pyopmnearwell.ml.upscale module#

Functionality to upscale data from an ensemble run on a radial grid to a cartesian grid.

Note: pylint: no-member is disabled, because it complains about the BaseUpscaler missing instance attributes, which are taken care of by the Upscaler protocol. pylint: pointless-string-statement is disabled, as it complains an attribute docstring in the Upscaler protocol.

class Upscaler(*args, **kwargs)[source]#

Bases: Protocol

Protocol class for upscalers.

This class is used for typing of abstract attributes of BaseUpscaler. MyPy will check if subclasses of BaseUpscaler implement the following instance attributes:

  • num_timesteps

  • num_layers

  • num_zcells

  • num_xcells

  • single_feature_shape

However, in comparison to missing abstract functions, no runtime error will be raised if they are missing.

See, e.g., https://stackoverflow.com/a/75253719 for an explanation.

Note: As of now (Python 3.11), each instance method of BaseUpscaler or a

subclass making use of one of the attributes, needs to have its self argumented annotated with Upscaler. In future python versions it should be possible to use class BaseUpscaler[Upscaler](ABC):... instead and remove these annotations.

property num_timesteps: int#

Return the number of simulation time steps.

Returns:

Result produced by the operation.

Return type:

int

property num_layers: int#

Return the number of geological layers.

Returns:

Result produced by the operation.

Return type:

int

property num_zcells: int#

Return the number of vertical grid cells.

Returns:

Result produced by the operation.

Return type:

int

property num_xcells: int#

Return the number of radial grid cells.

Returns:

Result produced by the operation.

Return type:

int

property single_feature_shape: tuple#

Return the expected shape of one upscaled feature.

Returns:

Result produced by the operation.

Return type:

tuple

property angle: float#

Return the angular extent of the cake grid in radians.

Returns:

Result produced by the operation.

Return type:

float

_abc_impl = <_abc._abc_data object>#
_is_protocol = True#
class BaseUpscaler[source]#

Bases: ABC

Extract and upscale data from an array of ensemble data.

This base class provides several methods to extract features from fine-scale radial simulations and upscale to coarse cartesian cells. Depending on the type of data, this is done by averaging/summing/etc. values along all cells that correspond to a coarse cell.

Additionally, the sparsity of the dataset can be increased by taking only some timesteps/horizontal cells.

The upscaled data is usually provided in form of two np.ndarrays, one for features and one for targets.

Subclasses need to implement __init__ and (if needed) create_ds methods.

The feature array will have shape (num_ensemble_runs, num_timesteps/step_size_t, num_layers, num_xcells/step_size_x, num_features). The target array will have shape (num_ensemble_runs, num_timesteps/step_size_t, num_layers, num_xcells/step_size_x, 1)

Note: All methods assume that all cells have the same height. if this is not the

case, the methods must be overridden.

abstractmethod create_ds()[source]#

Create the upscaled feature and target dataset.

Returns:

Result produced by the operation.

Return type:

Any

Parameters:

self (Upscaler)

reduce_data_size(feature, step_size_x=1, step_size_t=1, random=False)[source]#

Reduce the size of the input feature array by selecting elements with a fixed step size.

Parameters:

  • feature (np.ndarray) -- The input feature array.

  • step_size_x (int, optional) -- The step size for the x-axis. Defaults to 1.

  • step_size_t (int, optional) -- The step size for the t-axis. Defaults to 1.

  • random (bool, optional) -- If True, select elements randomly instead of using a fixed step size. Defaults to False. Not implemented yet.

  • self (Upscaler)

Returns:

The reduced feature array.

Return type:

np.ndarray

get_vertically_averaged_values(features, feature_index, disregard_first_xcell=True)[source]#

Average a selected feature over vertical cells within each layer.

Parameters:

  • features (np.ndarray) -- Ensemble feature array.

  • feature_index (Any) -- Index of the feature to process.

  • disregard_first_xcell (bool, optional) -- Whether to remove the innermost well cell.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

np.ndarray

get_radii(radii_file)[source]#

Read radial-cell centers and boundaries for upscaling.

Parameters:
Returns:

Result produced by the operation.

Return type:

tuple[np.ndarray, np.ndarray]

get_timesteps(simulation_length)[source]#

Create uniformly spaced simulation times.

Parameters:

  • simulation_length (float) -- Total simulation duration in days.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

np.ndarray

get_horizontically_integrated_values(features, cell_center_radii, cell_boundary_radii, feature_index, disregard_first_xcell=True)[source]#

Integrate a vertically averaged feature into equivalent Cartesian blocks.

cartesian block area.

Parameters:

  • features (np.ndarray) -- Ensemble feature array.

  • cell_center_radii (np.ndarray) -- Radii at radial-cell centers.

  • cell_boundary_radii (np.ndarray) -- Radii at radial-cell boundaries.

  • feature_index (int) -- Index of the feature to process.

  • disregard_first_xcell (bool, optional) -- Whether to remove the innermost well cell.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

Any

get_homogeneous_values(features, feature_index, disregard_first_xcell=True)[source]#

Extract a feature that is homogeneous within each layer.

Parameters:

  • features (Any) -- Ensemble feature array.

  • feature_index (Any) -- Index of the feature to process.

  • disregard_first_xcell (bool, optional) -- Whether to remove the innermost well cell.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

Any

get_analytical_PI(permeabilities, cell_heights, radii, well_radius)[source]#

Calculate the single-phase analytical Peaceman productivity index.

Parameters:

  • permeabilities (np.ndarray) -- Permeability values.

  • cell_heights (np.ndarray) -- Grid-cell heights.

  • radii (np.ndarray) -- Cell radii.

  • well_radius (float) -- Wellbore radius.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

np.ndarray

get_analytical_WI(pressures, saturations, permeabilities, temperature, surface_density, radii, well_radius, OPM)[source]#

Calculate a two-phase analytical well index from pressure-dependent fluid properties.

Parameters:

  • pressures (np.ndarray) -- Pressure values.

  • saturations (np.ndarray) -- Non-wetting saturation values.

  • permeabilities (np.ndarray) -- Permeability values.

  • temperature (float) -- Fluid temperature.

  • surface_density (float) -- Reference surface density.

  • radii (np.ndarray) -- Cell radii.

  • well_radius (float) -- Wellbore radius.

  • OPM (pathlib.Path) -- Path to the OPM installation.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

np.ndarray

get_data_WI(features, pressure_index, inj_rate_index, angle=1.0471975511965976)[source]#

Calculate a data-driven well index from pressure and injection-rate results.

Similar functionality to ensemble.calculate_WI, but can additionally treat multiple vertical cells in a layer correctly.

Parameters:

  • features (np.ndarray) -- Ensemble feature array.

  • pressure_index (int) -- Feature index containing pressure.

  • inj_rate_index (int) -- Feature index containing injection rate.

  • angle (float, optional) -- Cake-grid angle in radians.

  • self (Upscaler)

Returns:

Result produced by the operation.

Return type:

np.ndarray

_abc_impl = <_abc._abc_data object>#