% Some notes on using the code % Jānis Erdmanis % November 18 2016
Magnetic droplet surface evolution boundary integral algorithms (BIE or BEM) according to magnetostatics and Stokes equations in general 3D case are presented in this repository.
The code originaly was written with julia 0.4 but now essentials are ported for julia 1.0 and will run on julia 1.0 when warnings will be adressed. It is now required that LINUX is being used for running simulation which is due to a binary dependency ElTopo.
To run the code first repoitory needs to be cloned whch will be in MDrop folder. Then in julia 0.7 one executes a following commands in the REPL (assuming that . is in repository folder)
]activate .
resolve
which will install all the necessary dependencies for the code. To proceed then with a calculation one then executes a following commands:
include(".juliarc.jl")
include("config.jl")
include("FastField.jl")
The code is written in julia 0.4 and tested extensively on Ubuntu 16.04. Main dependencies of the code are JLD.jl which stores and loads meshes and simulations; SurfaceGeometry.jl which deals with mesh generation, stabilization and others which are used for processing simulation results;
Assuming you are Ubuntu install required dependencies with sudo apt-get install -y libcgal-dev liblapack-dev libblas-dev, sudo apt-get install csh hdf5-tools cmake and if you want to use Distmesh for mesh generation you also need Matlab. In julia REPL execute following lines
Pkg.init()
Pkg.add("JLD")
Pkg.add("ArgParse")
Pkg.add("FastGaussQuadrature")
Pkg.clone("https://github.com/akels/SurfaceGeometry.jl")
Pkg.build("SurfaceGeometry")
Since packages might break features in the future I provide backup in libs.zip. Then installation is just unzipping this folder in julia package directory.
Optional dependencies which are needed to view and process simulation data are installed with sudo apt-get install python-matplotlib and in julia REPL
Pkg.add("ThreeJS")
Pkg.add("Escher")
Pkg.add("PyPlot")
Pkg.add("Elliptic")
For running simulation execute in MDrop directory julia -L .juliarc.jl mdrop.jl --simulation=FastField --Bm=20 --mu=10 which will run FastField.jl and sore results in ~/SimulationData. Instead of passing command line arguments you can run simulation also as julia -L .juliarc.jl -L config.jl Fastfield.jl.
Erdmanis, J. & Kitenbergs, G. & Perzynski, R. & Cebers, A. (2017)
Magnetic droplet in rotating field: numerical simulation and comparison with experiment