☆ 4.7 Article

JAX-FEM: A differentiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science

COMPUTER PHYSICS COMMUNICATIONS (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Editorial Material Mathematics, Interdisciplinary Applications

Special issue of computational mechanics on machine learning theories, modeling, and applications to computational materials science, additive manufacturing, mechanics of materials, design and optimization

Wing Kam Liu et al.

COMPUTATIONAL MECHANICS (2023)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

JAX-Fluids: A fully-differentiable high-order computational fluid dynamics solver for compressible two-phase flows

Deniz A. Bezgin et al.

Summary: Physical systems are described by partial differential equations (PDEs), and machine learning (ML) methods provide new approaches for solving PDEs. JAX-Fluids is a fully-differentiable CFD package intended for ML-supported CFD research, enabling simulation and optimization of complex fluid dynamics.

COMPUTER PHYSICS COMMUNICATIONS (2023)

添加到收藏夹

Article Engineering, Multidisciplinary

Mapped shape optimization method for the rational design of cellular mechanical metamaterials under large deformation

Tianju Xue et al.

Summary: Soft cellular mechanical metamaterials (CMMs) have unique mechanical properties under large deformation, but the challenge lies in the strong nonlinearities and complex instabilities brought by the deformation. A mapped shape optimization method is proposed as a computational framework for inverse designs of soft CMMs, utilizing a fixed referential configuration and a differentiable shape map to reflect geometric changes and compute the deformation of the structures. The method avoids altering the background mesh and efficiently calculates the gradient of objective functions using the adjoint method. The effectiveness of the method is demonstrated through numerical examples.

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING (2022)

添加到收藏夹

Article Chemistry, Physical

Two-dimensional Stiefel-Whitney insulators in liganded Xenes

Mingxiang Pan et al.

Summary: This study proposes a large class of liganded Xenes, namely hydrogenated and halogenated 2D group-IV honeycomb lattices, as 2D Stiefel-Whitney insulators (SWIs). The nontrivial topology of liganded Xenes is determined by the bulk topological invariant and the existence of protected corner states. Moreover, the liganded Xenes have a large and tunable bandgap, facilitating the experimental characterization of the 2D SWI phase.

NPJ COMPUTATIONAL MATERIALS (2022)

添加到收藏夹

Article Engineering, Multidisciplinary

Mechanistically informed data-driven modeling of cyclic plasticity via artificial neural networks

Daoping Liu et al.

Summary: A mechanistically informed data-driven approach is proposed to simulate the complex plastic behavior of microstructured/homogenized solids subjected to cyclic loading, showing effectiveness and reliability. The method, integrating data science methods with mechanics principles, requires less data compared to pure data-driven methods, potentially leading to greater computational efficiency.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2022)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Physics constrained learning for data-driven inverse modeling from sparse observations

Kailai Xu et al.

Summary: This article presents a new approach that trains deep neural networks (DNNs) while numerically satisfying partial differential equation (PDE) constraints. The algorithm developed allows differentiation of both explicit and implicit numerical solvers in reverse-mode automatic differentiation. The approach demonstrates faster convergence and better stability in relatively stiff problems compared to the penalty method.

JOURNAL OF COMPUTATIONAL PHYSICS (2022)

添加到收藏夹

Review Computer Science, Interdisciplinary Applications

Automatic Differentiation for Solid Mechanics

Andrea Vigliotti et al.

Summary: Automatic differentiation (AD) is utilized in this paper to simplify the statement and solution of solid mechanics problems by computing numerical derivatives accurately. Techniques involve obtaining residual force and tangent stiffness matrix using AD, with a focus on convenience for the solid mechanics community.

ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING (2021)

添加到收藏夹

Article Mathematics, Applied

A kernel method for learning constitutive relation in data-driven computational elasticity

Yoshihiro Kanno

Summary: This paper introduces a kernel method for extracting data points lying on a manifold in a material dataset, which is achieved by solving a regularized least-squares problem. The method requires only simple computational tasks and is easy to implement. Experiments demonstrate the effectiveness and robustness of the proposed method.

