N-body simulation of binary star mass transfer using NVDIA GPUs

Shaukat Goderya,Edward Smith,Baylor Fain,Taylor Hutyra,Mason McCallum,Bryant Wyatt

2023-03-22

Abstract:Binary star systems are of particular interest to astronomers because they can be used as astrophysical laboratories to study the properties and processes of stars. Between 70% to 90% of the stars in our galaxy are part of a binary star system. Among the many types of binary systems observed, the dynamics of semi-detached and contact systems are the most interesting because they exhibit mass transfer, which changes the composition and life cycle of both stars. The time scales of the mass transfer process are extremely large which makes the process impossible to capture through physical observation. Computer simulations have proved invaluable in refining our understanding of the mass transfer processes. Here we introduce an intuitive, computationally efficient, gravity centered model that simulates the filling of the Roche lobe of an expanding star and its transfer of mass through the first Lagrangian point.

Solar and Stellar Astrophysics,Astrophysics of Galaxies,Instrumentation and Methods for Astrophysics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to simulate the mass transfer process in binary star systems. Specifically, the paper focuses on the mass transfer phenomenon in semi - detached binary star systems, where one star expands to its Roche lobe boundary and transfers mass to its companion star through the first Lagrangian point (L1). Since the time scale of this process is very long and it is difficult to capture it through actual observations, computer simulations are crucial for understanding the mass transfer process.

The paper proposes a new simulation model based on the Discrete Element Method (DEM) to replace the traditional Smoothed Particle Hydrodynamics (SPH) method. This model focuses on the gravitational interactions between particles to drive the simulation, aiming to provide a more intuitive, simple and equally effective model, enabling researchers to observe the mass transfer process of binary star systems in real - time.

By using NVIDIA GPU - accelerated computing, this model can run efficiently on modern graphics processing units, reducing the cost and complexity of the simulation, making it possible to carry out such simulations in most research departments. This not only helps to gain a deeper understanding of the evolution process of binary star systems, but also provides new tools and methods for research in related fields.