Adaptive physics-informed neural networks for dynamic coupled thermo-mechanical problems in large-size-ratio functionally graded materials

Applied Mathematical Modelling, Год журнала: 2024, Номер unknown, С. 115906 - 115906

Опубликована: Дек. 1, 2024

Язык: Английский

---

A Hybrid Collocation Method for Long‐Time Simulation of Heat Conduction in Anisotropic Functionally Graded Materials

International Journal for Numerical Methods in Engineering, Год журнала: 2025, Номер 126(3)

Опубликована: Фев. 7, 2025

ABSTRACT This study proposes a hybrid collocation approach for simulating heat conduction problems in anisotropic functionally graded materials over extended time intervals. In this approach, the Krylov deferred correction (KDC) scheme is employed temporal discretization of dynamic problems, featuring novel numerical implementation designed to ensure precise satisfaction boundary conditions. The localized radial basis function (LRBF) method modified and utilized solve resulting value problems. A new developed combined with an optimization strategy distribution source points enhance performance LRBF scheme. synergizes KDC technique, which supports large step sizes, method, characterized by its truly meshless nature, address long durations. Additionally, coefficient matrix produced sparse depends solely on spatial distances between points, advantageous long‐term simulations. Numerical simulations spanning thousands steps demonstrate accuracy, stability, convergence approach. framework shows significant improvements existing methods, particularly handling substantial temperature variations.

Язык: Английский

---

Physics-informed radial basis function network based on Hausdorff fractal distance for solving Hausdorff derivative elliptic problems

Computers & Mathematics with Applications, Год журнала: 2025, Номер 183, С. 271 - 286

Опубликована: Фев. 26, 2025

Язык: Английский

---

Localized Fourier collocation method for 2D transient heat conduction problems

Applied Mathematics Letters, Год журнала: 2024, Номер unknown, С. 109331 - 109331

Опубликована: Окт. 1, 2024

Язык: Английский

---

Least-squares stabilized collocation method for the parameter identification in transient inverse heat conduction problems

Applied Mathematical Modelling, Год журнала: 2025, Номер unknown, С. 116093 - 116093

Опубликована: Март 1, 2025

Язык: Английский

---

Efficient BEM for thin-walled inhomogeneous potential problems: Theory and MATLAB code

Engineering Analysis with Boundary Elements, Год журнала: 2025, Номер 176, С. 106241 - 106241

Опубликована: Апрель 2, 2025

Язык: Английский

---

Spectral integrated neural networks with large time steps for 2D and 3D transient elastodynamic analysis

Neural Networks, Год журнала: 2025, Номер 188, С. 107559 - 107559

Опубликована: Апрель 27, 2025

Язык: Английский

---

A spatio-temporal radial basis function collocation method based on Hausdorff fractal distance for Hausdorff derivative heat conduction equations in three-dimensional anisotropic materials

Applied Mathematics and Computation, Год журнала: 2025, Номер 502, С. 129501 - 129501

Опубликована: Май 2, 2025

Язык: Английский

---

Enriched Physics-informed Neural Networks for Dynamic Poisson-Nernst-Planck Systems

Mathematics and Computers in Simulation, Год журнала: 2025, Номер unknown

Опубликована: Май 1, 2025

Язык: Английский

---

Numerical Simulation of Transient Heat Conduction With Moving Heat Source Using Physics Informed Neural Networks

International journal of mechanical system dynamics, Год журнала: 2025, Номер unknown

Опубликована: Июнь 5, 2025

ABSTRACT In this article, the physics informed neural networks (PINNs) is employed for numerical simulation of heat transfer involving a moving source under mixed boundary conditions. To reduce computational effort and increase accuracy, new training method proposed that uses continuous time‐stepping through learning. A single network initialized used as sliding window function across time domain. On each interval trained with initial condition iteration solution obtained at iteration. Thus, framework enables computation large temporal intervals without increasing complexity itself. The to estimate temperature distribution in homogeneous medium source. results from compared traditional finite element good agreement seen.

Язык: Английский

---