DeepPhysics: A physics aware deep learning framework for real‐time simulation

DOI10.1002/NME.6943MaRDI QIDQ6129672zbMATH OpenFDO

Authors Ryadh Haferssas, Stéphane Cotin

Publication date 17 April 2024

Published in International Journal for Numerical Methods in Engineering (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2109.09491

zbMATH Keywords

deep learningfinite element methodneural networkNewton-Raphsonphysics informed neural networkreal-time

Mathematics Subject Classification ID

Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Multigrid methods; domain decomposition for boundary value problems involving PDEs (65N55) Artificial neural networks and deep learning (68T07) Nonlinear elasticity (74B20)

Abstract: Real-time simulation of elastic structures is essential in many applications, from computer-guided surgical interventions to interactive design in mechanical engineering. The Finite Element Method is often used as the numerical method of reference for solving the partial differential equations associated with these problems. Yet, deep learning methods have recently shown that they could represent an alternative strategy to solve physics-based problems 1,2,3. In this paper, we propose a solution to simulate hyper-elastic materials using a data-driven approach, where a neural network is trained to learn the non-linear relationship between boundary conditions and the resulting displacement field. We also introduce a method to guarantee the validity of the solution. In total, we present three contributions: an optimized data set generation algorithm based on modal analysis, a physics-informed loss function, and a Hybrid Newton-Raphson algorithm. The method is applied to two benchmarks: a cantilever beam and a propeller. The results show that our network architecture trained with a limited amount of data can predict the displacement field in less than a millisecond. The predictions on various geometries, topologies, mesh resolutions, and boundary conditions are accurate to a few micrometers for non-linear deformations of several centimeters of amplitude.

Recommendations

• A deep energy method for finite deformation hyperelasticity
• Deep learning-accelerated computational framework based on physics informed neural network for the solution of linear elasticity
• A deep learning energy method for hyperelasticity and viscoelasticity
• Meshless physics-informed deep learning method for three-dimensional solid mechanics
• Deep learning acceleration of total Lagrangian explicit dynamics for soft tissue mechanics

Cited in

(8)

• Thermodynamics-informed neural networks for physically realistic mixed reality
• Towards real-time fluid dynamics simulation: a data-driven NN-MPS method and its implementation
• Predicting dynamic responses of continuous deformable bodies: a graph-based learning approach
• Real-time topology optimization via learnable mappings
• Speeding up the reservoir simulation by real time prediction of the initial guess for the Newton-Raphson's iterations
• -FEM-FNO: a new approach to train a neural operator as a fast PDE solver for variable geometries
• A study of mechanism-data hybrid-driven method for multibody system via physics-informed neural network
• Accelerating the convergence of Newton's method for nonlinear elliptic PDEs using Fourier neural operators

This page was built for publication: DeepPhysics: A physics aware deep learning framework for real‐time simulation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6129672)