WS III “FAIRification put into practise”
The third EERAdata workshop (WS3) is organized jointly with the EERA transversal Joint Programme "Digitalization for energy" (DfE) as part of their annual conference. The workshop is held on 12 and 14 October 2021. Online webinars and hands-on sessions will be focussed on data interoperability and workflows.
Contents
Objectives
- Present activities of the tJP Digitalization for Energy: from FAIR data governance to AI applications and Exascale computing to support advanced research in the field of Energy
- Meet EERA JPs activities in DfE:
- Meet the EU projects that are supporting the tJP DfE activities: EERAdata and EoCoE
Moreover:
- Discuss and develop workflows for large-scale FAIR and open (simulation) data
- Synthesize community insights and formulate action points
- Participate in a writing hackaton: contribute to a community paper!
Read & watch aheads
In order to meet the WS3 objectives and to prepare for the paper writing hackathon, please have a look to:
- Preparatory workshop on the 6th July 2021, agenda and presentations
Agenda - 12 October: 14.00-17.00
| Towards High-Performance Computing (HPC) and the FAIRification of energy data | |||||
|---|---|---|---|---|---|
| Chair: Massimo Celino (ENEA & EERA DfE Deputy Coordinator) | |||||
| 14:00 | Welcome and introduction | Giorgio Graditi | ENEA, Head of Department 'Energy Technologies and Renewable Sources' | ||
| 14:20 | The EU energy policy context | Vincent Berrutto | Head of Unit 'Innovation, clean technologies and competitiveness', DG ENER, EC | ||
| 14:40 | The EERA vision for a Clean Energy Transition | Adel El Gammal | EERA Secretary General | ||
| 15:00 | EERA DfE: Maximizing the impact of digitalisation as a cross-cutting activity within EERA | Rafael Mayo Garcia | CIEMAT & EERA DfE Coordinator | ||
| 15:20 | Break | ||||
| 15.40 | SP1 (EoCoE): Toward exascale for a clean and decarbonized energy | Edouard Audit | CEA & EERA DfE SP1 Coordinator & EoCoE PI | ||
| 16:00 | SP2 (EERAdata): Machine-actionable data as a prerequisite for AI-applications | Valeria Jana Schwanitz | HVL & EERA DfE SP2 Coordinator & EERAdata PI | ||
| 16:20 | OntoCommons Project | Yann Le Franc | CEO and Scientific Director e-Science Data Factory | ||
| 16:40 | Enabling dynamic and intelligent workflows in the future EuroHPC ecosystem | Rosa M Badia | BSC, Workflows and Distributed Computing Group Manager | ||
Agenda - 13 October: 09.00-12.15
Continuation "Towards High-Performance Computing (HPC) and the FAIRification of energy data"
| Perspectives on HPC and data: EERA tJP challenges | |||||
|---|---|---|---|---|---|
| Chair: Rafael Mayo Garcia (CIEMAT & EERA DfE Coordinator) | |||||
| 09:00 | Towards an energy data space - the FEEDS project idea | Johanna Schmidt | Head of Visual Analytics Group VRVis Forschungs-GmbH & FEEDS PI | ||
| 09:15 | FAIR data and FAIR workflows with AiiDA: common workflows for materials properties with 11+ quantum engines | Giovanni Pizzi | EPFL & NCCR Marvel | ||
| 09:30 | Digitalization in nuclear materials research: facts and challenges | Marjorie Bertolus | CEA & DfE SP Nuclear Matter | ||
| 09:45 | The role of digitalization at the development of sustainable hydropower | Eduard Doujak | TU Wien & DfE SP Hydro | ||
| 10:00 | Seeking for new/improved materials for energy transition, impact of HPC and digitalization use | David Lacroix | Lorraine University & DfE SP AMPEA | ||
| 10:15 | Digitalisation challenges for an effective Energy Systems Integration | Peter Breuhaus | Norce Research & DfE SP ESI | ||
| 10:30 | ENTENTE – Defining and implementing a European Database for Multiscale Modelling | Marta Serrano Garcia | CIEMAT, Head of the Division of Materials of Energy Interest | ||
| 10.45 | The EERAdata platform - ideas for tools and services | Astrid Unger | AIT & EERAdata | ||
| 11:00 | Break | ||||
| Perspectives on workflows and interoperable data: EERAdata Use Cases | |||||
| Chair: Valeria Jana Schwanitz (HVL & DfE SP2 Coordinator & EERAdata PI) | |||||
| 11:15 | FAIRification challenges: Evidence from real life implementations | Mehmet Efe Biresselioglu | IUE & EERAdata UC1 "Buildings Efficiency" | ||
| 11:30 | FAIRification put into practice - The code of the FAIR COMETS inventory on energy communities across Europe | August Wierling | HVL & EERAdata UC2 "Energy Markets" | ||
| 11:45 | Experiences with establishing workflows for the material science database | Massimo Celino | ENEA & EERAdata UC3 "Materials for energy" | ||
| 12:00 | Challenges in implementing FAIR principles in energy policy domain | Mariusz Kruczek | GIG & EERAdata UC4 "Energy and energy efficiency policies" | ||
Agenda - 13 October: 13.30-17.30
EERAdata hands-on!
