Plasma-prescribed Active Region Static Extrapolation (PARSE) Dataset

PARSE数据集是由美国宇航局戈达德太空飞行中心创建的，包含超过五千个数据立方体，基于Spaceweather HMI Active Region Patch (SHARP)库的磁图图像。每个立方体解析了磁场矢量和等离子体强制力，分布在约100,000个散点上，这些散点在磁通量高的区域附近自适应聚集。数据集的创建过程利用了先进的磁静态外推方法，通过Radial-Basis-Function Finite-Difference (RBF-FD) MHS求解器实现。PARSE数据集主要用于训练和验证物理信息神经网络，解决太阳活跃区域磁场结构的复杂问题。

The PARSE dataset was created by NASA Goddard Space Flight Center, comprising over 5,000 data cubes based on magnetogram images from the Spaceweather HMI Active Region Patch (SHARP) repository. Each cube resolves magnetic field vectors and plasma forcing terms, distributed across approximately 100,000 sampling points that are adaptively clustered near regions with high magnetic flux. The dataset creation process leverages advanced magnetostatic extrapolation methods, implemented via the Radial-Basis-Function Finite-Difference (RBF-FD) MHS solver. The PARSE dataset is primarily used for training and validating physics-informed neural networks to address complex problems related to the magnetic field structures of solar active regions.