This project aims to standardise gyrokinetic analysis.
A general pyro object can be loaded either from simulation/experimental data or from an existing gyrokinetics file.
In general pyrokinetics can do the following
- Read data in from:
- Gyrokinetic input files
- Simulations outputs
- Write input files for various GK codes
- Generate N-D pyro object for scans
- Read in gyrokinetic outputs
- Standardise analysis of gk outputs
At a minimum pyrokinetics needs the local geometry and species data. Example scripts can be found in the examples folder
Documentation can be found at readthedocs.
Pyrokinetics requires a minimum Python version of 3.9. It may be necessary to upgrade
pip to install:
$ pip install --upgrade pipTo install the latest release:
$ pip install pyrokineticsOtherwise, to install from source:
$ git clone https://github.com/pyro-kinetics/pyrokinetics.git
$ cd pyrokinetics
$ pip install .If you are planning on developing pyrokinetics use the following instead to install:
$ pip install -e .[docs,tests]Note that currently the installation of pyrokinetics requires an available Fortran compiler
To run the tests:
$ pip install -e .[tests]
$ pytest --cov .The simplest action in Pyrokinetics is to convert a gyrokinetics input file for code
'X' into an equivalent input file for code 'Y'. The easiest way to achieve this is to
use a Pyro object, which manages the various other classes in the API. For example,
to convert a GS2 input file to a CGYRO input file:
>>> from pyrokinetics import Pyro
>>> pyro = Pyro(gk_file="my_gs2_file.in") # file type is automatically inferred
>>> pyro.write_gk_file("input.cgyro", gk_code="CGYRO")There are many other features in Pyrokinetics, such as methods for building gyrokinetics input files using global plasma equilibria and/or kinetics profiles. There are also methods for analysing and comparing the results from gyrokinetics code runs. Please read the docs for more information.
After installing, simple pyrokinetics operations can be performed on the command line using either of the following methods:
$ python3 -m pyrokinetics {args...}
$ pyro {args...}For example, to convert a GS2 input file to CGYRO:
$ pyro convert CGYRO "my_gs2_file.in" -o "input.cgyro"You can get help on how to use the command line interface or any of its subcommands
by providing -h or --help:
$ pyro --help
$ pyro convert --helpPyrokinetics provides a Dockerfile from which you can build and run Docker containers.
To do so, you must have Docker installed on
your system. To build a local container:
$ docker build . -t pyrokineticsIt can then be run using:
$ docker run -it --rm -v ./path/to/local:/mymount pyrokineticswhere:
-itruns an interactive shell.--rmdeletes the Docker instance after use.-v ./path/to/local:/mymountmounts the local directory./path/to/localto the directory/mymountwithin the Docker container.
The container runs an IPython interpreter, with Pyrokinetics already installed. Note
that you will need to import Pyrokinetics before it can be used.
Pyro object comprised of
- Equilibrium
- LocalGeometry
- Miller
- Fourier (to be added)
- LocalGeometry
- Kinetics
- LocalSpecies
- Numerics
- GKCodes
- GKOutput
- For nonlinear simulations
- Fields (field, kx, ky, theta, time)
- Fluxes (field, species, moment, ky, theta, time)
- For linear simulations
- Fields (field, kx, ky, theta, time)
- Fluxes (field, species, moment, ky, theta, time)
- Eigenfunctions (field, ky, theta, time)
- Eigenvalues - growth rate and mode freq (ky, time)
- For nonlinear simulations
There also exists the PyroScan object which allows you to make a N-D parameter scan of Pyro objects
pyrokinetics currently supports
Future formats to be added are
- CHEASE
Sources of kinetic profile data currently supported are
- SCENE
- JETTO
- TRANSP
- GACODE
- PFILE
- IMAS
- ELITEINP
Future codes to be add
- SimDB
- OMFIT
The following gk codes are supported in pyrokinetics
- CGYRO
- GS2
- GENE
- TGLF
- GKW
- STELLA
- GX
The pyro object uses standardised reference values to normalise the results. It will automatically handle converting to a GK codes standard units.
Note any scans/parameter changes made will be in standard pyro units so please account for this.
Reference values
-
$T_{ref} = T_e$ Electron temperature at flux surface -
$n_{ref} = n_e$ Electron density at flux surface -
$m_{ref} = m_D$ Deuterium mass -
$v_{ref} = c_s = \sqrt{T_e/m_D}$ Ion sound speed at flux surface -
$B_{ref} = B_0$ Toroidal field at centre of the flux surface -
$L_{ref} = a$ Minor radius of the last closed flux surface -
$t_{ref} = a/c_s$ Ion sound time at flux surface -
$\rho_{ref} = \frac{c_s}{eB_0/m_D}$ Ion Larmor radius at flux surface
It is possible to change the reference units but proceed with caution
This project is used by the following institutions
- UKAEA
- University of York
Copyright owned by UKAEA. Pyrokinetics is licensed under LGPL-3.0, and is free to use, modify, and distribute.