Skip to content

Latest commit

 

History

History
74 lines (54 loc) · 3.55 KB

README.rst

File metadata and controls

74 lines (54 loc) · 3.55 KB
Raw

FAROES

FAROES is "Fusion Analysis, Research, and Optimization for Energy Systems". It is a Python package for optimizing fusion power plants, and especially for optimizing properties like their capital cost or levelized cost of energy. It is built in a modular fashion to allow user-developers to modify or add equations or analyses of their own design. It uses the OpenMDAO framework (openmdao.org) to provide interfaces to third-party nonlinear solvers and optimizers, especially gradient-based optimizers. The framework has advanced logging features to record solution data and metadata. This framework also allows users to easily specify the design variables, constraints, and optimization targets, or to define their own.

forthebadge made-with-python

Pytest build status Unittest build status MIT licensed

Requirements

  • numpy>=1.21.0
  • scipy
  • openmdao == 3.15.0. Can be installed with pip install "openmdao==3.15.0".
  • ruamel >= 0.16, a yaml parser.
  • plasmapy (https://www.plasmapy.org/).

All these can be installed automatically using the setup.py file.

Installation

Download the repo, cd to the folder and run pip install -e .. This will download and install requirements and install FAROES in 'editable' mode.

Recommended Packages

The two below can be installed on linux using the script available here: https://github.com/OpenMDAO/build_pyoptsparse/

  • pyoptsparse (https://github.com/mdolab/pyoptsparse) in order to access the pyOptSparseDriver which allows use of more powerful optimizers like IPOPT.
  • IPOPT. This is optimizer seems to perform a bit better than the implemntations of COBYLA or SLSQP which are included with scipy.

Note that an environment variable must be set in every terminal instance using IPOPT. We recommend adding a line to ~/.bashrc to do this automatically for every terminal instance: export LD_LIBRARY_PATH=$HOME/ipopt/lib:$LD_LIBRARY_PATH

Documentation Packages

  • docutils<0.17,>=0.15
  • nbclient<0.6,>=0.2
  • sphinx >= 4.4.0
  • jupyter-sphinx==0.3.2
  • Jinja2<3.1
  • sphinxcontrib-bibtex >= 2.3
  • jsonschema < 4
  • sphinx_rtd_theme

The above packages are used to build the documentation for FAROES, and can be downloaded automatically by downloading the repos, cd to the folder, and running pip install -e .[docs], which will install FAROES in 'editable' mode alongside with the packages needed to build the documentation.

Funding

This work was supported by the U.S. Department of Energy under contract number DE-AC02-09CH11466.

The developer grants to the Government, and others acting on its behalf, a nonexclusive, paid-up, irrevocable, world-wide license in such copyrighted data to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government.