggested lower bound is satisfied, the GNN parameterisation allows the model to accurately capture the underlying phenomenology, resulting in solvers of adequate accuracy. Examples are provided for four different examples of equations that show the sharpness of the proposed lower bounds.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Variational rank reduction autoencoders for generative thermal design

  • Tierz, Alicia
  • Mounayer, Jad
  • Moya, Beatriz
  • Chinesta, Francisco
Generative thermal design for complex geometries is fundamental in many areas of engineering, yet it faces two main challenges: the high computational cost of high-fidelity simulations and the limitations of conventional generative models. Approaches such as autoencoders (AEs) and variational autoencoders (VAEs) often produce unstructured latent spaces with discontinuities, which restricts their capacity to explore designs and generate physically consistent solutions. To address these limitations, we propose a hybrid framework that combines Variational Rank-Reduction Autoencoders (VRRAEs) with Deep Operator Networks (DeepONets). The VRRAE introduces a truncated SVD within the latent space, leading to continuous, interpretable, and well-structured representations that mitigate posterior collapse and improve geometric reconstruction. The DeepONet then exploits this compact latent encoding in its branch network, together with spatial coordinates in the trunk network, to predict temperature gradients efficiently and accurately. This hybrid approach not only enhances the quality of generated geometries and the accuracy of gradient prediction, but also provides a substantial advantage in inference efficiency compared to traditional numerical solvers. Overall, the study underscores the importance of structured latent representations for operator learning and highlights the potential of combining generative models and operator networks in thermal design and broader engineering applications.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Can transformers overcome the lack of data in the simulation of history-dependent flows?

  • Urdeitx, Pau
  • Alfaro, Icíar
  • González, David
  • Chinesta, Francisco
  • Cueto, Elías
It is well known that the lack of information about certain variables necessary for the description of a dynamical system leads to the introduction of historical dependence (lack of Markovian character of the model) and noise. Traditionally, scientists have made up for these shortcomings by designing phenomenological variables that take into account this historical dependence (typically, conformational tensors in fluids). Often these phenomenological variables are not easily measurable experimentally. In this work we study to what extent Transformer architectures are able to cope with the lack of experimental data on these variables. The methodology is evaluated on three benchmark problems: a cylinder flow with no history dependence, a viscoelastic Couette flow modeled via the Oldroyd-B formalism, and a non-linear polymeric fluid described by the FENE model. Our results show that the Transformer outperforms a thermodynamically consistent, structurepreserving neural network with metriplectic bias in systems with missing experimental data, providing lower errors even in low-dimensional latent spaces. In contrast, for systems whose state variables can be fully known, the metriplectic model achieves superior performance.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Variational rank reduction autoencoders

  • Mounayer, Jad
  • Tierz, Alicia
  • Tomezyk, Jerome
  • Ghnatios, Chady
  • Chinesta, Francisco
We introduce Variational Rank Reduction Autoencoders (VRRAEs), a generative model that combines the structured latent regularization of deterministic Rank Reduction Autoencoders (RRAEs) with the probabilistic framework of Variational Autoencoders (VAEs). VRRAEs apply truncated singular value decomposition (SVD) in the latent space while incorporating Kullback–Leibler (KL) divergence regularization, enabling both controlled representation learning and effective stochastic generation. Empirically, VRRAEs outperform both RRAEs and VAEs across multiple benchmarks, achieving improvements in FID score on MNIST, CelebA, and CIFAR-10 on most of the tasks (up to % improvement on CelebA). Additionally, VRRAEs demonstrate increased robustness to posterior collapse, reducing test reconstruction error by 62 % on a synthetic dataset designed to expose this failure mode. Qualitative results further show smoother and more coherent latent interpolations. These findings suggest that SVD-based latent regularization not only enhances generative quality but also stabilizes training in variational frameworks. We provide an open-source JAX (Equinox) implementation at https://github.com/JadM133/RRAEs.