3D GVEC equilibria in Python
PyPI install (not yet :-):
pip install gvec_to_pythonOr from source:
git clone git@gitlab.mpcdf.mpg.de:spossann/gvec_to_python.git
cd gvec_to_python
pip install -e .Compile kernels for faster evaluations:
compile_gvec_to_pythonUsage
The Galerkin Variational Equilibrium Code (GVEC) calculates magneto-hydrodynamic (MHD) equilibria for Tokamaks and Stellarators. gvec_to_python parses the GVEC output file (.dat) and collects the data in a .json file:
from gvec_to_python.reader.gvec_reader import create_GVEC_json
create_GVEC_json(dat_file_in, json_file_out) # give absolute paths to the filesIn a second step, callables of MHD equilibrium quantities are created as methods of the GVEC class:
from gvec_to_python import GVEC
gvec = GVEC(json_file_out, mapping='gvec')Profiles can be evaluated via
gvec.profiles.profile(s, name='phi') # toroidal flux profile (radial coordinate s~sqrt(phi_norm)
gvec.profiles.profile(s, name='chi') # poloidal flux profile
gvec.profiles.profile(s, name='iota') # iota profile
gvec.profiles.profile(s, name='pressure') # pressure profilewhere the radial coordinate s is the square-root of the normalized toroidal flux. Profile derivatives are callable via
gvec.profiles.profile(s, name='phi', der='s') # first derivative
gvec.profiles.profile(s, name='phi', der='ss') # second derivative
The mapping and metric coefficients are called via
gvec.f(s, a1, a2) # mapping
gvec.df(s, a1, a2) # Jacobian matrix
gvec.det_df(s, a1, a2) # Jacobian determinant
gvec.df_inv(s, a1, a2) # inverse Jacobian matrix
gvec.g(s, a1, a2) # metric tensor
gvec.g_inv(s, a1, a2) # inverse metric tensorThe radial coordinate denotes s is always the square-root of the normalized toroidal flux, a1 is the poloidal angle and a2 denotes the toroidal angle.
Five different mappings can be invoked by the mapping.setter:
# gvec standard coordinates: (s, th, ze) -> (x, y, z)
# from (s,a1,a2)=(s,theta,zeta) in [0,1],[0,2pi],[0,2pi] to cartesian coordinates (x,y,z)
gvec.mapping = 'gvec'
# gvec straight-field-line mapping (PEST) (s, theta*, zeta*) -> (x, y, z)
# from (s,a1,a2)=(s,theta*,zeta*) in [0,1],[0,2pi],[0,2pi] to cartesian coordinates (x,y,z)
gvec.mapping = 'pest'
# gvec with unit cube as logical domain (s, u, v) -> (x,y,z)
# from (s,a1,a2)=(s,u,v) in [0,1],[0,1],[0,1] to cartesian coordinates (x,y,z)
gvec.mapping = 'unit'
# gvec straight-field-line (PEST) with unit cube as logical domain (s, u*, v*) -> (x,y,z)
# from (s,a1,a2)=(s,u*,v*) in [0,1],[0,1],[0,1] to cartesian coordinates (x,y,z)
gvec.mapping = 'unit_pest'
# gvec without hmap (s,th,ze) -> (X1,X2,zeta)
# from (s,a1,a2)=(s,theta,zeta) in [0,1],[0,2pi],[0,2pi] to GVECs internal coordinates (X1,X2,zeta)
# if default torus (hmap=1) is used in GVEC, then (R,Z,phi)=(X1,X2,-zeta)
gvec.mapping = 'wo_hmap'
The MHD quantities are called via
gvec.p0(s, a1, a2) # pressure as 0-form
gvec.p3(s, a1, a2) # pressure as 3-form
gvec.bv(s, a1, a2) # contra-variant B-field
gvec.b1(s, a1, a2) # co-variant B-field (1-form)
gvec.b2(s, a1, a2) # 2-form B-field
gvec.b_cart(s, a1, a2) # Cartesian B-field
gvec.av(s, a1, a2) # contra-variant vector potential
gvec.a1(s, a1, a2) # co-variant vector potential (1-form)
gvec.a2(s, a1, a2) # 2-form vector potential
gvec.a_cart(s, a1, a2) # Cartesian vector potential
gvec.jv(s, a1, a2) # contra-variant current
gvec.j1(s, a1, a2) # co-variant current (1-form)
gvec.j2(s, a1, a2) # 2-form current
gvec.j_cart(s, a1, a2) # Cartesian current
