GPU-Enabled General Relativistic Simulations of Misaligned Black Hole Accretion Systems

支持 GPU 的未对准黑洞吸积系统的广义相对论模拟

• 批准号: 1811605
• 负责人: Alexander Chekhovskoy
• 金额: $ 1.5万
• 依托单位: Northwestern University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-05-01 至 2019-10-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1811605&HistoricalAwards=false
• 关键词: GPU; Enabled; General; Relativistic; Simulations

A black hole accretion disk is a structure formed by diffused materials in orbital motion around a black hole (BH). From observations, the presence of tilted accretion disks around a BH is detected in some systems. However, the physics of accretion disks is poorly understood. This project will run very large scale simulations on the Blue Waters supercomputer to improve our fundamental understanding of BH accretion disks. Results of the proposed simulations will address long-standing questions in the way supermassive black holes consume and expel gas, and thereby exert feedback on their environment. These results will allow the physics community to gain first-principles understanding of disk physics in typical tilted BH accretion systems.Gas falling into a BH from large distances is unaware of BH spin direction, and misalignment between the accretion disk and BH spin is expected to be common. However, the physics of tilted disks is poorly understood, even for the "standard", geometrically thin, radiatively efficient accretion disks that power active galactic nuclei known as quasars and thought to provide the best observational tests of general relativity and disk physics. In particular, it is still not understood how the curved space-time of a spinning black hole imprints itself on the structure of the tilted disks. This project will make use of the fact that, at their core, BH accretion disks are well-described by the general relativistic magnetohydrodynamics (GRMHD) equations of motion. By carrying out direct GRMHD simulations of tilted thin and thick disks, the project will obtain the first-principles understanding of disk physics in typical, tilted BH accretion systems. To surmount the prohibitively expensive nature of these simulations, the project has constructed the first GPU-accelerated GRMHD code, H-AMR, which is capable of adaptive mesh refinement and is ideally suited for the Blue Waters supercomputer.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

黑洞吸积盘是由围绕黑洞(BH)轨道运动的扩散物质形成的结构。根据观测，在某些系统中，观测到黑洞周围存在倾斜的吸积盘。然而，人们对吸积盘的物理学知之甚少。这个项目将在Blue Waters超级计算机上进行非常大规模的模拟，以提高我们对BH吸积盘的基本理解。拟议的模拟结果将解决超大质量黑洞消耗和排出气体的长期问题，从而对其环境产生反馈。这些结果将使物理界对典型的倾斜BH吸积系统中的盘物理有第一性原理的了解。从大距离落入BH的气体不知道BH自旋方向，吸积盘和BH自旋之间的不对准预计是常见的。然而，人们对倾斜盘的物理学知之甚少，即使是为被称为类星体的活动星系核提供动力的几何上薄的、辐射高效的吸积盘，也是如此，被认为提供了广义相对论和盘物理的最佳观测测试。特别是，人们仍然不知道旋转黑洞的弯曲时空是如何在倾斜的圆盘结构上留下自己的印记的。这个项目将利用这样一个事实，即BH吸积盘的核心可以用广义相对论磁流体动力学(GRMHD)运动方程很好地描述。通过对倾斜的薄盘和厚盘进行直接GRMHD模拟，该项目将获得对典型的倾斜的BH吸积系统中的盘物理的第一性原理的理解。为了克服这些模拟的昂贵特性，该项目构建了第一个由GPU加速的GRMHD代码H-AMR，该代码能够进行自适应网格优化，非常适合Blue Waters超级计算机。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。