Graph neural network-accelerated Lagrangian fluid simulation

• Published in: Computers & graphics Vol. 103; pp. 201 - 211
• Main Authors: Li, Zijie, Farimani, Amir Barati
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.04.2022; Elsevier Science Ltd
• Subjects: Acceleration; Data-driven modeling; Fluid flow; Forecasting; Graph neural networks; Graphical representations; Incompressible flow; Incompressible fluids; Iterative methods; Neural networks; Particle-based fluid dynamics; Physical properties; Simulation; Data-driven modeling; Particle-based fluid dynamics; Graph neural networks
• ISSN: 0097-8493; 1873-7684
• DOI: 10.1016/j.cag.2022.02.004

We present a data-driven model for fluid simulation under Lagrangian representation. Our model, Fluid Graph Networks (FGN), uses graphs to represent the fluid field. In FGN, fluid particles are represented as nodes and their interactions are represented as edges. Instead of directly predicting the acceleration or position correction given the current state, FGN decomposes the simulation scheme into separate parts — advection, collision, and pressure projection. For these different predictions tasks, we propose two kinds of graph neural network structures, node-focused networks and edge-focused networks. We show that the learned model can produce accurate results and remain stable in scenarios with different geometries. In addition, FGN is able to retain many important physical properties of incompressible fluids, such as low velocity divergence, and adapt to time step sizes beyond the one used in the training set. FGN is also computationally efficient compared to classical simulation methods as it operates on a smaller neighborhood and does not require iteration at each timestep during the inference. [Display omitted] •A data-driven model which learns to simulate fluids by learning advection, collison, and pressure projection respectively with different sub-networks.•An efficient and lightweight graph neural network architecture for learning particle-based fluid dynamics.•Our method can predict accurate fluid dynamics and generate realistic animation, with an order of magnitude speed-up compared to the classical method.