Fluid Properties in pandapipes

The pipelines in the pandapipes network are run with a certain fluid. This can be chosen individually for the net from the fluid library or by own creation. The fluids are defined by certain properties.

The following fluids currently exist in the library:

  • hgas and lgas (high and low calorific natural gas),

  • hydrogen,

  • methane,

  • water,

  • biomethane_pure and biomathane_treated (compositions below)

  • air.


Biomethane(s): A particularity of injecting biomethane in the gas grid in Germany is the addition of LPG to enhance the Wobbe-Index and the heating value of the gas and thus make it equivalent to the (high-calorific) natural gas transported in the grid. This addition is done in a gas treatment unit (Biogaseinspeiseanlage) upstream of the biomethane feed-in junction. The properties of the biomethane(s) were computed based on the package CoolProp.

Biomethane pure

Biomethane treated





0.75 %

0.69 %

carbon dioxide

2.9 %

2.68 %


0.2 %

0.19 %


0 %

5 %


0 %

0.5 %

Wobbe-Index (normal conditions) \([\frac{KWh}{m^3}]\)



HHV (normal conditions) \([\frac{KWh}{m^3}]\)



The Fluid Class

Inside this class, different properties with their values are implemented. These properties can be called by different functions. There exists a general function, which returns the values of the requested property. Furthermore, there are different specified functions to return directly the value of the density, viscosity and heat capacity.

In addition to the already existing properties there is a function, which allows to add new properties. It also warns if there is already a property with the same name and can overwrite an existing property with a new value.


The idea behind the properties is a functional correlation between physical quantities, e. g. a linear correlation between pressure and temperature based on the ideal gas law. This way the pandapipes components can ask for a specific fluid property, such as the density, at a given operation point. Some classes for different functional correlations have already been implemented. All properties have to inherit from the Property base class in order to be used in a fluid. These classes shall be introduced in the following. They all inherit from the pandapower JSONSerializableClass, as they shall also be saved and reloaded savely (see also the chapter Save and Load Networks).

Property Base Class

Property With Constant Value

Property With Linear Correlation

Property With Interpolated Measure Points