Airfoil Flow Simulation Dataset

该数据集是一组针对多种不同气动力几何形状的稳态雷诺平均纳维尔斯托克斯（RANS）模拟的全面集合。它被用于从表面压力测量中重建完整的压力和速度场。数据集采用了分层分割方法，针对训练、验证和测试集进行了划分，特别关注几何变化和具有挑战性的气动力情况。规模上，该数据集包含了2907个收敛的模拟，其中2469个用于训练，239个用于验证，199个用于测试。任务目标是：从稀疏的表面测量中重建气动力学流场。

This dataset is a comprehensive collection of steady Reynolds-Averaged Navier-Stokes (RANS) simulations targeting a wide range of distinct aerodynamic geometries. It is utilized to reconstruct full pressure and velocity fields from surface pressure measurements. The dataset adopts a stratified splitting method to partition data into training, validation, and test sets, with particular focus on geometric variations and challenging aerodynamic scenarios. In terms of scale, the dataset contains 2907 converged simulations, among which 2469 are allocated for training, 239 for validation, and 199 for testing. The task objective is to reconstruct aerodynamic flow fields from sparse surface measurements.