The TITUS framework offers open source solvers for time and memory efficient deterministic proton and electron dose calculations in 2D/3D. We solve the continuous slowing down approximation (CSD) to transport equations using the dynamical low rank approximation. See [1] for a detailed description of the methodology and some initial results for 2D electron transport. Functionalities of TITUS include collided-uncollided splitting methods, high-order time integration, high-order spatial discretizations and GPU accelerations.
- Install julia and open REPL
- Add packages: Open package manager by typing
]
and
pkg> add ProgressMeter LinearAlgebra LegendrePolynomials QuadGK FastGaussQuadrature SparseArrays SphericalHarmonicExpansions SphericalHarmonics TypedPolynomials GSL MultivariatePolynomials Einsum CUDA Distributions PyCall PyPlot DelimitedFiles WriteVTK Interpolations Images FileIO
- Set dimensions, rank, solver and particle type of your problem in main.jl and choose test case (defined in settings.jl)
- run main.jl, in REPL:
include("main.jl")
or from command line
julia main.jl
- Results are written to the output folder
There are many branches containing subprojects and updates we are working on, the names should be mostly self-explanatory. master contains the code used for [1] and 3DProtons, e.g. has the latest extensions to 3D proton transport, which are however still in work.
To cite our associated paper use @article{csd-dlra2021, title={A robust collision source method for rank adaptive dynamical low-rank approximation in radiation therapy}, author={Kusch, Jonas and Stammer, Pia}, journal={arXiv preprint arXiv:2111.07160}, year={2021} }