pandapipes

An external grid is used to denote nodes with fixed values of pressure or temperature that should not be solved for during calculations. They often represent connections to higher-level grids (e.g., a medium pressure level connecting to a low pressure grid).

Naming Convention

The type of external grid is determined by its name suffix:

• p: Fixed pressure.
• t: Fixed temperature.
• pt: Both pressure and temperature are fixed at the connected junction.

Calculation Requirements

• Hydraulic calculations: At least one node with a fixed pressure must be present for each separate part of the grid.
• Heat transfer calculations: At least one node with a fixed temperature must be present for each separate part of the grid.
• Mass Flow: For nodes with fixed pressure, the mass flow into or out of the system is not known prior to calculation; it is a result of the pipeflow calculation.

Note: Connectivity checks will validate if these requirements are met.

In pandapipes, a network is represented by a pandapipesNet object. This object acts as a container for all network elements, organized into pandas.DataFrame tables. Each table corresponds to a specific element type (e.g., pipe, sink, source, junction).

Key characteristics of pandapipesNet:

• Data Storage: Elements are stored in DataFrames where each row represents one element and each column represents a parameter.
• Access Patterns: The pandapipesNet object is an ADict (Attribute Dictionary). You can access element tables using either dictionary syntax or attribute syntax:
  • Dictionary syntax: net['pipe']
  • Attribute syntax: net.pipe
• Inspection: Calling repr(net) provides a human-readable summary of the network, including the fluid type and the count/types of elements included.

import pandapipes as pp

# Create a network
net = pp.create_empty_network()

# Accessing a table via dictionary syntax
pipe_df_dict = net['pipe']

# Accessing a table via attribute syntax
pipe_df_attr = net.pipe

Calculations in pandapipes do not run directly on the user-facing pandas tables. Instead, they use an internal structure called pandapipes internal tables (pit), accessible via net["_pit"].

Key Components:

• pit: The standard internal structure consisting of two numpy arrays: one for all network nodes and one for all branches.
• active_pit: An optimized internal structure (net["_active_pit"]) that contains only the 'active' parts of the network. It excludes nodes and branches that were set out of service by the user or by the connectivity check.

Important Note on Node Indices: In the branch tables, FROM_NODE and TO_NODE refer to the indices of junctions in the pipe table. However, within the pit and active_pit, these indices refer to the positions within the node arrays. When nodes are removed to create an active_pit, these indices are re-mapped to ensure they point to the correct locations in the reduced tables.

The compressor component in pandapipes models pressure lift and compression power based on the following physical principles:

Pressure Lift

For a compressor between from_junction i and to_junction j, the pressure lift is calculated using the absolute pressure (relative pressure + ambient pressure) and the pressure_ratio $\Pi$.

If the mass flow $\dot{m}{ij} > 0$ kg/s: $$p_j = (p_i + p{amb}) \cdot \Pi - p_{amb}$$

Otherwise, $p_j = p_i$.

Compression Power

For $\dot{m}_{ij} > 0$, the required compression power (in Megawatts) is estimated via the ideal adiabatic change in enthalpy:

$$P_{\text{compr},ij} = \dot{m}_{ij}\frac{\kappa}{\kappa - 1} R_s z(p_i) T_i \cdot \left(\Pi ^{\frac{\kappa -1}{\kappa}}-1\right)\cdot 10^{-6}$$

Where:

• $\kappa$ is the isentropic exponent (assumed as 1.4).
• $R_s$ is the specific gas constant.
• $z(p)$ is the compressibility.
• $T_i$ is the temperature.