Variable presets and data transformation

ogstools.variables provides classes (Scalar, Vector, Matrix) which encapsulate unit handling and data transformation for simplified processing of mesh data. There are also several predefined variables.

import ogstools as ot
from ogstools import examples

ot.variables.get_dataframe()

	data_name	data_unit	output_unit	output_name	type
preset
heatflowrate	HeatFlowRate			HeatFlowRate	Scalar
massflowrate	MassFlowRate			MassFlowRate	Scalar
material_id	MaterialIDs			MaterialIDs	Scalar
nodal_forces	NodalForces			NodalForces	Vector
...	...	...	...	...	...
temperature	temperature	K	°C	temperature	Scalar
temperature_BHE	temperature_BHE	K	°C	temperature_BHE	Components_BHE
time	timevalues	s	s	time	Scalar
velocity	velocity	m/s	m/s	darcy_velocity	Vector

22 rows × 5 columns

Scalar, Vector and Matrix inherit from the class Variable with its transform() function. This function converts the argument from data_unit to output_unit and applies a function if specified. In this case we convert from K to °C:

ot.variables.temperature.transform(273.15, strip_unit=False)

0.0 °C

You can also create your own variables by creating a Scalar, Vector or Matrix variable. The following doesn’t do any unit conversion.

custom_temperature = ot.variables.Scalar(
    data_name="temperature", data_unit="K", output_unit="K"
)
custom_temperature.transform(273.15, strip_unit=False)

273.15 K

Or use existing presets as a template and replace some parameters:

custom_temperature = ot.variables.temperature.replace(output_unit="°F")
custom_temperature.transform(273.15, strip_unit=False)

31.999999999999936 °F

Components of Vector variables and Matrix variables can be accessed with bracket indexing. Vector variables should be of length 2 or 3 corresponding to the dimension. Matrix variables likewise should be of length 4 [xx, yy, zz, xy] or 6 [xx, yy, zz, xy, yz, xz].

ot.variables.displacement[1].transform([0.01, 0.02, 0.03], strip_unit=False)

0.02 m

ot.variables.strain["xx"].transform(
    [0.01, 0.02, 0.03, 0.04, 0.05, 0.06], strip_unit=False
)

1.0 %

Magnitude of a 2D displacement vector:

ot.variables.displacement.magnitude.transform([0.03, 0.04], strip_unit=False)

0.05 m

We suggest specifying the variables and their transformations once. These can be reused in different kind of post processing. When plotting with ogstools.plot we can use these presets to simplify the task of processing the data (e.g. calculate the von Mises stress):

fig = ot.plot.contourf(
    examples.load_mesh_mechanics_2D(), ot.variables.stress.von_Mises
)
fig.show()

Some variables store functions with arguments or parameters.

ot.variables.fluid_pressure_criterion.function

Function(callable=<function fluid_pressure_criterion at 0x7f0e2b5c5f80>, args=['pressure'], params={'biot': 1.0})

These can be changed in the following way.

create a copy of the variables to not modify the original:

fpc = ot.variables.fluid_pressure_criterion.copy()
fpc.function.params["biot"] = 0.6  # change a parameter
fpc.function

Function(callable=<function fluid_pressure_criterion at 0x7f0e2b5c5f80>, args=['pressure'], params={'biot': 0.6})

Have a look at Plotting the data for more examples.