Do you sometimes wonder who else uses your favourite open-source energy modelling tools, and what they do with it? Come join us at the PyPSA User Meeting!
Dear PyPSA community,
We are excited to announce the second PyPSA User Meeting on May 27, 2024: 3 pm - 6 pm (CEST)!
Youtube recording
What to do?
- Register here for the online event: Zoom Registration (Link)
- To stay up to date and to find the recordings, please revisit this thread on openmod (Link)
- Propose your lightning talk: Closed.
Updated Agenda
| Time | Youtube | Presenter | Topic |
|---|---|---|---|
| 15:00 | 00:00:00 | ENSYS TU Berlin |
PyPSA group welcomes all participants |
| 15:05 | 00:05:36 | Tom Adams Haast energy |
Modelling security constrained dispatch in the Australian Grid |
| 15:18 | 00:20:25 | Malcolm Moss Virginia Tech |
Open ERCOT: Using PyPSA to Understand Generation Stack Modeling |
| 15:31 | 00:32:01 | Tapio Schmidt-Achert FfE |
Modelling of Global Levelized Cost of Hydrogen under the Open-Source Modelling Environment PyPSA |
| 15:44 | 00:43:38 | Andrew Lyden University of Edinburgh |
PyPSA-GB: Open-source future Great Britain model applied to security of supply and locational pricing |
| 15:57 | 00:58:06 | Agnès François FEMTO-ST Institute |
Integrating hydrogen in non-interconnected islands |
| 16:10 | 01:10:45 | Peter Klein Meridian Economics |
Independent IRP modelling for South Africa |
| 16:30-16:40 | Break | ||
| 16:40 | 01:26:22 | Martha Frysztacki Open Energy Transition (OET) |
Enhancing Energy System Analysis: OET’s Role in Supporting PyPSA Workflows and Development |
| 16:53 | 01:40:03 | Ganesh Doluweera Canada Energy Regulator |
PyPSA-Can: The Electricity Supply Model of Canada’s Long-term Energy Projection System |
| 17:06 | 01:56:05 | Carlos Fernandez Université de Liège |
The PyPSA-BO case: How we used PyPSA-Earth to create a curated version of the Bolivian electrical system |
| 17:19 | 02:11:35 | Rowan Tunnicliffe Mutual Energy |
Guiding a Northern Irish gas TSO through the energy transition using PyPSA |
| 17:32 | 02:26:24 | Tom Welfonder Universität Stuttgart |
Stochastic Unit Commitment on Multiple Power Markets using PyPSA |
| 17:45 | 02:40:52 | Shuwei (Shawn) Zhang Draworld (Beijing) Center |
Applying PyPSA to understand heterogeneity and promote transparency in highly-regulated systems: a case for China with 33 de facto balance areas |
| 17:58 | 02:53:40 | Adrian Odenweller Potsdam Institute for Climate Impact Research |
REMIND-PyPSA-Eur: Coupling an integrated assessment model to an hourly power system model |
| 18:05-18:10 | 03:09:31 | Conclusion | |
| 18:10-18:40 | Breakout sessions (informal) |
Abstracts
1. Tom Adams (Haast energy): Modelling security constrained dispatch in the Australian Grid
We have been using PyPSA’s dispatch optimisation combined with an adaptation of n-1 line-overload constraints to model generation dispatch in the Australian National Electricity Market (NEM). This talk will give a high-level demonstration of how PyPSA can be used to forecast binding constraints, and how these may be affected by line and plant outages, as well as network augmentations.
2. Malcolm Moss (Virginia Tech): Open ERCOT: Using PyPSA to Understand Generation Stack Modeling
In this talk we present Open ERCOT, a basic model of the ERCOT generation stack using historical data and a simplified 8 bus system. Our aim was to create a simple model using open-source data and PyPSA to investigate grid dispatch. Using our model we looked into the impact of battery energy storage solutions and how they change grid dispatch. We also aim to discuss the use of PyPSA as a tool for understanding the grid and how it operates. In addition we aim to present how PyPSA can be used to help teach electricity markets concepts through hands-on experience. Through our analysis we aim to show some of the potential impacts of battery energy storage solutions as well as some potential future work using our model.