The afternoon is organized in a slot with two breakout groups, followed by two plenaries.
13.30-15.00 Parallel breakout groups (BOG):
BOG "FAIR data challenges in Individual level data: Cases and evidences from H2020 Projects". Organized by UC1 "Buildings efficiency". Format: Presentations, Q&A and discussion. Agenda with speakers and topics to follow.
| FAIR data challenges in Individual level data:
Cases and evidences from H2020 Projects Link to workshop page with notes and results | |||||
|---|---|---|---|---|---|
| Moderator: Prof. Dr. Mehmet Efe Biresselioglu, Izmir University of Economics, Turkey | |||||
| |||||
| Discussant: Dr. Muhittin Hakan Demir, Izmir University of Economics, Turkey | |||||
BOG "FAIRIfication in practice: Discussing solutions for technical challenges - integrating standards, supporting workflows." Organized by UC2 "Power transmission grids, electricity markets". Format: Presentations, Q&A, discussions.
| FAIRIfication in practice: Discussing solutions for technical challenges - integrating standards, supporting workflows. | |||||
|---|---|---|---|---|---|
| Moderator: August Wierling | |||||
Links:
| |||||
| Main outcome: Hands-on knowledge for the FAIRification process | |||||
Break of 30 min
15.30-16.15 Collecting feedback from data users and stakeholders (to prepare a draft to survey user interests) organized by UC 4 "Energy and energy efficiency policies" (GIG team)
| FAIR data challenges in energy and energy efficiency policies | |||||
|---|---|---|---|---|---|
| Moderators: Małgorzata Markowska, Aleksandra Zgórska, GIG, Katowice | |||||
|
Discussion questions:
| |||||
| Main outcome: Recommendations for the community survey | |||||
Break of 15 min
16.30-17.30 Conclusions for the EERAdata platform: Which FAIR data services and tools to offer? organized by WP3 (AIT team)
| EERAdata platform implementation strategy | |||||
|---|---|---|---|---|---|
| Moderators: Manfred Paier, Michael Barber, Astrid Unger (AIT, Vienna) | |||||
|
Discussion points:
| |||||
| Main outcome: Action plan for preparing and implementing Workshop 4 | |||||
Agenda - 14 October: 09.00-12.30
EERAdata hands-on!
| Title | Moderator | |
|---|---|---|
| 09:00 | Discussion across use cases: How to advance the state of FAIR in the use cases? What are the plans for uploading use cases demos to the European Open Science Cloud? | Use case leaders |
| 11:30 | Plenary to wrap up discussions. Organisation of feedback to the Digitalisation of Energy Action Plan OPC (DG Ener) (Foreseen closing date is 24/01/2022). |
Preparatory workshop
In order to meet the WS3 objectives and to prepare for the paper writing hackathon, please have a look to the preparatory workshop. The preparatory workshop was needed to discuss about:
- best practices in FAIRfication processes
- state of the art and needs for data management and workflows
- briefing participants for the upcoming workshop No. 3 and the planned community paper
Preparatory workshop was on July 6th, 2021. It is part of a series of workshops dedicated to the FAIRification and opening of energy research data in and with the low carbon energy research experts and communities at large. The preparatory workshop takes the form of an online discussion with invited talks and a panel discussion.
