The purpose of this project is to understand whether fluid motion can be accurately represented as a very large collection of rigid spheres. The motion of fluids is traditionally modeled using the Navier-Stokes equations, whereas rigid body motion is governed by the Newton-Euler equations. This work attempts to achieve a high-resolution discrete representation of a continuum problem using a large count of bodies interacting via frictional contact. If possible, the use of rigid-body dynamics to simulate fluid dynamics would potentially open doors to new modeling approaches in many fields, including turbulence and fluid-solid interaction. Moreover, this technique would allow for faster simulations since its solution algorithm maps very well to modern Graphics Processing Unit (GPU) computing architectures.
Contributors: Contain Kelly