Finite Volume Graph Network(FVGN): Predicting unsteady incompressible fluid dynamics with finite volume informed neural network

Abstract

The rapid development of deep learning has significant implications for the advancement of Computational Fluid Dynamics (CFD). Currently, most pixel-grid-based deep learning methods for flow field prediction exhibit significantly reduced accuracy in predicting boundary layer flows and poor adaptability to geometric shapes. Although Graph Neural Network (GNN) models for unstructured grids based unsteady flow prediction have better geometric adaptability, these models suffer from error accumulation in long-term predictions of unsteady flows. More importantly, fully data-driven models often require extensive training time, greatly limiting the rapid update and iteration speed of deep learning models when facing more complex unsteady flows. Therefore, this paper aims to balance the demands for training overhead and prediction accuracy by integrating physical constraints based on the finite volume method into the loss function of the graph neural network. Additionally, it incorporates a twice-massage aggregation mechanism inspired by the extended stencil method to enhance the unsteady flow prediction accuracy and geometric shape generalization ability of the graph neural network model on unstructured grids. We focus particularly on the model's predictive accuracy within the boundary layer. Compared to fully data-driven methods, our model achieves better predictive accuracy and geometric shape generalization ability in a shorter training time.

References in corpus (19)

• Neural Message Passing for Quantum Chemistry
• NSFnets (Navier-Stokes Flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations
• PhyGeoNet: Physics-Informed Geometry-Adaptive Convolutional Neural Networks for Solving Parameterized Steady-State PDEs on Irregular Domain
• Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows
• Learning to Simulate Complex Physics with Graph Networks
• Physics-informed deep learning for incompressible laminar flows
• Physics-informed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse problems
• Graph networks as learnable physics engines for inference and control
• Uncovering near-wall blood flow from sparse data with physics-informed neural networks
• Physics-Informed Neural Operator for Learning Partial Differential Equations
• Graph neural networks for laminar flow prediction around random 2D shapes
• Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View
• Teaching the Incompressible Navier-Stokes Equations to Fast Neural Surrogate Models in 3D
• Flow Completion Network: Inferring the Fluid Dynamics from Incomplete Flow Information using Graph Neural Networks
• A Finite Element-Inspired Hypergraph Neural Network: Application to Fluid Dynamics Simulations
• Physics-Embedded Neural Networks: Graph Neural PDE Solvers with Mixed Boundary Conditions
• Finite Volume Neural Network: Modeling Subsurface Contaminant Transport
• Predicting Physics in Mesh-reduced Space with Temporal Attention
• Learning Incompressible Fluid Dynamics from Scratch -- Towards Fast, Differentiable Fluid Models that Generalize

Cited by in corpus (3)

• Finite-difference-informed graph network for solving steady-state incompressible flows on block-structured grids
• Unstructured Moving Least Squares Material Point Methods: A Stable Kernel Approach With Continuous Gradient Reconstruction on General Unstructured Tessellations
• A graph neural network surrogate model for mesh-based crashworthiness prediction of vehicle panel components