Fast Simulation of Particulate Suspensions Enabled by Graph Neural Network

  title={Fast Simulation of Particulate Suspensions Enabled by Graph Neural Network},
  author={Zhan Ma and Zisheng Ye and Wenxiao Pan},



Learning to Simulate Complex Physics with Graph Networks

A machine learning framework and model implementation that can learn to simulate a wide variety of challenging physical domains, involving fluids, rigid solids, and deformable materials interacting with one another, and holds promise for solving a wide range of complex forward and inverse problems.

Unveiling the predictive power of static structure in glassy systems

This work determines the long-time evolution of a glassy system solely from the initial particle positions and without any handcrafted features, using graph neural networks as a powerful model, and shows that this method outperforms current state-of-the-art methods, generalizing over a wide range of temperatures, pressures and densities.

Numerical prediction of colloidal phase separation by direct computation of Navier–Stokes equation

Numerical prediction of out-of-equilibrium processes in soft and bio matter containing liquids is highly desirable. However, it is quite challenging primarily because the motions of the components at

Spectral Ewald Acceleration of Stokesian Dynamics for polydisperse suspensions

Fast Stokesian dynamics

In FSD, the standard system of linear equations for SD is reformulated using a single saddle point matrix and this reformulation is generalizable to a host of particular simulation methods enabling the self-consistent inclusion of a wide range of constraints, geometries and physics in the SD simulation scheme.

Simulation of concentrated suspensions using the force-coupling method

Accelerated Stokesian Dynamics simulations

A new implementation of the conventional Stokesian Dynamics (SD) algorithm, called accelerated Stokesian Dynamics (ASD), is presented. The equations governing the motion of N particles suspended in a

Rotne–Prager–Yamakawa approximation for different-sized particles in application to macromolecular bead models

Abstract The Rotne–Prager–Yamakawa (RPY) approximation is a commonly used approach to model the hydrodynamic interactions between small spherical particles suspended in a viscous fluid at a low

Propagation Networks for Model-Based Control Under Partial Observation

Propagation Networks is introduced, a differentiable, learnable dynamics model that handles partially observable scenarios and enables instantaneous propagation of signals beyond pairwise interactions and achieves superior performance on various control tasks.