Physics-Embedded Neural Networks: Graph Neural PDE Solvers with Mixed Boundary Conditions
This addresses the need for reliable and fast PDE solvers in physics and engineering, though it is incremental in improving boundary condition handling and long-term prediction in GNNs.
The paper tackles the problem of accurately solving partial differential equations (PDEs) with boundary conditions using graph neural networks (GNNs), achieving a better speed-accuracy trade-off than classical solvers and state-of-the-art machine learning models.
Graph neural network (GNN) is a promising approach to learning and predicting physical phenomena described in boundary value problems, such as partial differential equations (PDEs) with boundary conditions. However, existing models inadequately treat boundary conditions essential for the reliable prediction of such problems. In addition, because of the locally connected nature of GNNs, it is difficult to accurately predict the state after a long time, where interaction between vertices tends to be global. We present our approach termed physics-embedded neural networks that considers boundary conditions and predicts the state after a long time using an implicit method. It is built based on an E(n)-equivariant GNN, resulting in high generalization performance on various shapes. We demonstrate that our model learns flow phenomena in complex shapes and outperforms a well-optimized classical solver and a state-of-the-art machine learning model in speed-accuracy trade-off. Therefore, our model can be a useful standard for realizing reliable, fast, and accurate GNN-based PDE solvers. The code is available at https://github.com/yellowshippo/penn-neurips2022.