Some python codes to generate various aggregate particles (e.g. ballistic particle cluster, and cluster-cluster) and characterise them in a variety of ways. Note that this is not done via dynamic simulation, but takes a ray-tracing approach.
A conda environment file is provided, so after checking out the repo you can install all dependencies with:
conda env create
Note that currently simulation.show() can use either matplotlib or mayavi, but the inclusion of mayavi holds back various packages (e.g. scipy).
The builder
package contains routines to build different aggregate types. Currently BPCA (particle-cluster) and BCCA (cluster-cluster) aggregates are implemented. Future improvements will include:
- polydisperse monomers (using a variety of distributions)
- mixed models where either single monomers or clusters can be added
The returned simulation
object has a variety of methods to calculate aggregate properties, including:
- centre of mass
- radius of gyration
- characteristic radius
- porosity
- density
- fractal dimension (various methods)
There are also methods to display or write the simulation data to various formats:
show
- displays the aggregate in 3D usingmayavi
ormatplotlib
projection
- calculates a 2D projection of the aggregate in a given directionto_csv
- output to a simple comma separated value fileto_vtk
- output to a VTK file (uses evtk!)to_afm
- calculates a simulated AFM image assuming an infinite tipto_gsf
- usesto_afm
and outputs data to a Gwyddion simple file formatto_liggghts
- writes a data file that can be read by LIGGGHTS
The most simple usage is as shown below:
In [1]: from aggregate import builder
In [2]: agg = builder.build_bpca()
Generating particle 1024 of 1024
In [3]: agg.porosity_gyro()
Out[3]: 0.6464342819868891
In [4]: agg.com()
Out[4]: array([-0.29366738, -0.84070369, 0.81448045])
In [5]: agg.show(using='mpl')