Skip to main content
We use two types of modelling frameworks at TransitionZero, each with its own purpose based on their strengths and limitations. They are:

​
Long-Term Capacity Expansion (OSeMOSYS)

This is a least-cost optimisation model employed for long-term strategic planning. Its primary objective is to determine the optimal configuration of future capacity additions to an existing system over a defined planning horizon. This involves identifying the specific types, scales, and deployment schedules of new assets required to satisfy projected future demand while optimising a defined objective function, typically the minimisation of total discounted costs. Consequently, it serves as a critical analytical tool for evaluating alternative investment pathways and informing strategic infrastructure development decisions. The Capacity Expansion model used here is TZ-OSeMOSYS - a Python package developed by TransitionZero using the open source OSeMOSYS modelling framework as the basis.

​
Single-Year Dispatch (PyPSA)

Dispatch modelling focuses on determining the optimal operation of a power system’s resources (generators, storage, and controllable loads) to meet electricity demand at each point in time, typically over a short-term horizon (e.g., hourly or sub-hourly for the next day or week). The goal is to minimise operational costs (like fuel costs for generators) while satisfying demand and respecting various technical and network constraints. The Dispatch model used here is PyPSA. It provides a comprehensive and flexible open-source environment for conducting sophisticated dispatch modelling of power systems, considering technical constraints, economic objectives, and the increasing complexity introduced by renewable energy sources and sector coupling. It allows users to analyse short-term system operation and evaluate the impact of different dispatch strategies and technologies.

​
Single-Year Capacity Expansion (PyPSA)

Capacity expansion is optionally available for all PyPSA scenarios, and enabled by default when creating scenarios in future years. This extends our PyPSA dispatch scenarios with single-investment-period capacity expansion.

​
Choosing a Capacity Expansion Framework:

  • Use PyPSA with single-investment-period capacity expansion for higher temporal resolution: better models of demand and weather variability, more realistic transmission and storage utilisation.
  • Use OSeMOSYS for more realistic long-term planning: growth constraints, stock turnover, etc.
Growth rates and long-term considerations may lead to OSeMOSYS and PyPSA choosing a very different capacity mix in 2050. Why not try both and compare the results?

​
Soft-Linking

A common workflow is to run dispatch models based on long-term capacity expansion plans. This allows you to account for long-term dependencies using OSeMOSYS, while still seeing more detailed system dynamics with PyPSA. Feeding the optimised capacities from a long-term capacity expansion scenario into a dispatch scenario is called soft-linking. In Scenario Builder:
  • Run an OSeMOSYS scenario
  • Download the results
  • Create a PyPSA scenario for your year of choice.
  • Update the Installed Capacity template with the optimal capacities for that year. Optionally: turn off capaciy expansion by setting Maximum Additional Capacity to zero.
  • Solve and inspect the results.
  • The increased time resolution may lead to infeasibilities now that demand and supply must be matched for every hour of the year. From here, you may want to:
    • Re-run the long-term expansion model with modified inputs (higher demand, lower renewable availability)