perigrain

A high-fidelity granular media simulation and analysis platform with grain deformation and fracture using peridynamics and discrete element method

Installation

# system packages
sudo apt-get install -y libhdf5-serial-dev gmsh python3-tk

# eigen library
wget -c https://gitlab.com/libeigen/eigen/-/archive/3.3.9/eigen-3.3.9.tar.gz 
mkdir lib
tar -xvf eigen-3.3.9.tar.gz -C lib/
rm eigen-3.3.9.tar.gz

# python packages
python3 -m venv env
source env/bin/activate
pip3 install -r requirements.txt

# generate the executable
make

Dependencies

Eigen (Specify the location of the Eigen library path in makefile if using other versions)
gmsh to generate mesh:
To read hdf5 files on Linux using h5dump:
Python packages
- matplotlib, numpy as usual
- mesh related python dependencies (pip3 install): pygmsh, gmsh, meshio
- optimization using gekko
- parallel processing using multiprocessing, pathos

Tested on:

Ubuntu 18.04
Ubuntu 20.04
Ubuntu 22.04

Experiment setup and running

Sample simulations

There are sample scripts in the /scripts directory. Just call the executable run.sh.

Specify the experiment in gen_setup.py. This uses dictionaries:
- exp_dict.py
- shape_dict.py
- material_dict.py

Custom simulations

Write your own simulation setup in a function called new_function() in the file exp_dict.py. Particle geometry and material and contact parameters are set here.
In gen_setup.py, add the line exp = exp_dict.new_function().
Specify the timesteps and simulation parameters in config/main.conf. Here, you can toggle various runtime features.

Then run:

# Generate the setup and setup.png
python3 gen_setup.py

# Copy `meshdata/all.h5` to `data/hdf5/all.h5`
make getfresh_py

# run the time integration steps (or ex3 for 3d)
make ex2

# generation simulation plots
make genplot

Cite

@article{doi:10.1137/21M1439389,
author = {Bhattacharya, Debdeep and Lipton, Robert P.},
title = {Simulating Grain Shape Effects and Damage in Granular Media Using PeriDEM},
journal = {SIAM Journal on Scientific Computing},
volume = {45},
number = {1},
pages = {B1-B26},
year = {2023},
doi = {10.1137/21M1439389},

URL = { 
        https://doi.org/10.1137/21M1439389
},
eprint = { 
        https://doi.org/10.1137/21M1439389
}
,
    abstract = { Abstract. We provide a numerical platform for the analysis of particle shape and topology effect on the macroscopic behavior of granular media. We work within a discrete element method (DEM) framework and apply a peridynamic model for deformable particles accounting for deformation and damage of individual particles. To accommodate arbitrary particle shapes including nonconvex ones as well as particle topology, an efficient method is developed to keep intraparticle peridynamic interaction within particle boundaries. Particle contact with the rigid boundary wall is computed analytically to improve accuracy. To speed up simulations with particles of different shapes and sizes the initial configuration is chosen using security disks containing different particle shapes that are placed in a jammed state using an optimization-based method. The effect of particle shape and topology on settling and compaction of the aggregate for deformable particles is analyzed. }
}

Perigrain

Install / Use

README