Developed by the Computational Physics Group at the University of Michigan.
http://www.umich.edu/~compphys/index.html
List of contributors:
Greg Teichert (Lead Developer)
Shiva Rudraraju
Koki Sagiyama
Krishna Garikipati
The mechanoChemIGA code is an isogeometric analysis based code used to solve the partial differential equations describing solid mechanics (including gradient elasticity) and chemistry (including the Cahn-Hilliard phase field model). It is built on the PetIGA [https://bitbucket.org/dalcinl/petiga/] and PETSc [https://www.mcs.anl.gov/petsc/] libraries, and it uses the automatic differentiation capabilities of the Sacado package from the Trilinos library [https://trilinos.github.io/].
This is version 0.2.2.
GNU Lesser General Public License (LGPL). Please see the file LICENSE for details. Note that the functions IGAElementNextFormFunction and IGAComputeProjectionFunction in the file src/output.cc, as well as all functions
in the file src/petigasnes_mod.h were derived from the PetIGA/src/petigasnes.c source code
in the PetIGA library [https://bitbucket.org/dalcinl/petiga/]. Accordingly,
we include the license/copyright notice for the PetIGA library here in the file LICENSE_PetIGA
to apply to the above functions.
This code has been developed under the support of the following:
Toyota Research Institute, Award #849910 "Computational framework for data-driven, predictive, multi-scale and multi-physics modeling of battery materials"
NSF DMREF grant: DMR1436154 "DMREF: Integrated Computational Framework for Designing Dynamically Controlled Alloy-Oxide Heterostructures"
NSF CDI Type I grant: CHE1027729 "Meta-Codes for Computational Kinetics"
DOE BES, Division of Materials Sciences and Engineering: Award #DE-SC0008637 that funds the PRedictive Integrated Structural Materials Science (PRISMS) Center at University of Michigan
If you write a paper using results obtained with the help of this code, please consider citing one or more of the following:
"A variational treatment of material configurations with application to interface motion and microstructural evolution" (Journal of the Mechanics and Physics of Solids)
G. Teichert, S. Rudraraju, K. Garikipati
\@article{Teichert2016a,
title = "A variational treatment of material configurations with application to interface motion and microstructural evolution ",
journal = "Journal of the Mechanics and Physics of Solids ",
volume = "99",
pages = "338 - 356",
year = "2017",
issn = "0022-5096",
doi = "https://doi.org/10.1016/j.jmps.2016.11.008",
url = "http://www.sciencedirect.com/science/article/pii/S0022509616305221",
author = "Gregory H. Teichert and Shiva Rudraraju and Krishna Garikipati",
}
"A comparison of Redlich-Kister polynomial and cubic spline representations of the chemical potential in phase field computations" (Computational Materials Science)
G. Teichert, H. Gunda, S. Rudraraju, A. Natarajan, B. Puchala, K. Garikipati, A. Van der Ven
\@article{Teichert2016b,
title = "A comparison of Redlich-Kister polynomial and cubic spline representations of the chemical potential in phase field computations ",
journal = "Computational Materials Science ",
volume = "128",
pages = "127 - 139",
year = "2017",
issn = "0927-0256",
doi = "https://doi.org/10.1016/j.commatsci.2016.11.024",
url = "http://www.sciencedirect.com/science/article/pii/S0927025616305754",
author = "Gregory H. Teichert and N.S. Harsha Gunda and Shiva Rudraraju and Anirudh Raju Natarajan and Brian Puchala and Krishna Garikipati and Anton Van der Ven",
}
"Mechano-chemical spinodal decomposition: A phenomenological theory of phase transformations in multi-component, crystalline solids" (Nature npj Computational Materials)
S. Rudraraju, A. Van der Ven, K. Garikipati
\@article{Rudraraju2016,
Title = {Phenomenological treatment of chemo-mechanical spinodal decomposition},
Author = {S. Rudraraju and A. Van der Ven and K. Garikipati},
Journal = {npj Computational Materials},
Year = {2016},
Volume = {2},
Doi = {10.1038/npjcompumats.2016.12}
}
A Dockerfile is included that creates a Docker image with all necessary libraries in an Ubuntu environment. Alternative installation instructions are as follows:
- Install PETSc:
-Download and extract PETSc source code.
