Ember-PyPSA

Introduction

Ember-PyPSA is a dataset that can be used with the PyPSA open energy modelling framework to analyze the mechanics of power systems - currently covering all European Union countries and the UK. The model was created by Ember and described in a briefing on UK 2030 clean power and another one on wind and solar targets in Central and Eastern Europe. This github repository replaces PyPSA-CEE and PyPSA-UK, merging both models into a common, flexible framework.

Citation and license

The recommended citation is:

Ember. (2023). Ember-PyPSA - An open energy model of the European power sector based on the PyPSA framework. https://github.com/ember-climate/ember-pypsa

Ember-PyPSA is based on the PyPSA open energy modelling framework used by many research institutions around the world. Whenever using PyPSA-UK, please also give credit to the authors of PyPSA following the guidelines described in the PyPSA documentation.

Similarly to PyPSA, Ember-PyPSA is distributed under the MIT license.

Setup

To use the model you will need to have a basic understanding of Python and conda, an LP solver and a computer with enough memory (32 GB RAM). The number of nodes in the model was kept at a minimum to make it possible to run with the open solver HiGHS.

Before you start working with Ember-PyPSA, please follow the installation guidelines for PyPSA.

The basic installation steps can be summarized as follows:

1. Clone the Ember-PyPSA repository
2. Create a new conda environment
3. Install an LP solver (make sure it is available in the Python terminal)
4. Install the PyPSA package through conda

Getting started

To make the model as accessible as possible Ember-PyPSA is contained in just one excel sheet (per scenario) and one main Python file (and a set of helper functions). The excel sheet contains all the input data, and the Python file is used to import them into the model, run the optimization and analyse the results.

After you complete the installation procedue, it is quite easy to get first results:

1. Open the ember-pypsa.ipynb file script in the ember-pypsa folder using your IDE.
2. Set your solver name in cell 12: network.optimize(solver_name='highs').
3. Save the file and run it (if an error pops up, make sure the terminal points to the correct directory and Python can see your solver).
4. Once the optimization is finished, you can view selected results in the bottom of the Jupyter notebook file.

The notebook shows only basic outputs like electricity generation, but you there are several examples of processing results in the PyPSA documentation - such as plotting storage discharge profiles, SRMC, emissions, line loading etc. You can change scenarios by changing the data file path in cell 2.

Model design

Please see Ember's reports, e.g. the briefing on UK 2030 clean power for a full explanation of the assumptions behind the modelled scenarios. An excerpt is provided below.

For the UK, the scenarios provided with the model use 2030 capacity and demand expectations from National Grid’s Future Energy Scenarios (Falling short and Leading the Way pathway). For other EU countries, scenarios are based on the scenarios provided with the model use our own 2030 capacity assumptions and demand expectations from ENTSO-E's TYNDP (Distributed Energy pathway). The capacity assumptions come from a variety of sources - cross-checked gas and coal unit databases from Global Energy Monitor and Beyond Fossil Fuels for the existing fossil fuel fleet, national coal/nuclear phaseout announcements. The unit capacities for the existing fleet were aggregated and checked against Ember's data, IRENA, SolarPower Europe, WindEurope and others. Planned/in construction units were included. Wind and solar capacities come from a variety of sources, as described in detail in Ember's CEE report. Renewable capacities besides wind and solar were sources from NECPs and the existing fleet.

The model is run for each hour of the year 2030, with 12 nodes representing the regions of the UK and 29 nodes representing representing all EU countries except Luxembourg, Malta and Cyprus, as well as the Norway, Switzerland, Turkey and Russia. All existing interconnections were represented, with their capacities based on 2021 NTCs as well as ENTSO-E's 2025 reference grid. Current and planned power station locations are used (based on DUKES data, planned hydro pumped-storage units, the REPD database, among others), with some generators being decommissioned and some upgraded.

The input data includes weather and demand profiles from ENTSO-E's Pan-European Climate Database (PECD) for the most critical climatic year: 2008 (the baseline year). Other climate years from PECD can be used by changing the profiles manually in the Excel sheet.

