Heat Equation Simulation Data

该数据集由经典热方程求解器生成的温度场组成，这些温度场是通过使用隐式欧拉方案的有限差分方法进行近似得到的。数据通过改变初始和边界温度条件生成，这些条件在[100, 500]开尔文的范围内随机采样。该数据集规模达到8TB，由多次求解器运行产生。其任务是用深度代理模型来预测温度场。

This dataset comprises temperature fields generated by a classical heat equation solver, where the fields are approximated using the finite difference method with an implicit Euler scheme. The data is generated by varying initial and boundary temperature conditions, which are randomly sampled within the range of [100, 500] Kelvin. This dataset has a total size of 8 TB and is produced from multiple solver runs. The downstream task of this dataset is to predict temperature fields using deep surrogate models.