3. Tapio Schmidt-Achert (FfE): Modelling of Global Levelized Cost of Hydrogen under the Open-Source Modelling Environment PyPSA
Hydrogen is central to the global energy transition when produced at low emissions. This paper introduces a renewable hydrogen production system model (HPSM) that optimizes a hybrid hydrogen production system (HPS) on a worldwide 50 km × 50 km grid, considering country-specific interest rates. Besides the renewable energy’s impact on the HPS design, we analyze the effect of country-specific interest rates on the levelized cost of hydrogen (LCOH) production. LCOH production ranges between 2.7 AC/kg and 28.4 AC/kg, with an average of 9.1 AC/kg. Over one third (40.0%) of all cells have an installed PV capacity share between 50% and 70%, and 76.4% have a hybrid configuration. Hybrid HPSs can significantly reduce the LCOH production compared to non-hybrid designs, whereas country-specific interest rates lead to significant increases in the LCOH production. Hydrogen storage is deployed rather than battery storage to balance production and demand.
4. Andrew Lyden (University of Edinburgh): PyPSA-GB: Open-source future Great Britain model applied to security of supply and locational pricing
PyPSA-GB is an open-source energy model and dataset designed for simulating Great Britain using the well-established National Grid’s Future Energy Scenarios. This talk will present research studies applying PyPSA-GB to investigate research on security of supply in Scotland and the impact of locational pricing on heat pump deployment.
5. Agnès François (FEMTO-ST Institute): Integrating hydrogen in non-interconnected islands
The HyLES project aims to study the integration of hydrogen in non-interconnected islands. In this project, my work focuses on the study of the electricity grid of a French island in the Indian Ocean: Reunion Island. The aim of this work is to assess whether, in the medium term, the local electricity resource will be sufficient to meet the island’s inhabitants needs. Indeed, the studied island is aiming for energy autonomy in the coming years. Namely, 100% local supply for electricity and mobility, whereas the island is currently 86% dependent on fossil imports.
The first aim is to assess the island’s electrical autonomy: can local (renewable) production meet the inhabitants’ consumption needs? What size of installations would be needed? How should the electricity mix be broken down? What additional storage requirements will be needed with the development of intermittent energies? PyPSA enabled us to model our network and answer all these questions. Then, the issue of mobility was addressed, and hydrogen was introduced. Hydrogen as a bus or train fuel, hydrogen for the manufacture of e-kerosene or e-ammonia, etc. Once all parameters had been defined, the placement, sizing and energy management of the hydrogen technologies were studied, again using PyPSA. Minimum annual use, limit on installed storage, constraint on the location of installations, second objective function, etc. Although PyPSA was used as the basis for our modelling, its adaptability enabled us to add various constraints to our problem, in order to match reality as closely as possible.
6. Peter Klein (Meridian Economics): Independent IRP modelling for South Africa
The South African government recently released a draft Integrated Resource Plan for public comment. This presentation covers independent modelling done by Meridian Economics, using the PyPSA-RSA model (forked from PyPSA-ZA) as part of our technical submission.
7. Martha Frysztacki (Open Energy Transition - OET): Enhancing Energy System Analysis: OET’s Role in Supporting PyPSA Workflows and Development
The PyPSA ecosystem provides powerful tools to address various questions related to energy systems. However, due to the high complexity of the topic, many users need support in managing workflows, defining constraints or integrating data. In this talk, we demonstrate how OET assists PyPSA users and provide an overview of current projects. Specifically, we show how OET helps TransnetBW, a German transmission system operator, to synchronize its customized PyPSA-Eur workflow with the original upstream software. We present the main strategies discussed to accomplish this task, as well as general technical aspects that need to be considered to ensure sustainable compatibility between software developments. Additionally, we discuss OET’s recent collaborative work, “Flattening the Energy Curve,” which evaluates energy efficiency measures in the European energy system. Using PyPSA-Eur, we explore how renovation, demand-side management, and efficiency improvements impact overall energy prices and system robustness.
