2D图网格数据集

2D图网格数据集是由索邦大学和Extrality合作创建的，旨在研究高雷诺数下的不可压缩稳态纳维-斯托克斯方程。该数据集包含230个样本，涵盖了从10^6到5*10^6的雷诺数范围，用于训练和验证几何深度学习模型。数据集通过OpenFOAM软件生成，提供了详细的流体动力学模拟数据，包括速度、压力和湍流粘度等物理量。此数据集的应用领域主要集中在流体动力学模拟和物理系统学习，旨在通过深度学习方法提高流体模拟的效率和准确性。

The 2D graph-grid dataset was developed in collaboration between Sorbonne University and Extrality, with the aim of studying incompressible steady-state Navier-Stokes equations at high Reynolds numbers. This dataset contains 230 samples, covering a Reynolds number range from 10^6 to 5×10^6, and is intended for training and validating geometric deep learning models. The dataset is generated using the OpenFOAM software, and provides detailed hydrodynamic simulation data including physical quantities such as velocity, pressure, and turbulent viscosity. The application scenarios of this dataset mainly focus on hydrodynamic simulation and physical system learning, with the goal of improving the efficiency and accuracy of fluid simulation via deep learning methods.