Pore-Scale CO2-Water Interaction Benchmark Dataset
收藏资源简介:
该数据集由赫瑞瓦特大学地球能源工程学院创建,是一个包含624个二维样本的数据集,每个样本大小为512×512像素,分辨率高达35微米,覆盖了100个时间步长的CO2和水相互作用的模拟结果。数据集体现了不同粒度大小和随机间隔变化的异质性,为机器学习模型提供了一个强大的测试平台。数据集以HDF5格式提供,包含了流速度分量、压力场、饱和度场和物理域表示的高分辨率数据,适用于科学和工程应用。
This dataset was created by the School of Geoscience and Energy Engineering at Heriot-Watt University. It includes 624 2D samples, each with a size of 512×512 pixels and a maximum resolution of 35 micrometers, covering simulation results of CO₂-water interactions across 100 time steps. The dataset exhibits heterogeneity with varying grain sizes and randomly spaced variations, serving as a robust testbed for machine learning models. Provided in HDF5 format, it contains high-resolution data of flow velocity components, pressure fields, saturation fields and physical domain representations, applicable for scientific and engineering applications.