8. Ganesh Doluweera (Canada Energy Regulator): PyPSA-Can: The Electricity Supply Model of Canada’s Long-term Energy Projection System
A core mission of the Canada Energy Regulator (CER) is to provide timely and relevant energy information and analysis. Canada’s Energy Future series (EF series), which provides long-term energy supply and demand projections, is the CER’s flagship energy information publication. The EF series is used by a broader spectrum of decision-makers, including those from governmental bodies and industry. The EF series relies on a suite of energy system models, each meticulously tailored to examine various facets of Canada’s energy landscape. The latest iteration of the series, Canada’s Energy Future 2023: Energy Supply and Demand Projections to 2050 (EF2023), is the CER’s first long-term outlook that modeled scenarios where Canada reaches net-zero greenhouse gas (GHG) emissions by 2050. In order to generate these net-zero energy projections, the CER undertook a comprehensive overhaul of its energy system modeling suite. One significant enhancement involves the complete redesign and implementation of the electricity supply model using PyPSA.
The redesigned model, PyPSA-Can, interacts seamlessly with other models within the EF Series suite, including a detailed energy demand projection model, a macro-economic analysis model, and several fuel supply models, enabling a comprehensive understanding of Canada’s energy landscape in a net-zero future. The insights derived from EF2023, particularly those pertaining to electricity, have been met with acclaim and are already influencing policy design and analysis. The presentation will describe the model setup, highlight key findings, and discuss the modelling lessons learned.
9. Carlos Fernandez (Université de Liège): The PyPSA-BO case: How we used PyPSA-Earth to create a curated version of the Bolivian electrical system
This presentation provides the context on how PyPSA-Earth, a global range energy system model, works and how it was used to create a customized version of the model for the case of Bolivia (PyPSA-BO). Additionally some of the results from implementing the model and its adaptation are shown, as well as future work and contribution opportunities.
10. Rowan Tunnicliffe (Mutual Energy): Guiding a Northern Irish gas TSO through the energy transition using PyPSA
Mutual Energy owns and operates natural gas transmission assets in Northern Ireland (NI), as well as the 500MW HVDC electricity link between Northern Ireland and Scotland – the Moyle Interconnector. As a mutual company, these assets are underwritten by consumers in NI, and the company exists to manage assets in the long-term interests of NI consumers.
With the energy transition picking up pace, and significant open policy questions about how we transition away from fossil fuels, Mutual Energy needs to understand the impact of different pathways to Net Zero to ensure that the potential value of our assets is realised and consumer interests are protected. As a transmission system operator (TSO) ensuring ongoing security of supply is also a key concern. These emerging problems required a comprehensive modelling solution, including a power dispatch and expansion model capability. After assessing different options, the company landed on PyPSA as a solution. This talk will explain how Mutual Energy are currently using PyPSA, as well as our plans to utilise it in the future.
11. Tom Welfonder (Universität Stuttgart): Stochastic Unit Commitment on Multiple Power Markets using PyPSA
In this talk, we will present a stochastic unit commitment (UC) optimization application for PyPSA which we have developed and tested on an example use case in which UC optimization is done for a waste-to-energy plant with heat storage and a battery energy storage system (BESS) in Germany. The power market mechanisms underlying day-ahead (DA) and balancing power (aFRR) markets are implemented. Uncertainty in the market prices as well as heat load uncertainty is considered. The tool consists of multiple modular extensions for the Python for Power System Analysis (PyPSA) framework, namely the implementation of market and bidding mechanisms, stochastic optimization and multistaging.
