PiNDiff-CVI

Probabilistic Physics-integrated Neural Differentiable Modeling for Isothermal Chemical Vapor Infiltration Process

The paper can be found here

Data generation for training PiNDiff:

Provide simulation conditions to generate training synthetic data in "case_setup.yaml" with gen_syn_data: True, train_model: True.

Train on synthetic data:

To train PiNDIff on synthetic data, Provide the following arguments in "case_setup.yaml" gen_syn_data: False, ExpData_name: False, load_syn_data: True, train_model: True.

Data generation for testing PiNDiff:

Provide simulation conditions to generate training synthetic data in "case_setup.yaml" with gen_syn_data: True, train_model: False.

Test on synthetic data:

To train PiNDIff on synthetic data, Provide the following arguments in "case_setup.yaml" gen_syn_data: False, ExpData_name: False, load_syn_data: True, train_model: False.

Train on Experimental data (Benzinger):

To train PiNDIff on experimental data, Provide the following arguments in "case_setup.yaml" gen_syn_data: False, ExpData_name: Benzinger2 or Benzinger3, load_syn_data: False, train_model: True.

Acknowledgments

We would like to acknowledge the funds from the Air ForceOffice of Scientific Research (AFOSR), United States of America, under award number FA9550-22-1-0065. J.X.W. would also like to acknowledge the funding support from the Office of Naval Research under award number N00014-23-1-2071 and the National Science Foundation under award number OAC-2047127 in supporting this study.

Citation

Find this useful or like this work? Cite us with:

@article{akhare2024probabilistic,
  title={Probabilistic physics-integrated neural differentiable modeling for isothermal chemical vapor infiltration process},
  author={Akhare, Deepak and Chen, Zeping and Gulotty, Richard and Luo, Tengfei and Wang, Jian-Xun},
  journal={npj Computational Materials},
  volume={10},
  number={1},
  pages={120},
  year={2024},
  publisher={Nature Publishing Group UK London}
}