CAREER: A Novel and Fast Open-Source Code for Global Simulation of Stratified Convection and Magnetohydrodynamics of the Sun

职业生涯：用于太阳分层对流和磁流体动力学全局模拟的新颖且快速的开源代码

• 批准号: 1554005
• 负责人: Chunlei Liang
• 金额: $ 49.72万
• 依托单位: George Washington University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-05-15 至 2019-10-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1554005&HistoricalAwards=false
• 关键词: CAREER; Novel; Fast; Open; Source

Non-technical: The goal of this project is to create a unique capability for predicting density-stratified magnetohydrodynamics of the Sun. This research is expected to lay a foundation for developing methods for predicting extreme space weather, e.g. the event of a "super solar flare" followed by an extreme geomagnetic storm. Scientific results of this research can help resolve several contradictory predictions from previous studies of the solar convection zone. The Principal Investigator (PI) will develop and disseminate a powerful open-source software package to the space weather and solar physics communities. The success of predicting severe space weather events has significant societal and economic impacts. PI will design high-order accurate computational algorithms suitable for exascale simulations that can perform a billion billion calculations per second. This software will run on massively parallel distributed-memory computers to predict coupled global and local dynamics of the sun. PI will reach out to K-12 students and demonstrate that science of the sun and high-performance computing are exciting and important to society. Furthermore, PI will leverage outreach efforts with the High Altitude Observatory of the National Center for Atmospheric Research and other research centers. This project, thus, serves the national interest as stated by NSF's mission: to promote the progress of science and to advance the national welfare.Technical: The goal of this research program is to develop a novel, fully compressible model and an open-source community code for global simulations of the solar convection zone that includes the top near surface shear layer of the Sun. Current leading global simulations use an elastic approximation whose computational domains extend from the base of the solar convection zone and must stop at about 0.96 solar radius, stopping short of the top near surface shear layer where Mach number could reach unity. This research program will create a powerful open-source community code CHORUS++ to simulate magnetohydrodynamics of the solar convection zone. CHORUS stands for Compressible High-ORder Unstructured-grid Spectral difference code which has been co-developed by the PI for hydrodynamics of the solar convection zone. CHORUS++ will be equipped with variable mesh resolution capability to focus on targeted regions of interests. A fast local time-stepping algorithm will be designed and equipped for CHORUS++ for long-period time integration on massively parallel computers. These technical accomplishments can accelerate the original CHORUS code by a factor over 100. The PI will conduct a series of global simulations of magnetohydrodynamics of the solar convection zone with unprecedented resolutions for predicting the differential rotation, meridional circulation, giant cells, and super-granulation of the sun.

非技术性：该项目的目标是建立预测太阳密度分层磁流体动力学的独特能力。预计这项研究将为制定预测极端空间天气的方法奠定基础，例如“超级太阳耀斑”事件和随后的极端地磁暴。这项研究的科学成果可以帮助解决以前对太阳对流区研究中的几个相互矛盾的预测。首席研究员将开发一个功能强大的开放源码软件包，并将其分发给空间气象和太阳物理学界。成功预测严重空间气象事件具有重大的社会和经济影响。PI将设计适用于百亿亿级模拟的高阶精确计算算法，每秒可执行10亿亿次计算。该软件将在大规模并行分布式存储器计算机上运行，以预测太阳的耦合全局和局部动力学。PI将接触K-12学生，并证明太阳科学和高性能计算对社会是令人兴奋和重要的。 此外，PI将利用与国家大气研究中心高海拔观测站和其他研究中心的外联工作。 因此，该项目符合国家利益，正如NSF的使命所述：促进科学进步，提高国家福利。技术：该研究计划的目标是开发一种新颖的，完全可压缩的模型和一个开放源代码的社区代码，用于全球模拟太阳对流区，包括太阳近地表剪切层的顶部。目前领先的全球模拟使用弹性近似，其计算域从太阳对流区的底部延伸，必须在大约0.96太阳半径处停止，在接近表面剪切层的顶部处停止，马赫数可以达到1。该研究计划将创建一个强大的开源社区代码CHORUS++来模拟太阳对流区的磁流体动力学。 CHORUS代表可压缩高阶非结构网格光谱差分代码，该代码由PI共同开发，用于太阳对流区的流体动力学。 CHORUS++将配备可变网格分辨率功能，以专注于目标感兴趣区域。设计了一种快速的局部时间步进算法，并将其装备到CHORUS++上，用于大规模并行计算机上的长周期时间积分。这些技术成就可以将原始CHORUS代码加速100倍以上。 PI将对太阳对流区的磁流体动力学进行一系列全球模拟，其分辨率前所未有，用于预测太阳的差异旋转，纬向环流，巨细胞和超级颗粒。