Python code for the numerical experiments presented in BvG17.
The code is stored in the subfolder /python_code. All experiments can be reproduced with a single run from the command line. For example:
python3 elast_squares.py --ndims=2 --dx=2.5 --dy=2.5 --dz=2.5 --freq=[1.0,9.0] --Nom=5 --degree=1 --damping=0.005 --maxit=300 --maxit_i=20 --tol=1e-8 --tol_i=1e-1 --dg_pp=3 --tau_re=0.7 --tau_im=-0.3 --iLU=True --fill_factor=10.0 --rot=True --block=True --plots=True --plot_resnrm=True --solver_flag=0 --nprocs=8
- nutils:
pip install git+https://github.com/joostvanzwieten/nutils@955bc67d219496e26b037f47855709a222850d7c - PyAMG:
pip install --upgrade pyamg - NumPy [v 1.8.2], SciPy [v 0.14.0], matplotlib [v 1.4.2]
The author is a PhD student in Numerical Analysis at TU Delft. My research is focused on linear solvers and preconditioning for the elastic wave equation. Feel free to get in touch.
- Manuel Baumann and Martin B. van Gijzen (2017). Efficient iterative methods for multi-frequency wave propagation problems: A comparison study. Procedia Computer Science, Volume 108, Pages 645-654.