Skip to content

Adding new technologies in PyPSA-SPICE

This note explains how to add new or custom technologies in PyPSA-SPICE that are not included in the default set. In PyPSA-SPICE, some energy technologies are modeled by combining multiple PyPSA components to capture the characteristics of a specific plant type. Below are a few common examples for reference.

Info

Adding new or custom technologies often requires a good understanding of both the PyPSA framework and PyPSA-SPICE to avoid errors during model execution.

If you are uncertain about making these changes, please reach out to us at modelling@agora-thinktanks.org. We would be happy to help you get started.

Example 1: Adding a multi-output plant (e.g., a new fossil plant with CO₂ emissions)

In PyPSA-SPICE, a multi-output plant is modeled using a PyPSA Link, where the fuel bus (bus0) provides the input and the output buses (bus1, bus2, etc.) represent the plant’s different outputs. For example, a gas-fired power plant (CCGT) can be implemented as a Link that connects the gas bus (bus0) to the electricity/power sector bus (bus1) as its main output, and to an atmosphere bus (bus2) as a second output to represent CO₂ emissions released as a by-product of combustion; additional outputs can be included by connecting to further buses such as bus3, bus4, and so on.

To add a new gas-fired power plant type (e.g. CCGT_Mod), follow these steps:

  • In technologies.csv: Add a new row for CCGT_Mod and fill in the technical parameters. Specify Link as class. Note: efficiency applies to the conversion from bus0 → bus1, efficiency2 to bus0 → bus2, and so on for additional outputs.
  • In power_plant_costs.csv: Add a new row for CCGT_Mod for each modeled year and enter the cost assumptions.
  • In power_links.csv: Add a new row for CCGT_Mod, fill in the required inputs, and make sure to set the buses correctly: bus0 = fuel, bus1 = power sector, bus2 = atmosphere.
  • In decomission_capacity.csv: Add a decommissioning plan for CCGT_Mod, if applicable.

Tip

In most cases, the new plant type is only slightly different from an existing one (e.g. CCGT_Mod vs CCGT).

To save time and reduce errors, copy the CCGT rows in all four files, change the relevant text to CCGT_Mod, and then modify only the parameters that differ.

Example 2: Adding a single-output plant (e.g., new renewable plant with time-dependent availability)

In PyPSA-SPICE, renewable plants with time-dependent availability are modeled as Generator component (hydro dams are the exception and use StorageUnit). To add a new renewable plant (e.g. Wind-PV-Hybrid), do the following:

  • In technologies.csv: Add a new row for Wind-PV-Hybrid and fill in the technical parameters. Specify Generator as class. Leave p_max_pu and p_min_pu empty, as the renewable plant's availability is defined in availability.csv.
  • In power_plant_costs.csv: Add a new row for each modeled year and enter the cost assumptions for Wind-PV-Hybrid.
  • In renewables_technical_potential.csv: Add new rows for Wind-PV-Hybrid that define the maximum buildable capacity by region (per country).
  • In availability.csv: Add the time-dependent availability profile (per unit) for Wind-PV-Hybrid by region/country.
  • In power_generator.csv: Add a new row for Wind-PV-Hybrid, fill in the required inputs, and make sure to set the bus information correctly.

Example 3: Adding a storage plant with a fixed energy-to-power ratio

In PyPSA-SPICE, by default, all storage plants are built from a combination of PyPSA Store and Link components, which lets the model optimize energy capacity and power capacity independently. If you want a custom storage type with a fixed energy-to-power ratio, use a PyPSA StorageUnit instead. For example, to add a battery storage plant (BATS_8) with an energy-to-power ratio of 8, follow these steps:

  • In technologies.csv: Add a new row for BATS_8 and fill in the technical parameters. Set max_hours = 8. Note that storage can have different charge/discharge efficiencies: use efficiency for discharging and efficiency_store for charging. Specify StorageUnit as class.
  • In power_plant_costs.csv: Add a new row for each modeled year and enter cost assumptions for BATS_8. CAPEX and FOM in the CSV are in \$/MW. If your costs are in \$/MWh, multiply by 8 to convert to \$/MW.
  • In storage_inflows.csv: Add the time-dependent inflow profile (MW) for BATS_8 by region/country (not applicable for batteries, but might apply for technologies like hydro dam).
  • In storage_capacity.csv: Add a new row for BATS_8, fill in the required inputs, and make sure to set the bus information correctly.