Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userClassifying the Equation of State from Rotating Core Collapse Gravitational Waves with Deep Learning
91
0
0.0
(
0
)
Added by
Matthew Charles Edwards
Publication date
2020
fields
Physics
and research's language is
English
Authors
Matthew C. Edwards
Ask ChatGPT about the research
No Arabic abstract
In this paper, we seek to answer the question given a rotating core collapse gravitational wave signal, can we determine its nuclear equation of state?. To answer this question, we employ deep convolutional neural networks to learn visual and temporal patterns embedded within rotating core collapse gravitational wave (GW) signals in order to predict the nuclear equation of state (EOS). Using the 1824 rotating core collapse GW simulations by Richers et al. (2017), which has 18 different nuclear EOS, we consider this to be a classic multi-class image classification and sequence classification problem. We attain up to 72% correct classifications in the test set, and if we consider the top 5 most probable labels, this increases to up to 97%, demonstrating that there is a moderate and measurable dependence of the rotating core collapse GW signal on the nuclear EOS.
rate research
Read More
Gravitational waves (GWs) generated by axisymmetric rotating collapse, bounce, and early postbounce phases of a galactic core-collapse supernova will be detectable by current-generation gravitational wave observatories. Since these GWs are emitted from the quadrupole-deformed nuclear-density core, they may encode information on the uncertain nuclear equation of state (EOS). We examine the effects of the nuclear EOS on GWs from rotating core collapse and carry out 1824 axisymmetric general-relativistic hydrodynamic simulations that cover a parameter space of 98 different rotation profiles and 18 different EOS. We show that the bounce GW signal is largely independent of the EOS and sensitive primarily to the ratio of rotational to gravitational energy, and at high rotation rates, to the degree of differential rotation. The GW frequency of postbounce core oscillations shows stronger EOS dependence that can be parameterized by the cores EOS-dependent dynamical frequency $sqrt{Gbar{rho}_c}$. We find that the ratio of the peak frequency to the dynamical frequency follows a universal trend that is obeyed by all EOS and rotation profiles and that indicates that the nature of the core oscillations changes when the rotation rate exceeds the dynamical frequency. We find that differences in the treatments of low-density nonuniform nuclear matter, of the transition from nonuniform to uniform nuclear matter, and in the description of nuclear matter up to around twice saturation density can mildly affect the GW signal. We find that approximations and uncertainties in electron capture rates can lead to variations in the GW signal that are of comparable magnitude to those due to different nuclear EOS. This emphasizes the need for reliable nuclear electron capture rates and for self-consistent multi-dimensional neutrino radiation-hydrodynamic simulations of rotating core collapse.
Gravitational waves are theorized to be gravitationally lensed when they propagate near massive objects. Such lensing effects cause potentially detectable repeated gravitational wave patterns in ground- and space-based gravitational wave detectors. These effects are difficult to discriminate when the lens is small and the repeated patterns superpose. Traditionally, matched filtering techniques are used to identify gravitational-wave signals, but we instead aim to utilize machine learning techniques to achieve this. In this work, we implement supervised machine learning classifiers (support vector machine, random forest, multi-layer perceptron) to discriminate such lensing patterns in gravitational wave data. We train classifiers with spectrograms of both lensed and unlensed waves using both point-mass and singular isothermal sphere lens models. As the result, classifiers return F1 scores ranging from 0.852 to 0.996, with precisions from 0.917 to 0.992 and recalls ranging from 0.796 to 1.000 depending on the type of classifier and lensing model used. This supports the idea that machine learning classifiers are able to correctly determine lensed gravitational wave signals. This also suggests that in the future, machine learning classifiers may be used as a possible alternative to identify lensed gravitational wave events and to allow us to study gravitational wave sources and massive astronomical objects through further analysis.
We have carried out an extensive set of two-dimensional, axisymmetric, purely-hydrodynamic calculations of rotational stellar core collapse with a realistic, finite-temperature nuclear equation of state and realistic massive star progenitor models. For each of the total number of 72 different simulations we performed, the gravitational wave signature was extracted via the quadrupole formula in the slow-motion, weak-field approximation. We investigate the consequences of variation in the initial ratio of rotational kinetic energy to gravitational potential energy and in the initial degree of differential rotation. Furthermore, we include in our model suite progenitors from recent evolutionary calculations that take into account the effects of rotation and magnetic torques. For each model, we calculate gravitational radiation wave forms, characteristic wave strain spectra, energy spectra, final rotational profiles, and total radiated energy. In addition, we compare our model signals with the anticipated sensitivities of the 1st- and 2nd-generation LIGO detectors coming on line. We find that most of our models are detectable by LIGO from anywhere in the Milky Way.
While gravitational waves have been detected from mergers of binary black holes and binary neutron stars, signals from core collapse supernovae, the most energetic explosions in the modern Universe, have not been detected yet. Here we present a new method to analyse the data of the LIGO, Virgo and KAGRA network to enhance the detection efficiency of this category of signals. The method takes advantage of a peculiarity of the gravitational wave signal emitted in the core collapse supernova and it is based on a classification procedure of the time-frequency images of the network data performed by a convolutional neural network trained to perform the task to recognize the signal. We validate the method using phenomenological waveforms injected in Gaussian noise whose spectral properties are those of the LIGO and Virgo advanced detectors and we conclude that this method can identify the signal better than the present algorithm devoted to select gravitational wave transient signal.
The next galactic core-collapse supernova (CCSN) has already exploded, and its electromagnetic (EM) waves, neutrinos, and gravitational waves (GWs) may arrive at any moment. We present an extensive study on the potential sensitivity of prospective detection scenarios for GWs from CCSNe within 5Mpc, using realistic noise at the predicted sensitivity of the Advanced LIGO and Advanced Virgo detectors for 2015, 2017, and 2019. We quantify the detectability of GWs from CCSNe within the Milky Way and Large Magellanic Cloud, for which there will be an observed neutrino burst. We also consider extreme GW emission scenarios for more distant CCSNe with an associated EM signature. We find that a three detector network at design sensitivity will be able to detect neutrino-driven CCSN explosions out to ~5.5 kpc, while rapidly rotating core collapse will be detectable out to the Large Magellanic Cloud at 50kpc. Of the phenomenological models for extreme GW emission scenarios considered in this study, such as long-lived bar-mode instabilities and disk fragmentation instabilities, all models considered will be detectable out to M31 at 0.77 Mpc, while the most extreme models will be detectable out to M82 at 3.52 Mpc and beyond.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