Agenda
| 6th July, 10-12 CET | Agenda |
|---|---|
| 10:00 - 10:05 | Welcome and introduction, Massimo Celino (ENEA) |
| 10:05 - 10:20 | FAIR data collective, Nikola Vasiljevic (DTU Wind Energy, GOFAIR Facilitator Team) |
| 10:20-10:35 | FAIR data resources - examples from the life sciences, Susanna-Assunta Sansone (University of Oxford) |
| 10:35-10:50 | Data management in the BIG-MAP project and its relation with the BATTERY 2030+ community, Ivano Castelli (DTU Energy) |
| 10:50-11:00 | Break |
| 11:00-11:15 | From energy community principles to the EERAdata platform, Manfred Paier (AIT) |
| 11:15-12:00 | Panel discussion with Q&A |
Summary
The objective of the online discussion was to prepare the 3rd EERAdata community workshop. In this workshop participants explored use cases to:
- manage the implementation of FAIR and open principles,
- identify and test workflows for the FAIRification and opening of databases,
- collaborate on a community paper “AI & data-driven science for the energy transition”.
Gathering 40 participants, this EERAdata online discussion focused on the state of the art and needs for data management and workflows through best-practice lectures about the organisation of the FAIRification processes. The workshop was part of the the third EERAdata workshop which continue in October.
Three external speakers showcased their best practices on FAIR workflows:
- Nikola Vasiljevic (DTU Wind Energy) presented approaches from the initiative GOFAIR.
- Susanna-Assunta Sansone (University of Oxford) gave insights into data processing by the initiative FAIRsharing.
- Ivano Castelli (DTU Energy) discussed FAIR data workflows applied in material science. Lastly Manfred Paier (AIT) introduced participants to the concept of the EERAdata platform.
Dr. Nikola Vasiljevic highlighted the FAIR data collective, an informal group collaborating on the FAIRification lifecycle. He stressed that it is pivotal to allow non-web-semantic experts to execute the FAIRification lifecycle, for which they develop controlled vocabularies, metadata templates and open sources tools, such as the CEDAR workbench.
In her lecture, Prof. Dr. Susanna-Assunta Sansone went back to the roots, clarifying that FAIR principles mean “good data management under common standards”. The general FAIRification process goes through the cycle of ‘define’, ‘implement’, and ‘embed & sustain’. She introduced the initiatves ELIXIR and FAIRsharing as well as the support tool FAIR cookbook. ELIXIR is a platform for FAIR life science data in Europe. FAIRsharing is an educational resource on data and metadata standards, targeting all disciplines as well as funders, publishers, and other stakeholders. The FAIR cookbook supports the FAIR workflows with recipes and applied examples for each of the FAIR principles.
Dr. Ivano Castelli introduced the European project BIG-MAP, which focusses on improving battery technology and is part of the Battery2030+ initiative. The project gathers not only numerical data but also documents, films, and photographs. The data is collected in an app store ready for end-use processing. The data sharing process has been tested in the Battery2030+ and BIG-MAP projects and will be crucial for the StoRIES project starting in the second half of 2021.
Finally, Manfred Paier presented the concept for the EERAdata community platform. He introduced the idea of FAIR digital objects and FAIR Implementation Profiles as EERAdata’s choice for supporting FAIR workflows and standards.
Lectures were wrapped up by a panel debate. The take-home message is distilled in a conclusion by Susanna-Assunta Sansone: “It is necessary to jump onto the wagon and try out for the domain (…) Use what exists. There is no way around trying out”. In this spirit, the EERAdata project will continue its work to steer the FAIR train for the low-carbon energy domain. The next possibility for researchers and policymakers to join the efforts of the EERAdata project is the online workshop organized together with the annual conference of EERA’s Joint Program Digitalization for Energy. The event will take place between 11 and 15 October 2021.
