Physical Types

A physical type corresponds to physical quantities with dimensionally compatible units. For example, the physical type mass corresponds to physical quantities with units that can be converted to kilograms. Physical types are represented as instances of the PhysicalType class.

Accessing Physical Types

Using get_physical_type() lets us acquire PhysicalType instances from strings with a name of a physical type, units, Quantity instances, objects that can become quantities (e.g., numbers), and PhysicalType instances.

>>> import astropy.units as u
>>> u.get_physical_type('speed')  # from the name of a physical type
PhysicalType({'speed', 'velocity'})
>>> u.get_physical_type(u.meter)  # from a unit
PhysicalType('length')
>>> u.get_physical_type(1 * u.barn * u.Mpc)  # from a Quantity
PhysicalType('volume')
>>> u.get_physical_type(42)  # from a number
PhysicalType('dimensionless')

The physical type of a unit can be accessed via its physical_type attribute:

>>> u.coulomb.physical_type
PhysicalType('electrical charge')
>>> (u.meter ** 2).physical_type
PhysicalType('area')

Using Physical Types

An equality comparison between a PhysicalType and a string will return True if the string is a name of the PhysicalType:

>>> acceleration = u.get_physical_type(u.m / u.s ** 2)
>>> acceleration == 'acceleration'
True

Some units may correspond to multiple physical types because compatible units can be used to quantify different phenomena:

>>> u.get_physical_type('pressure')
PhysicalType({'energy density', 'pressure', 'stress'})

We can iterate through the names of a PhysicalType:

>>> for name in u.J.physical_type: print(name)
energy
torque
work

We can test for membership or equality with a string that has the name of a PhysicalType:

>>> 'energy' == u.J.physical_type
True
>>> 'work' in u.J.physical_type
True

Dimensional Analysis

PhysicalType instances support multiplication, division, and exponentiation. Because of this, they can be used for dimensional analysis:

>>> length = u.get_physical_type('length')
>>> time = u.get_physical_type('time')
>>> length ** 2
PhysicalType('area')
>>> 1 / time
PhysicalType('frequency')

Dimensional analysis can be performed between a PhysicalType and a unit or between a PhysicalType and a string with a name of a PhysicalType:

>>> length ** 2 / u.s
PhysicalType({'diffusivity', 'kinematic viscosity'})
>>> length / 'time'
PhysicalType({'speed', 'velocity'})