Homec4science
Diffusion µSpectre (feat/implement_new_materials) Browse

muSpectre/README
b9f21ca7059dfeat/implement_new_mate…

Last Change
Raw File

README
View Options

= µSpectre =
Copyright © 2018 Till Junge
µSpectre is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License as
published by the Free Software Foundation, either version 3, or (at
your option) any later version.
µSpectre is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details.
You should have received a copy of the GNU General Public License
along with GNU Emacs; see the file COPYING. If not, write to the
Free Software Foundation, Inc., 59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA.
This README contains only a small quick start guide. Please refer to the [full documentation](http://muspectre.readthedocs.io/en/master/) for more help.
== Building µSpectre ==
µSpectre is a CMake project that uses C++14. It depends on the Boost unit test framework for testing and uses uses python3 as secondary API. You will need a modern C++ compiler (µSpectre was tested with gcc-6 , gcc-7, clang-4 and clang5) and CMake version 3.7.0 or higher.
=== Compilation ===
$ git clone https://c4science.ch/source/muSpectre.git
$ cd build
$ cmake -DCMAKE_BUILD_TYPE=Release ..
$ make
µSpectre makes use of expression templates, as a result, production code **must** be compiled with the `CMAKE_BUILD_TYPE` flag set to `Release` in order to get performance (under gcc, non-optimised code is about 50 times slower than the release). Be careful with parallel compilation: compiling µSpectre is quite memory-hungry because of the use of expression templates, a parallel `make -j` requires currently about 10 GB of RAM under GCC.
== Simple usage example ==
The following is a simple example for using µSpectre through its convenient Python interface
#!/usr/bin/env python3
import numpy as np
import muSpectre as µ
# setting the geometry
resolution = [51, 51]
center = np.array([r//2 for r in resolution])
incl = resolution[0]//5
lengths = [7., 5.]
formulation = µ.Formulation.small_strain
# creating the periodic cell
rve = µ.SystemFactory(resolution,
lengths,
formulation)
hard = µ.material.MaterialLinearElastic1_2d.make(
rve, "hard", 10e9, .33)
soft = µ.material.MaterialLinearElastic1_2d.make(
rve, "soft", 70e9, .33)
# assign a material to each pixel
for i, pixel in enumerate(rve):
if np.linalg.norm(center - np.array(pixel),2)<incl:
hard.add_pixel(pixel)
else:
soft.add_pixel(pixel)
tol = 1e-5
cg_tol = 1e-8
# set macroscopic strain
Del0 = np.array([[.0, .0],
[0, .03]])
if formulation == µ.Formulation.small_strain:
Del0 = .5*(Del0 + Del0.T)
maxiter = 401
verbose = 2
solver = µ.solvers.SolverCG(rve, cg_tol, maxiter, verbose=False)
r = µ.solvers.newton_cg(rve, Del0, solver, tol, verbose)
print("nb of {} iterations: {}".format(solver.name(), r.nb_fev))
You can find more examples using both the python and the c++ interface in the `bin/` and `tests` folder.
c4science · Help