English

Characterizing the gravitational wave temporal evolution of the gmode fundamental resonant frequency for a core collapse supernova: A neural network approach

General Relativity and Quantum Cosmology 2023-04-25 v1

Abstract

We present a methodology based on the implementation of a fully connected neural network to estimate the gravitational wave (GW) temporal evolution of the gmode fundamental resonant frequency for a Core Collapse Supernova (CCSN). To perform the estimation, we construct a training data set, using synthetic waveforms, that serves to train the ML algorithm, and then use several CCSN waveforms to test the model. According to the results obtained from the implementation of our model, we provide numerical evidence to support the classification of progenitors according to their degree of rotation. The relative error associated with the estimate of the slope of the resonant frequency versus time for the GW from CCSN signals is within 13%13\% for the tested candidates included in this study. This method of classification does not require priors or templates, it is based on physical modelling, and can be combined with studies that classify the progenitor with other physical features.

Keywords

Cite

@article{arxiv.2304.11498,
  title  = {Characterizing the gravitational wave temporal evolution of the gmode fundamental resonant frequency for a core collapse supernova: A neural network approach},
  author = {Alejandro Casallas Lagos and Javier M. Antelis and Claudia Moreno and Michele Zanolin and Anthony Mezzacappa and Marek J. Szczepańczyk},
  journal= {arXiv preprint arXiv:2304.11498},
  year   = {2023}
}

Comments

14 pages, 9 figures, 6 tables