Research news on Numerical simulations in gravitation & astrophysics

Numerical simulations in gravitation and astrophysics are computational techniques that solve the governing equations of gravitational dynamics—typically Einstein’s field equations of general relativity or the Newtonian/relativistic N-body problem—using discretization schemes such as finite difference, finite volume, spectral, or particle-based methods. They enable quantitative modeling of systems like compact-object binaries, gravitational-wave sources, stellar evolution, accretion flows, galaxy formation, and large-scale structure. These simulations incorporate additional physics (e.g., magnetohydrodynamics, radiation transport, nuclear microphysics) and employ high-performance computing, adaptive mesh refinement, and sophisticated time-integration algorithms to resolve multi-scale, nonlinear phenomena inaccessible to purely analytic approaches.

To discover new physics, AI may need to 'unlearn' the old one

A study in the Journal of Cosmology and Astroparticle Physics explores how a machine-learning strategy known as transfer learning could dramatically reduce the computational cost of searching for new physics beyond the standard ...

Mathematical method calculates most efficient Earth-moon route yet

Researchers have developed a mathematical method that enables more precise calculations of the most economical travel routes between the orbits of celestial bodies. To demonstrate this method, they calculated a more efficient ...

How a single star can reshape an entire galaxy

Astronomers who simulate galaxies do not always get the same result, even when they start from identical conditions. New research from Leiden University shows that this is not a flaw, but a consequence of how galaxies behave—and ...

page 1 from 5