WS3UC1
Case Study 1: PEAKapp Project (Collecting 15-min load profiles of households in Austria)
- Dr. Johannes Reichl, Energieinstitut an der Johannes Kepler Universität Linz, Austria
Dr. Johannes Reichl from the Energy Institute Johannes Kepler University Linz presented "Collecting 15-min load profiles of households in Austria" in case study 1. The focus of the presentation is to examine new ways of collecting data in residential buildings with the emergence of ICT in the energy system, focusing on smart metering systems. From this point of view, this case study describes the data gathering process in the EU funded research project PEAKapp (https://www.peakapp.eu/), which aimed the improvement of a smartphone app for allowing households to save electricity and to increase their consumption of clean and moderate-priced electricity in times of high renewable electricity production. The test of the app applied between 2016 - 2019 in four European countries, Austria, Latvia, Sweden, and Estonia was motivated to review their energy consumption via the app, and therefore to implement energy savings due to energy-saving measures. In PEEKapp, the households can observe how much electricity they consumed and at which time of the day their consumption is increasing. In addition, there is an additional benefit of the app in terms of a more conscious point of view about consumption. Moreover, the app alerts users to times of discounted electricity prices through push messages. The users can learn whether they consume renewable based electricity or fossil fuel based electricity. When users participated, the app collects every day consumption data as well as how households use the smart phone app. Electricity profiles of users are regarded as “sensitive data”. PEEKapp provides opportunity to observe the everyday practices and electricity consumption patterns of households from morning until the night. Therefore, such sensitive data should be safeguarded. To obtain such data, the collaboration with electricity provider is necessary. However, one of the challenges is that energy providers don’t usually know much about their customers’ profile (They know about the address of the households, their names, etc. However, they do not know about how many people live in a house, for example). The data obtained in PEEKapp provides all this socio-demographic data such as education. Peculiarities of PEEKapp Data are listed such as sensitive data, necessity to safeguard this sensitive data, full openness (Zenodo platform), providing the socio-demographic data that the electricity providers do not have, high statistical capacity and opportunity to observe daily consumption patterns. In the contrary, the app has some challenges; hard to convince the users to participate in this app, make the users contribute to the app (in terms of sharing their data), mindset of different cultures/countries, different prototypes, you may persuade the top-management; however, the operational process is not controlled by this top-management, different levers of people with different mindsets in utility companies, lack of interest from policy makers or industry for raw data; instead, mostly researchers are interested in (particularly for smart meter data which is huge and complex), the inability of electricity market to react fast, the approach of utility companies is more different than the approach of researchers. In terms of FAIRification, it is not really sufficient to make the data findable or accessible. To make it usable, we also need some intervention from the researcher’s side to make sure that it will be correctly utilized.
Case Study 2: Multinational Survey among 18,000 households in Europe from ECHOES Project
- Dr. Andrea Kollmann, Energieinstitut an der Johannes Kepler Universität Linz, Austria
Dr. Andrea Kollmann from the Energy Institute Johannes Kepler University Linz presented " Multinational Survey among 18,000 households in Europe from ECHOES Project" in case study 2. The major goals of the ECHOES project are to understand factors that influence individual and collective energy-related choices and behaviors, and illuminate differences between countries in energy choices and attitudes. For this aim, a large-scale survey, which was managed in 31 European countries (EU-26, UK, Turkey, Norway and Switzerland). This survey was presented to approximately 600 respondents per country (less in the Island countries of Malta and Cyprus) in their native language, which led to a total sample of 18,037 completed surveys. The survey took 20 minutes per response and representative sampling quotas in age, gender, income dimensions. When the features and advantages of the ECHOES Survey are examined, it has a large data set and did not produce sensitive data; The data is published in both ECHOES webpage and Zenodo platform; data is available in different formats (excel file, SPSS file format for econometric analysis); Working with a survey company (An international marketing survey company. Their experience to frame the questions and disseminate the survey contribute to the purposes of the project); Fairly easy process to work with a survey company and a more consolidated survey structure; Data security (Since the survey company takes the responsibility, even the researchers do not know about the respondents' profiles); The respondents are all volunteers; Refund for the participation; A good dataset as a teaching material. The difficulties encountered in the ECHOES project are summarized as follows; very few people are aware of the dataset; Not updated; Lack of dissemination of the data; No keywords provided for the ECHOES dataset on Zenodo; Outside of the original consortium, no other academic researchers are publishing with this dataset (A concern for reusability); Although the data is available on Zenodo platform and official web page of the project, the data is still not much utilized by different user groups (Visibility problem); It is hard for researchers to work on the data that is already publicly available for a long time. (Publicly available and accessible, but the risk of repetition in similar studies); Inability to integrate qualitative data into the dataset (Lack of original ideas for how to integrate such qualitative data, and issues with GDPR – because the more specific the data is, it tells more about the person - the risk of identifying the person.) and lastly the lack of visualization.
As a result of expert opinions from the Energy Institute Johannes Kepler University Linz, the following can be listed as general achievements:
- Data protection (safeguard) of data is needed for sensitive data
- Open access data - Zenodo platform and official web page of the project make the dataset accessible
- Observation of energy use patterns
- Available data in different formats (excel file, SPSS file format for econometric analysis)
When we consider what is mainly challenging are:
- Hard to find users to participate in the PEAKapp
- Different cultures, dynamics
- Lack of interest from policymakers (weak awareness to their customers)
- Lack of keywords and visualization
- Lack of interest to publish any academic study with the available ECHOES data (a visibility concern)
A number of recommendations can also be listed:
- Accurate and updated data is needed
- Visualization of data
- Land to find adequate data