Dispatching is optimised based on short-run marginal costs. 2030 price forecasts were based on the latest available futures contracts, except for lignite costs that were sourced from IEEFA and adapted according to the unit’s efficiency. Running costs for non-fossil fuel generators were set to represent their dispatching priority in the merit order. CHP operation is optimized only within boundaries based on average 2021 hourly generation for CHP units in the region. A minimum load of 40% for nuclear plants was assumed based on ENTSO-E’s ERAA input data, a maintenance schedule is also applied to match historical capacity factors. No ramp up/down limits were introduced due to their impact on model solving times, but these can be added in future works based on Ember-PyPSA.

Please note that Ember-PyPSA is not configured as a capacity expansion model (CEM), but rather a merit-order style dispatch model. The installed capacity for different technologies is provided manually as part of the scenario design. This decision was made to ensure that the implemented scenarios are realistic and aligned with national studies, industry forecasts, the existing potentials and the possible expansion within the limited 2030 timeframe. Running the model as a CEM means the capacity additions are set by the optimizer and do not always correspond to reality. It is possible to set the capex values for each technology and convert Ember-PyPSA to a CEM (an example is provided in the gen_gas_peak tab).

Changing input parameters

To change the scenario, access or clone the Excel sheet uk-2030-falling-short.xlsx.

A few categories of tabs can be seen:

• load - demand profiles for each region.
• pv, wind, wind_offshore, chp - solar, wind and chp generation profiles for all hours of the year.
• lines, links, buses, cbf - technical components of the model - the definition of the nodes, lines and interconnectors
• prices - price assumptions for fossil fuels and hydrogen
• gen_XXX, st_XXX - capacity assumptions for each generator and storage technology. Most parameters should be self-explanatory, the p_nom parameter is the installed capacity, for more explanation please refer to the PyPSA components documentation

Further improvements

Many improvements can be made in the Ember-PyPSA model, some of them include:

• Hydrogen - including electrolyzers in all countries, currently they are implemented only in the UK
• Thermal unit flexibility - adding ramp-up/down and minimum utilization constraints for thermal units. This was removed from the model due to the significant increase in model solving times, but it is possible to implement these constraints in PyPSA.
• Demand disaggregation - the UK demand profiles are disaggregated into regions based on historical regional consumption data. It is possible the spatial distribution of the demand will change in the future, which should be represented in the model. The demand profile is also based on ENTSO-E's assumptions and will be changed to our own forecasts in the future.
• Unit maintenance schedule - the model only includes planned and unplanned outages of nuclear units, which could mean that the availability of especially older coal and gas plants is higher than in reality.
• DSR - demand side response is implemented as a generator which is a simplification, could potentially be implemented as a storage unit with variable availability

You are welcome to expand and modify the model according to your needs (honoring the license - see Citiation and license section). You can also submit pull requests onto the Ember-PyPSA repository, but due to a lack of resources we cannot ensure we will be able to review and approve those.

References

Please look at the Ember briefing cited in the Introduction for a detailed list of sources and a description of the methodology, the scenario design assumptions, price parameters etc.

The most significant model references are listed below:

• PyPSA: T. Brown, J. Hörsch, D. Schlachtberger, PyPSA: Python for Power System Analysis, 2018, Journal of Open Research Software, 6(1), arXiv:1707.09913, DOI:10.5334/jors.188
• National Grid Future Energy Scenarios 2022: https://www.nationalgrideso.com/future-energy/future-energy-scenarios
• ENTSO-E Ten-year Network Development Plan (TYNDP) 2022: https://2022.entsos-tyndp-scenarios.eu/download/
• ENTSO-E European Resource Adequacy Assessment (ERAA) 2021: https://www.entsoe.eu/outlooks/eraa/2021/eraa-downloads/index.html
• Climate Change Committee Sixth Carbon Budget (2020): https://www.theccc.org.uk/publication/sixth-carbon-budget/

Release notes

• 1.0.0 - initial release of Ember-PyPSA model with 2030 scenarios described in Ember's briefing on UK bills