Deep learning model to assist multiphysics conjugate problems

Haber, George El; Viquerat, Jonathan; Larcher, Aurélien; Ryckelynck, David; Alves, José; Patil, Aakash; Hachem, Elie

doi:10.1063/5.0077723

Cited by 9 publications

(1 citation statement)

References 47 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Convolutional neural networks (CNNs) are known to outperform traditional fully-connected ANNs, and this has been demonstrated in various domains, including physics-based simulations (Guo et al, 2016;Deshpande et al, 2022a;Krokos et al, 2022b;El Haber et al, 2022). CNNs work on the principle of parameter sharing and local convolution operations, which enables efficient training on large inputs.…”

Section: Introductionmentioning

confidence: 99%

Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics

et al. 2023

View full text Add to dashboard Cite

Deep learning surrogate models are being increasingly used in accelerating scientific simulations as a replacement for costly conventional numerical techniques. However, their use remains a significant challenge when dealing with real-world complex examples. In this work, we demonstrate three types of neural network architectures for efficient learning of highly non-linear deformations of solid bodies. The first two architectures are based on the recently proposed CNN U-NET and MAgNET (graph U-NET) frameworks which have shown promising performance for learning on mesh-based data. The third architecture is Perceiver IO, a very recent architecture that belongs to the family of attention-based neural networks–a class that has revolutionised diverse engineering fields and is still unexplored in computational mechanics. We study and compare the performance of all three networks on two benchmark examples, and show their capabilities to accurately predict the non-linear mechanical responses of soft bodies.

show abstract

Section: Introductionmentioning

confidence: 99%

Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics

et al. 2023

View full text Add to dashboard Cite

show abstract

Research on flame prediction in a scramjet combustor using a data-driven model

et al. 2022

View full text Add to dashboard Cite

Flame prediction using deep learning technology could promote the research and development of flame propagation in scramjet combustors. A data-driven prediction model is proposed to effectively predict a future flame based on the flame sequence at the previous moments. A convolutional neural network is used to construct the prediction model, and the network training is performed using an experimental dataset. Ground experiments are conducted in a scramjet combustor using different equivalence ratio variation laws, and the flame evolution in the experiments is recorded and processed into a dataset. The flame prediction accuracy of the proposed model under different equivalence ratio variation laws is analyzed in detail. Moreover, both subjective and objective analysis results show that the flame prediction well agrees with the experimental result, and the flame boundary and area are accurately predicted to a certain extent. The influence of the prediction span on the flame prediction accuracy is also discussed.

show abstract

Deep learning model for two-fluid flows

et al. 2023

View full text Add to dashboard Cite

Various industries rely on numerical tools to simulate multiphase flows due to the wide occurrence of this phenomenon in nature, manufacturing processes, or the human body. However, the significant computation burden required for such simulations directs the research interest toward incorporating data-based approaches in the solution loop. Although, these approaches returned significant results in various domains, incorporating them in the computational fluid dynamics field is wrangled by their casting aside of the already known governing constitutional laws along with the natural incompatibility of various models with unstructured irregular discretization spaces. This work suggests a coupling framework, between a traditional finite element CFD solver and a deep learning model, for tackling multiphase fluid flows without migrating the benefits of physics-enriched traditional solvers. The tailored model architecture, along with the coupling framework, allows tackling the required problem with a dynamically adapted unstructured irregular triangular mesh, thus dodging the limitation of traditional convolution neural networks. Moreover, the various ingredients that allowed the model to simulate the complex and computation-demanding Navier-Stokes flow equation, such as relying on a sequential validation dataset while exposing the model training to a noise inherited from the quality of its inferring, along with the proper choice of model inputs, are highlighted and elaborated throughout this paper. To the authors' knowledge, this work is the first of its type to introduce a data-based graph-based approach for solving multiphase flow problems with a level-set interface capturing method.

show abstract

Deep learning model to assist multiphysics conjugate problems

Cited by 9 publications

References 47 publications

Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics

Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics

Research on flame prediction in a scramjet combustor using a data-driven model

Deep learning model for two-fluid flows

Contact Info

Product

Resources

About