12. Shawn Zhang (Draworld (Beijing) Center): Applying PyPSA to understand heterogeneity and promote transparency in highly-regulated systems: a case for China with 33 de facto balance areas
Our series of completed and ongoing applied research employs Python for Power System Analysis (PyPSA) to model China’s electricity sector across 33 de-facto physical balance areas, focusing on regional heterogeneity and regulatory nuances. We aim to closely mirror real operations rather than achieving theoretical optimality, enhancing the understanding of the regulated markets and aiding policymakers and stakeholders with tools to navigate spatial and temporal heterogeneity. The presentation will address, using our Google Colab-based application, key issues such as:
- China-specific political constraints: How do extensive planning timelines and uniformity in coal units shape energy policy (e.g. ToU design, pricing resolution) and beyond?
- Coal inflexibility and the “counter-factual”: How does the rigidity of coal operations, affected by design and economic disincentives, affect renewable integration and system reliability?
- Cross-regional transmission patterns: To what degree should our modelling reflect the realities of China’s cross-regional electricity transmission?
- Charging-discharge asymmetry for EVs: Is it self-defeating or self-fulfilling in terms of “smart charge” and “sufficient V2G”?
- Mid-year capacity pro-rating: How does the alignment of renewable profiles, annual hours, and end-year capacities cope with the substantial GW capacity involved?
13. Adrian Odenweller (Potsdam Institute for Climate Impact Research): REMIND-PyPSA-Eur: Coupling an integrated assessment model to an hourly power system model
Low-cost renewable electricity will become a key enabler of the global energy transition. This requires a flexible power system that is closely interlinked with other end-use sectors. However, integrated assessment models (IAMs) – the main tool to inform global climate policymaking – struggle to properly represent hourly power system effects as they lack the necessary spatio-temporal detail. Vice versa, while power system models (PSMs) represent these crucial effects, they typically require exogenous techno-economic parameters for a given target year. Here, we leverage these synergies by a bi-directional price-based coupling of the IAM REMIND with the PSM PyPSA-Eur. In the coupling, REMIND takes care of realistic transformation pathways, incorporating detailed energy service demands from all end-use sectors. Based on total power demand, technology costs and minimum capacities, PyPSA-Eur is then executed for each 5-year time step of REMIND. The market values and capacity factors from each PyPSA-Eur run then again inform the intertemporal optimisation in REMIND. We demonstrate the feasibility and internal consistency of this iterative coupling by obtaining a high degree of convergence in terms of investment decisions in both models. We present preliminary results and highlight promising avenues for further research such as expanding the coupling approach to other world regions.
Additional information regarding the format of the user meeting:
To all the speakers
- Please prepare to hold a presentation of 8 minutes,
- Followed by a short Q&A of 3 minutes.
To all listeners
- Please feel free to participate in the talks by posting your questions into the Zoom chat.
- We will collect the questions and discuss them in the consecutive Q&As or in the breakout rooms after the lightning talks.
Recording permissions
We’d like to record the entire session and, for people who give their consent, make the recordings available after the session under a Creative Commons Attribution (CC BY 4.0) to those who were not able to make the meeting.
Participants can license their contribution to the meeting under a Creative Commons CC BY 4.0 license at the time of registration. This will permit the associated video recording to be shared with a wider audience in the Open Modelling community.
In addition, presenters can optionally license their presentations under a Creative Commons CC BY 4.0 license. This will also facilitate their dissemination and reuse.
Attendees and presenters should note that open licensing is optional. We will not publish anything without the consent of those being recorded. You may withdraw your consent afterwards as well. We will respect the wishes of anyone who asks to delete the recording of them during times when they were talking.
Inquiries
For any inquiries, please contact Bobby Xiong @bobbyxng // xiong@tu-berlin.de and Caspar SchauĂź @cpschauss // caspar.schauss@gmail.com
We look forward to a fruitful exchange with many of you!
Best regards,
Caspar, Bobby and the PyPSA team at TU Berlin
https://www.tu.berlin/ensys
https://tub-ensys.github.io/
https://pypsa.org/