Reference : Convolution, aggregation and attention based deep neural networks for accelerating si...

	Document type :	E-prints/Working papers : First made available on ORBilu
	Discipline(s) :	Engineering, computing & technology : Multidisciplinary, general & others
	Focus Areas :	Computational Sciences
	To cite this reference:	http://hdl.handle.net/10993/52915

Title :	Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics
Language :	English
Author, co-author :	Deshpande, Saurabh [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Engineering (DoE) >]
	Sosa, Raul Ian [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Physics and Materials Science (DPHYMS) >]
	Bordas, Stéphane [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Engineering (DoE) >]
	Lengiewicz, Jakub [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Engineering (DoE) >]
Publication date :	Nov-2022
Peer reviewed :	No
Keywords :	[en] Surrogate Modeling ; Deep Learning ; CNN U-NET ; Graph U-Net ; Perceiver IO ; Finite Element Method
Abstract :	[en] 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 nonlinear 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.
Target :	Researchers ; Professionals
Permalink :	http://hdl.handle.net/10993/52915
Framework Programme / European Project :	H2020 ; 764644 - RAINBOW - Rapid Biomechanics Simulation for Personalized Clinical Design

	File(s) associated to this reference
	Fulltext file(s): File Commentary Version Size Access Open access Convolution__aggregation_attention_based_deep_neural_networks.pdf Author preprint 10.8 MB View/Open

All documents in ORBi^lu are protected by a user license.

 

University of Luxembourg Library |  Feedback |  Legal notices Site Map