SeismicNet

Physics-informed neural networks for seismic wave modeling in semi-infinite domain

Paper link: [CPC], [arXiv]

Authors: Pu Ren, Chengping Rao, Su Chen, Jian-Xun Wang, Hao Sun, Yang Liu

Highlights

• Present a new PINN model for seismic wave modeling in semi-infinite domain without the need for labeled data.
• Introduce the absorbing boundary conditions into the network as a soft regularizer for handling truncated boundaries.
• Leverage a sequential training strategy via temporal domain decomposition to improve the scalability of the PINNs.
• Design a novel surrogate modeling strategy to account for parametric loading, which estimates the wave propagation in semi-infinite domain given the seismic loading at different locations.

Datasets

The dataset is provided via a Google Drive link.

Citation

If you find our research helpful, please consider citing us with：

@article{ren2023seismicnet,
  title={SeismicNet: Physics-informed neural networks for seismic wave modeling in semi-infinite domain},
  author={Ren, Pu and Rao, Chengping and Chen, Su and Wang, Jian-Xun and Sun, Hao and Liu, Yang},
  journal={Computer Physics Communications},
  pages={109010},
  year={2023},
  publisher={Elsevier}
}