CaltechTHESIS
  A Caltech Library Service

Black Hole Simulations: From Supercomputers to Your Laptop

Citation

Varma, Vadapalli Vijay S. (2019) Black Hole Simulations: From Supercomputers to Your Laptop. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/YT07-Q639. https://resolver.caltech.edu/CaltechTHESIS:05252019-040137906

Abstract

In this thesis, I will present various advancements in the modeling of binary black holes (BBHs): two black holes (BHs) that are in orbit around each other. The BHs lose energy to gravitational waves, causing them to spiral towards each other until they eventually merge and leave behind a single BH. BBHs are primary sources for ground based detectors such as the Laser Interferometer Gravitational-Wave Observatory (LIGO).

As the BHs are about to merge, they are moving at about half the speed of light and the spacetime is highly dynamical. All analytical methods break down at this stage, and numerical relativity (NR) simulations of the full Einstein’s equations are necessary. These simulations, however, are very expensive, with each simulation taking a month on a supercomputer. For direct data analysis applications with LIGO, we need a model that can be evaluated in a fraction of a second. Therefore, several approximate but fast models that are calibrated to NR simulations have been developed over the years.

Surrogate modeling provides a more powerful alternative: trained directly against the NR simulations without added assumptions, these models can reproduce the simulations as accurately as the simulations themselves, while taking only a fraction of a second to evaluate on a laptop. In short, surrogate models take BBH NR simulations from supercomputers to your laptop, without a loss of accuracy.

In this thesis, I will present several state-of-the-art surrogate models including (i) the first NR based surrogate model to span the full range of frequencies for ground based detectors, (ii) the first surrogate model for the mass, spin, and kick velocity of the final black hole after merger, and (iii) extension of an existing precessing surrogate model to higher mass ratios. In addition, I will present some work in improving the BBH initial data used in NR simulations, as well as in understanding the systematic biases introduced by approximate waveform models in LIGO data analysis.

As we head into the imminent era of high-precision gravitational wave astronomy, accurate yet fast models such as surrogate models will play a crucial role in maximizing the science output of our detectors.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:gravitational waves; numerical relativity; surrogate models; LIGO.
Degree Grantor:California Institute of Technology
Division:Physics, Mathematics and Astronomy
Major Option:Physics
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Scheel, Mark (advisor)
  • Chen, Yanbei (advisor)
Group:TAPIR, Astronomy Department
Thesis Committee:
  • Teukolsky, Saul A. (chair)
  • Weinstein, Alan Jay
  • Wise, Mark B.
  • Scheel, Mark
  • Chen, Yanbei
Defense Date:7 May 2019
Non-Caltech Author Email:vijay.varma392 (AT) gmail.com
Funders:
Funding AgencyGrant Number
NSFPHY–170212
NSFPHY–1708213
Sherman Fairchild FoundationUNSPECIFIED
Record Number:CaltechTHESIS:05252019-040137906
Persistent URL:https://resolver.caltech.edu/CaltechTHESIS:05252019-040137906
DOI:10.7907/YT07-Q639
Related URLs:
URLURL TypeDescription
https://journals.aps.org/prd/abstract/10.1103/PhysRevD.98.084032PublisherChapter 2
https://journals.aps.org/prd/abstract/10.1103/PhysRevD.98.104011PublisherChapter 3
https://journals.aps.org/prd/abstract/10.1103/PhysRevD.96.124024PublisherChapter 4
https://journals.aps.org/prd/abstract/10.1103/PhysRevD.99.064045PublisherChapter 5
https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.122.011101PublisherChapter 6
https://arxiv.org/abs/1905.09300arXivChapter 7
https://iopscience.iop.org/article/10.1088/1361-6382/ab0ee9/metaPublisherChapter 8
ORCID:
AuthorORCID
Varma, Vadapalli Vijay S.0000-0002-9994-1761
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:11544
Collection:CaltechTHESIS
Deposited By: Vadapalli Vijay Varma
Deposited On:31 May 2019 21:56
Last Modified:23 Sep 2020 20:54

Thesis Files

[img]
Preview
PDF (Complete thesis) - Final Version
See Usage Policy.

14MB

Repository Staff Only: item control page