Machine Learning Methods for Stellar Collisions. I. Predicting Outcomes of SPH Simulations

TL;DR

This paper addresses the challenge of predicting stellar-collision outcomes and remnant masses in dense stellar systems by constructing a large grid of SPH collisions (27,720 runs) across MS ages, masses, $v_ty$, and $r_p$, then training multiple ML models to classify outcomes and regress remnant masses. It compares kNN, SVM, and neural networks, with inputs transformed to standardized scales and results indicating strong performance: classification balanced accuracy reaches 98.4% and final-mass predictions achieve relative errors as low as 0.11% and 0.15%. The trained models are released in the collAIder package to enable rapid, on-the-fly collision predictions within N-body simulations. This work enhances the ability to model exotic stellar phenomena and transient events arising from collisions, blue straggler formation, and potential pathways to massive black hole seeds in dense environments.

Abstract

Stellar collisions can occur frequently in dense cluster environments, and play a crucial role in producing exotic phenomena from blue stragglers in globular clusters to high-energy transients in galactic nuclei. Successive collisions and mergers of massive stars could also lead to the formation of massive black holes, serving as seeds for supermassive black hole in the early universe. While analytic fitting formulae exist for predicting collision outcomes, they do not generalize across different energy scales or stellar evolutionary phases. Smoothed particle hydrodynamics (SPH) simulations are often used to compute the outcomes of stellar collisions, but, even at low resolution, their computational cost makes running on-the-fly calculations during an $N$-body simulation quite challenging. Here we present a new grid of $27,720$ SPH calculations of main-sequence star collisions, spanning a wide range of masses, ages, relative velocities, and impact parameters. Using this grid, we train machine learning models to predict both collision outcomes (merger vs disruption, or flyby) and final remnant masses. We compare the performance of nearest neighbors, support vector machines, and neural networks, achieving classification balanced accuracy of $98.4\%$, and regression relative errors as low as $0.11\%$ and $0.15\%$ for the final stars $1$ and $2$, respectively. We make our trained models publicly available as part of the package collAIder, enabling rapid predictions of stellar collision outcomes in $N$-body models of dense star cluster dynamics.

Figures (1)

• Figure 1: Complete grid of SPH stellar collision simulations. The left and top axes show the primary and secondary masses in the collisions, respectively. Within each square, the corresponding times at which the collisions were sampled are shown, colored to indicate different times. In the lower right corner, we shown an example of how the pericenter distance ($r_p$) and velocity at infinity ($v_{\infty}$) are sampled at each age, $M_1$, and $M_2$ dimensions. The colors illustrate the number of stellar remnants, where we distinguish cases where the stars merge from those that result in a stripped star.