Exploring Supernova Gravitational Waves with Machine Learning

<p>The gravitational wave strain from сore-collapse supernova simulations used in our analysis. The file contains 402 signals labeled as <em>s00A0O00</em> or s00A0O00.0, where:</p> <p><em>s00</em> -- corresponds to (zero-age) progenitor mass, e.g. s27 means 27 solar mass</p> <p><em>A0</em> -- corresponds for a degree of differential rotation</p> <p><em>O00 </em>or <em>O00.0 </em>-- corresponds to central angular velocity, e.g. O07 or O07.5 means that our model has a central angular velocity of 7 or 7.5 rad/s, respectively</p> <p>Our waveforms are represented as a quadrupole wave amplitude. One can get a strain <em>h</em> multiplied by the distance <em>D </em>(= 10 kpc) by the following formula:&nbsp;<em>hD</em> =&nbsp;<strong><em>our_data</em></strong>/3.66 cm; see Eq (20) of Dimmelmeier et al 2008 [<a href="https://journals.aps.org/prd/abstract/10.1103/PhysRevD.78.064056">link</a>] for more information. All waveforms are represented in the time range from -15 to 20 ms with a 0.001 ms step size. The time of zero corresponds to the time of bounce. See [<a href="https://arxiv.org/abs/2209.14542">https://arxiv.org/abs/2209.14542</a>] for more information.</p>