| 研究課題/領域番号 |
19F19317
|
|---|
| 研究機関 | 東京大学 |
|---|
研究代表者 |
藤井 通子 東京大学, 大学院理学系研究科(理学部), 准教授 (90722330)
|
|---|
| 研究分担者 |
Wang Long 東京大学, 理学(系)研究科(研究院), 外国人特別研究員
|
|---|
| 研究期間 (年度) |
2019-11-08 – 2022-03-31
|
| キーワード | star cluster / N-body simulation / binary black hole / gravitational wave |
|---|
| 研究実績の概要 |
The star-by-star N-body simulations of massive globular clusters (GCs) are important method to understand the long-term evolution of and predict the properties of gravitational wave (GW) events. The previous fastest N-body code, NBODY6++GPU, cannot handle dense massive GCs with many binaries. To overcome the bottleneck, we have developed the slow-down algorithmic regularization (SDAR) algorithm to efficiently handle the motions of multiple systems. Then, we developed a new N-body code, PeTar, which combines the particle-particle particle-tree method and the SDAR algorithm. Compared to NBODY6++GPU, PeTar can achieve 7-10 times faster computing speed for million-body simulations with many binaries. We have also implemented the code into ASURA-BRIDGE for simulating the formation of star clusters.
We have performed simulations of star clusters with different initial mass functions (IMFs) using our new code and found that the numbers of binary black holes merging with GW emission does not simply depends on the slope of the IMF. We also performed simulations of star clusters on Fugaku supercomputer. We are analyzing the data and will publish the results in the next fiscal year.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
The PeTar code is developed on time and shows excellent scaling on the XC50 supercomputer. Two papers related to it have been published. Now we have used it to carry out several studies related to GW events as the plan. One work on studying how the variation of IMF affects the GW events in GCs have finished and the paper is under review. The works on studying the impact of primordial binaries and the Galactic potential is ongoing. After implementing PeTar into the hydrodynamic code, ASURA-BRIDGE, we also studied how the dynamical formation and evolution of binaries affect the feedback to star formation.
|
| 今後の研究の推進方策 |
We will continue to carry out N-body models to investigate how the IMF, the primordial binaries and the Galactic potential affect the GW events in GCs. We will carry out the largest N-body model for GCs with 3 million stars by using the XC-50 and the Fugaku supercomputers. The models will be used to validate the approximate methods such as Monte-Carlo simulations and to predict the properties of GW events. In addition, we will carry out the simulation for nuclear star clusters with 10 million stars, to investigate whether the runaway collisions of stellar mass black holes can be the possible channel to form the seed of supermassive black hole. Moreover, we will continue to improve the performance and functions of the PeTar code.
|