JAPAN JOURNAL OF INDUSTRIAL AND APPLIED MATHEMATICS (2021)

添加到收藏夹

Article Nuclear Science & Technology

Automatic Differentiation in MetaPhysicL and Its Applications in MOOSE

Alexander Lindsay et al.

Summary: Efficient solution of nonlinear systems of equations using Newton's method requires an accurate representation of the Jacobian, and automatic differentiation (AD) can provide this with guaranteed accuracy at a relatively low cost. This approach is beneficial for complex systems arising from spatial discretization of partial differential equations in multiphysics settings.

NUCLEAR TECHNOLOGY (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Machine learning-accelerated computational fluid dynamics

Dmitrii Kochkov et al.

Summary: Numerical simulation of fluids is crucial in modeling physical phenomena like weather and aerodynamics, but solving the Navier-Stokes equations at scale remains challenging due to computational costs. This study uses end-to-end deep learning to improve approximations in computational fluid dynamics for modeling two-dimensional turbulent flows, achieving computational speedups and stability while generalizing well to different scenarios. Leveraging machine learning and hardware accelerators, this approach showcases the potential to enhance simulations without sacrificing accuracy or generalization.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

AuTO: a framework for Automatic differentiation in Topology Optimization

Aaditya Chandrasekhar et al.

Summary: This paper introduces the use of automatic differentiation (AD) for computing sensitivities in topology optimization (TO) problems. The framework called AuTO, implemented using the high-performance Python library JAX, is illustrated through several examples in compliance minimization, compliant mechanism design, and microstructural design.

STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION (2021)

添加到收藏夹

Article Biochemical Research Methods

SciPy 1.0: fundamental algorithms for scientific computing in Python

Pauli Virtanen et al.

NATURE METHODS (2020)

添加到收藏夹

Review Multidisciplinary Sciences

Array programming with NumPy

Charles R. Harris et al.

NATURE (2020)

添加到收藏夹

Article Engineering, Multidisciplinary

Adjoint optimization of pressurized membrane structures using automatic differentiation tools

Alexander Niewiarowski et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2020)

添加到收藏夹

Article Chemistry, Physical

A data-driven computational scheme for the nonlinear mechanical properties of cellular mechanical metamaterials under large deformation

Tianju Xue et al.

SOFT MATTER (2020)

添加到收藏夹

Article Mathematics, Applied

Non-commutative discretize-then-optimize algorithms for elliptic PDE-constrained optimal control problems

Jun Liu et al.

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS (2019)

添加到收藏夹

Article Engineering, Multidisciplinary

TIGAR: Automating isogeometric analysis with FENICS

David Kamensky et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2019)

添加到收藏夹

Article Multidisciplinary Sciences

Deep learning predicts path-dependent plasticity

M. Mozaffar et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2019)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Stress-constrained topology optimization based on maximum stress measures

Dixiong Yang et al.

COMPUTERS & STRUCTURES (2018)

添加到收藏夹

Review Multidisciplinary Sciences

Deep learning

Yann LeCun et al.

NATURE (2015)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

The NumPy Array: A Structure for Efficient Numerical Computation

Stefan van der Walt et al.

COMPUTING IN SCIENCE & ENGINEERING (2011)

添加到收藏夹

Article Mathematics, Applied

OPTIMAL SOLVERS FOR PDE-CONSTRAINED OPTIMIZATION

Tyrone Rees et al.

SIAM JOURNAL ON SCIENTIFIC COMPUTING (2010)

添加到收藏夹

Article Mathematics

Inverse problems: A Bayesian perspective

A. M. Stuart

ACTA NUMERICA (2010)

添加到收藏夹

Article Computer Science, Software Engineering

deal. II - A general-purpose object-oriented finite element library

W. Bangerth et al.

ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE (2007)

添加到收藏夹

Review Engineering, Multidisciplinary

Review of options for structural design sensitivity analysis. Part 1: Linear systems

F van Keulen et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2005)

添加到收藏夹

Article Mathematics, Applied

ADJOINT SENSITIVITY ANALYSIS FOR DIFFERENTIAL-ALGEBRAIC EQUATIONS: THE ADJOINT DAE SYSTEM AND ITS NUMERICAL SOLUTION

Yang Cao et al.

SIAM JOURNAL ON SCIENTIFIC COMPUTING (2003)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.