Rapid Prediction of Thermal Stress on Satellites via Domain Decomposition-based Hybrid Fourier Neural Operator
This repository contains code for the paper Rapid Prediction of Thermal Stress on Satellites via Domain Decomposition-based Hybrid Fourier Neural Operator.
Using conda and the environment.yml file:
conda env create --name HFNO --file=environment.yml
conda activate HFNO
We provide the datasets we used in the paper.
To divide the domain, use the following command:
python experiment/domain_decomposition/decomposition.py --PATH 'data/data_case1.npy'
--variable 'sigma_xy_2d'
--num_sims 3000
--num_dataset 5
--NUM_TURNS 6
--filename 'result/saved_KDtree/KDtree.npy'
python train.py --PATH 'data/data_case3.npy'
--PATH_kdtree 'result/saved_KDtree/KDTree_3.npy'
--saved_model 'result/saved_model/model.pt'
--variable 'sigma_xy_2d'
--epochs 201
--wandb False
python evaluate.py --PATH 'data/data_case3.npy'
--PATH_kdtree 'result/saved_KDtree/KDTree_3.npy'
--saved_model 'result/saved_model/model3.pt'
--variable 'sigma_xy_2d'
This work is built on top of other open source projects, including Fourier Neural Operator with Learned Deformations for PDEs on General Geometries (GEO-FNO), and NUNO: A General Framework for Learning Parametric PDEs with Non-Uniform Data (NU-FNO). We thank the original contributors of these works for open-sourcing their valuable source codes.
For any questions or issues, you are welcome to open an issue in this repo, or contact us at zhoukangruinudt@163.com, and wendy0782@126.com.