An energy approach to the solution of partial differential equations in computational mechanics via machine learning: Concepts, implementation and applications

E Samaniego, C Anitescu, S Goswami… - Computer Methods in …, 2020‏ - Elsevier
Abstract Partial Differential Equations (PDEs) are fundamental to model different
phenomena in science and engineering mathematically. Solving them is a crucial step …

Deep learning for plasticity and thermo-viscoplasticity

DW Abueidda, S Koric, NA Sobh, H Sehitoglu - International Journal of …, 2021‏ - Elsevier
Predicting history-dependent materials' responses is crucial, as path-dependent behavior
appears while characterizing or geometrically designing many materials (eg, metallic and …

Meshless physics‐informed deep learning method for three‐dimensional solid mechanics

DW Abueidda, Q Lu, S Koric - International Journal for …, 2021‏ - Wiley Online Library
Deep learning (DL) and the collocation method are merged and used to solve partial
differential equations (PDEs) describing structures' deformation. We have considered …

Enhanced physics‐informed neural networks for hyperelasticity

DW Abueidda, S Koric, E Guleryuz… - International Journal for …, 2023‏ - Wiley Online Library
Physics‐informed neural networks have gained growing interest. Specifically, they are used
to solve partial differential equations governing several physical phenomena. However …

The magneto-electro-elastic coupling isogeometric analysis method for the static and dynamic analysis of magneto-electro-elastic structures under thermal loading

L Zhou, F Qu - Composite Structures, 2023‏ - Elsevier
Many engineering problems show some degree of coupling or interaction between different
physics fields. It is necessary to improve the calculation efficiency and accuracy for magneto …

An analytical solution for vibration response of CNT/GPL/fibre/polymer hybrid composite micro/nanoplates

H Salehipour, MA Shahmohammadi… - Mechanics of …, 2024‏ - Taylor & Francis
In the present article, a closed-form solution is carried out for the vibration response of
CNT/GPL/fiber/polymer hybrid composite macro/micro/nanoplates resting on elastic support …

Deep learning modeling strategy for material science: from natural materials to metamaterials

W Li, P Chen, B **ong, G Liu, S Dou… - Journal of Physics …, 2022‏ - iopscience.iop.org
Computational modeling is a crucial approach in material-related research for discovering
new materials with superior properties. However, the high design flexibility in materials …

Explicit error estimates for spline approximation of arbitrary smoothness in isogeometric analysis

E Sande, C Manni, H Speleers - Numerische Mathematik, 2020‏ - Springer
In this paper we provide a priori error estimates with explicit constants for both the L^ 2 L 2-
projection and the Ritz projection onto spline spaces of arbitrary smoothness defined on …

Improving the accuracy of the deep energy method

C Chadha, J He, D Abueidda, S Koric, E Guleryuz… - Acta Mechanica, 2023‏ - Springer
The deep energy method (DEM), a type of physics-informed neural network, is evolving as
an alternative to finite element analysis. It employs the principle of minimum potential energy …

Collaborative design of fiber path and shape for complex composite shells based on isogeometric analysis

P Hao, X Liu, Y Wang, D Liu, B Wang, G Li - Computer Methods in Applied …, 2019‏ - Elsevier
Composite shells with complex geometry are widely used in aerospace structures. Due to
the complexity of geometry and curvilinear fiber path, the analysis and optimization based …