-Quick installation as follows (the symbol $ denotes the command prompt):
$ ./configure --with-cc=gcc --with-cxx=g++ --with-fc=gfortran --download-fblaslapack --download-mpich --download-metis --download-parmetis --download-superlu_dist
$ make all test
-Set the appropriate PETSC_DIR and PETSC_ARCH environment variables, e.g.
$ export PETSC_DIR=/path/to/petsc-3.8.3
$ export PETSC_ARch=arch-linux2-c-debug
Note that this also installs mpich. It is possible to use an existing version of mpi by including a flag to its directory. If you will be using the local PETSc installation of mpich, set the following:
$ alias mpirun=$PETSC_DIR/$PETSC_ARCH/bin/mpirun
Download: https://bitbucket.org/petsc/petsc/get/v3.8.3.tar.gz
Installation instructions: http://www.mcs.anl.gov/petsc/documentation/installation.html
- Install PetIGA:
-Clone the PetIGA source code (confirmed to work with the mechanoChemIGA code at commit 5ecf484).
-Enter the PetIGA top directory and install using make:
$ git clone https://bitbucket.org/dalcinl/PetIGA.git
$ cd PetIGA
$ git reset --hard 5ecf484
$ make all
$ make test
-Export path to PetIGA directory:
$ export PETIGA_DIR=/path/to/petiga/
Installation instructions: https://bitbucket.org/dalcinl/petiga/
- Install CMake:
Download: https://cmake.org/download/
- Install the Sacado package from Trilinos (version 11.8.1 recommended)
-Download and extract or clone the Trilinos source code.
-Simple installation of Sacado as follows (from the Trilinos top directory), with the desire /path/to/trilinos/installation/:
$ mkdir build
$ cd build
$ cmake -DTrilinos_ENABLE_Sacado=ON -DTrilinos_ENABLE_Teuchos=OFF -DCMAKE_INSTALL_PREFIX=/path/to/trilinos/installation/ ../
$ make install
-Export path to Trilinos:
$ export TRILINOS_DIR=/path/to/trilinos/installation/
Download: http://trilinos.csbsju.edu/download/files/trilinos-11.8.1-Source.tar.gz
-
Install igakit (required to convert binary output files to .vtk files; also requires numpy and scipy)
$ pip install https://bitbucket.org/dalcinl/igakit/get/default.tar.gz
Download and installation instructions: https://bitbucket.org/dalcinl/igakit
To run the example initial boundary value problems, navigate to the desired example folder. Create a makefile using cmake:
$ cmake CMakeLists.txt
To compile the code (default is debug mode):
$ make
To switch the compilation to release mode:
$ make release
To switch the compilation to debug mode:
$ make debug
To run the code (replace "nproc" with the number of processors to be use; superlu_dist is used by default if available, gmres is used otherwise):
$ mpirun -np nproc ./main
To use use gmres instead of superlu_dist:
$ mpirun -np nproc ./main -ksp_type gmres -pc_type none
The ouput files created by the code are .dat binary files and a fieldInfo.txt file. These files can be converted to .vtk files and visualized using tools such as VisIt or ParaView using the igakit package (see step 5 of the installation instructions). To do this file conversion, run the initBounValProbs/writeVTKFile.py script from the directory containing the output files. For example, if the .dat and fieldInfo.txt ouput files were located in the initBounValProb/nonGradientMechanics/3D folder, the following commands would create the .vtk files:
$ cd initBounValProb/nonGradientMechanics/3D
$ python ../../writeVTKFiles.py