Macroscopic auxiliary asymptotic preserving neural networks for the linear radiative transfer equations
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914700948668416 |
|---|---|
| author | Li, Hongyan Jiang, Song Sun, Wenjun Xu, Liwei Zhou, Guanyu |
| author_facet | Li, Hongyan Jiang, Song Sun, Wenjun Xu, Liwei Zhou, Guanyu |
| contents | We develop a Macroscopic Auxiliary Asymptotic-Preserving Neural Network (MA-APNN) method to solve the time-dependent linear radiative transfer equations (LRTEs), which have a multi-scale nature and high dimensionality. To achieve this, we utilize the Physics-Informed Neural Networks (PINNs) framework and design a new adaptive exponentially weighted Asymptotic-Preserving (AP) loss function, which incorporates the macroscopic auxiliary equation that is derived from the original transfer equation directly and explicitly contains the information of the diffusion limit equation. Thus, as the scale parameter tends to zero, the loss function gradually transitions from the transport state to the diffusion limit state. In addition, the initial data, boundary conditions, and conservation laws serve as the regularization terms for the loss. We present several numerical examples to demonstrate the effectiveness of MA-APNNs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_01820 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Macroscopic auxiliary asymptotic preserving neural networks for the linear radiative transfer equations Li, Hongyan Jiang, Song Sun, Wenjun Xu, Liwei Zhou, Guanyu Numerical Analysis Machine Learning We develop a Macroscopic Auxiliary Asymptotic-Preserving Neural Network (MA-APNN) method to solve the time-dependent linear radiative transfer equations (LRTEs), which have a multi-scale nature and high dimensionality. To achieve this, we utilize the Physics-Informed Neural Networks (PINNs) framework and design a new adaptive exponentially weighted Asymptotic-Preserving (AP) loss function, which incorporates the macroscopic auxiliary equation that is derived from the original transfer equation directly and explicitly contains the information of the diffusion limit equation. Thus, as the scale parameter tends to zero, the loss function gradually transitions from the transport state to the diffusion limit state. In addition, the initial data, boundary conditions, and conservation laws serve as the regularization terms for the loss. We present several numerical examples to demonstrate the effectiveness of MA-APNNs. |
| title | Macroscopic auxiliary asymptotic preserving neural networks for the linear radiative transfer equations |
| topic | Numerical Analysis Machine Learning |
| url | https://arxiv.org/abs/2403.01820 |