gvec.core.compute#

GVEC Postprocessing - Compute Functions

gvec.core.compute.Evaluations(rho: Literal['int'] | int | float | DataArray | ndarray | Sequence | None = 'int', theta: Literal['int'] | int | float | DataArray | ndarray | Sequence | None = 'int', zeta: Literal['int'] | int | float | DataArray | ndarray | Sequence | None = 'int', state: State | None = None, nfp: int | None = None)#

Create an Evaluations dataset with a grid in Boozer coordinates.

This factory function generates a mesh in logical coordinates (rho, theta, zeta) based on a grid in Boozer coordinates. The grid has dimensions (“rad”, “pol”, “tor”), corresponding to the radial, poloidal, and toroidal directions.

If a 2D or 3D array for theta_B or zeta_B is passed, the corresponding coordinate for the poloidal/toroidal dimension needs to be set manually afterwards (e.g. ev[“alpha”] = (“pol”, values) and ev = ev.set_coords(“alpha”).set_xindex(“alpha”)).

Parameters:

rho ("int" | int | float | 1D array (DataArray, ndarray, list)) – The specification of the radial, radius-like coordinate. “int” will use the integration points from the state object.
theta_B (int | float | 1D, 2D or 3D array (DataArray, ndarray, list)) – The specification of the poloidal, angle-like Boozer coordinate. 1D assumes dimension “pol”, 2D assumes (“pol”, “tor”), 3D assumes (“rad”, “pol”, “tor”).
zeta_B (int | float | 1D, 2D or 3D array (DataArray, ndarray, list)) – The specification of the toroidal, angle-like Boozer coordinate. 1D assumes dimension “tor”, 2D assumes (“pol”, “tor”), 3D assumes (“rad”, “pol”, “tor”).
state (State) – The gvec.State object to create the grid for. Used to perform the Boozer transform.
radial_derivative (bool, default True) – Whether to compute the radial derivatives of the LA and NU_B variables, at fixed GVEC angles \((\vartheta(\rho_i,\vartheta_{B,j},\zeta_{B,k}),\zeta(\rho_i,\vartheta_{B,j},\zeta_{B,k}))\). Computes boozer transform at additional radial points \(\rho-\epsilon\), and uses a first order Finite Difference for the derivatives.
epsilon_FD (float, optional) – The offset in rho used for the Finite Difference computation of the radial derivatives, default 1e-8.
boozer_kwargs (dict, optional) – Additional keyword arguments to pass to the get_boozer method of the state object. These can be used to specify the Boozer transform parameters. For example the maximum mode number factor boozer_kwargs={"MNfactor": 3}.

Return type:

xr.Dataset

gvec.core.compute.EvaluationsBoozerCustom(rho, theta_B, zeta_B, state, **boozer_kwargs)#: Create a custom EvaluationsBoozer dataset with Boozer coordinates.

Deprecated since version v1.2.

Create an Evaluations dataset with a grid in PEST coordinates.

PEST coordinates are straight-fieldline coordinates with both the radial and toroidal coordinate being identical to their logical coordinates, i.e. rho_P = rho and zeta_P = zeta. Note that for GVEC the toroidal coordinate is not necessarily the cylindrical angle.

This factory function generates a mesh in logical coordinates (rho, theta, zeta) based on a grid in PEST coordinates. The grid has dimensions (“rad”, “pol”, “tor”), corresponding to the radial, poloidal, and toroidal directions.

If a 2D or 3D array for theta_P or zeta is passed, the corresponding coordinate for the poloidal/toroidal dimension needs to be set manually afterwards (e.g. ev[“alpha”] = (“pol”, values) and ev = ev.set_coords(“alpha”).set_xindex(“alpha”)).

Parameters:

rho ("int" | int | float | 1D array (DataArray, ndarray, list)) – The specification of the radial, radius-like coordinate. “int” will use the integration points from the state object.
theta_P (int | float | 1D, 2D or 3D array (DataArray, ndarray, list)) – The specification of the poloidal, angle-like PEST coordinate. 1D assumes dimension “pol”, 2D assumes (“pol”, “tor”), 3D assumes (“rad”, “pol”, “tor”).
zeta (int | float | 1D, 2D or 3D array (DataArray, ndarray, list)) – The specification of the toroidal, angle-like logical coordinate. 1D assumes dimension “tor”, 2D assumes (“pol”, “tor”), 3D assumes (“rad”, “pol”, “tor”).
state (State) – The gvec.State object to create the grid for. Used to perform the PEST transform.

gvec.core.compute.compute(ev: Dataset, *quantities: str, state: State | None = None, registry: Mapping | None = None) → Dataset#

Compute the target quantity and add it to the given evaluation dataset.

This method will recursively determine prerequisites, compute them and add them to the dataset as needed.

Parameters:

ev (xr.Dataset) – The evaluation dataset with the target grid (rad, pol, tor), coordinates (rho, theta, zeta) and possibly some precomputed quantities.
*quantities (str) – One or more names of the quantities to compute. See the default table of available quantities or call table_of_quantities to see all options.
state (gvec.State | None, optional) – A gvec.State object that can be used by quantities which require it. Not necessary if the desired quantities only depend on already computed quantities.
registry (Mapping | None, optional) – The registry of computable quantites to use. Default: the gvec.core.compute.QUANTITIES registry used to evaluate a gvec.State object.