Exascale multiphase flow solver — 2025 Gordon Bell Prize Finalist | 200T grid points on 43K+ GPUs
-
Updated
Nov 9, 2026 - Fortran
Exascale multiphase flow solver — 2025 Gordon Bell Prize Finalist | 200T grid points on 43K+ GPUs
Library for performing multiphase simulations (based on the Shan-Chen model) in complicated geometries (i.e. porous media 3D images)
OpenFOAM solver for performing single- and two-phase flow simulations on hybrid-scale porous media.
A powerful multi-phase load flow solver, by Roseau Technologies
ListLBM is a sparse lattice Boltzmann solver for multiphase flow in porous media
Extension of OpenFOAM's "interfaceProperties" library. A run-time selection mechanism was implemented to select the model for calculating the surface tension force & curvature. These models can reduce spurious currents.
An OpenFOAM-based software tool for Internal Ballistic simulations of solid rocket motors.
Lattice Boltzmann codes
GPU-accelerated project for multicomponent computational fluid dynamics (CFD) simulations using the Lattice Boltzmann Method (LBM) with CUDA C++.
Multiphase dq Transformation Toolkit for Ansys Electronics Desktop
lattice Boltzmann method-theory and application
2D Three-Phase Lattice Boltzmann Method Solver Parallelised with OpenMP
Optimization environment for TimeNET
Repository for the determination of the most promising machine learning algorithm and the best set of dimensionless features able to solve the challenging classification problem of adiabatic flow pattern prediction
Runs OpenFOAM v14 cases entirely on one NVIDIA GPU: laminar two-phase VoF (incompressibleVoF), icoFoam-style PISO and potentialFoam, translated from the GPL source. Not affiliated with OpenCFD or The OpenFOAM Foundation.
Lattice Boltzmann codes
A simple blog to keep tracks of what I am doing.
Defects in ANSYS CFX wall-film breakage: a rotating-wall frame mismatch, the same mismatch in the injection path, and an undocumented generation cap that silently drops about 30% of fragment count. Box reproducers only.
A physics-informed neural network (PINN) designed for the automated inversion of non-Newtonian multiphase landslide and debris flow parameters using satellite observation tracking constraints. This repository provides a high-performance framework that integrates the modified Saint-Venant shallow water dynamics, Herschel-Bulkley/Bingham rheology...
OpenFOAM solver for transonic/supersonic compressible flow with Lagrangian particles (sonicFoam + DPM)
To associate your repository with the multiphase topic, visit your repo's landing page and select "manage topics."