Neural NILM Learning Paradigms: From Centralised to Decentralised Learning

Centralised vs collaborative learning

Non-intrusive Load Monitoring (NILM) has become a paramount in both industrial and residential sectors to achieve efficient energy consumption. Deep neural networks have been gaining the highest interest from the research community, commonly referred to as neural NILM. In most cases, neural NILM models follow a centralised based learning scheme, where the energy data is assumed to be available in a central node for training. This practice can, however, raise privacy and security concerns from the consumer’s side since energy data can reveal in-home activities and occupancy records if intercepted. In response, Federated Learning (FL) has been suggested as a viable solution to address these issues. In the paper "Neural NILM Learning Paradigms: From Centralised to Decentralised Learning", an overview of neural NILM models following both a centralised and a federated learning paradigm was presented while also identifying the main challenges with regard to both learning paradigms and potential future research directions for more robust, secure and privacy-preserving models in the neural NILM industry. Overall, as any other new technology, FL has its merits and limitations. Typically, FL provides promising perspectives to solve the privacy issues of energy disaggregation. However, it also opens doors for new challenges, especially those related to the (i) low disaggregation performance of FL-based NILM algorithms, (ii) susceptibility to noise, (iii) lack of labeled sub-metered data at the customer’s level, and (iv) need to adopt robust security mechanisms.

Further information can be found in the paper:

Hafsa Bousbiat, Christoph Klemenjak, Yassine Himeur, Wilfried Elmenreich, Abbes Amira, Wathiq Mansoor, and Shadi Atalla. Neural NILM learning paradigms: From centralised to decentralised learning. In Proceedings of the 2022 5th International Conference on Signal Processing and Information Security (ICSPIS), pages 138–142, December 2022. (doi:10.1109/icspis57063.2022.10002485)

The paper also won the best paper award at the 5th International Conference on Signal Processing and Information Security (ICSPIS) in December 2012.

The Role of Renewable Energies in the Arctic

Last week, Prof. David Finger from Sustainability Institute and Forum at Reykjavik University visited the University of Klagenfurt as a guest researcher. His inspiring talk at Energy Cluster Meeting XXXI, titled "Climate-Neutral Europe: the Role of Renewable Energies in the Arctic to decarbonize Europe and enhance energy independence", was truly captivating and gave us all a glimpse into the possibilities of an Austrian-Icelandic Energy Cooperation. 

Students and researchers alike were amazed by the future of renewable energy in Europe that Prof. Finger's talk highlighted and left the room with interesting insights into the role of renewable energy in the Arctic. It was a great opportunity to learn more about the progress of climate neutrality and energy independence in Europe and we look forward to further collaborations with Prof. Finger in the future.

Energy Informatics 2023 in Vienna -- Call for Papers

The EU aims to be climate-neutral by 2050 – an economy with net-zero greenhouse gas emissions. This objective is at the heart of the European Green Deal and in line with the EU’s commitment to global climate action under the Paris Agreement. The transition to a climate-neutral society is both an urgent challenge and an opportunity to build a better future for all. Energy informatics support in solving many challenges of the energy transition, by providing solutions for intelligent management and operation of energy systems and their assets.

The objective of the DACH+ conference series on Energy Informatics is to promote research, development, and implementation of information and communication technologies in the energy domain and to foster the exchange between academia, industry, and service providers in the German-Austrian-Swiss region and its neighbouring countries (DACH+).

We seek high-quality original contributions addressing the design, adoption, operation and management of smart energy systems, the integration of intermittent renewable generation and energy efficiency gains through ICT, market approaches and mechanisms for ICT-enabled energy systems, and research on associated (decentralised) data-driven decisions. We welcome theoretical contributions as well as publications addressing system design, implementation, and experimentation. The list of topics of interest to the conference includes, but is not limited to:

• ICT for future energy systems, sector coupling and the integration of intermittent renewable generation
• Information and decision support systems for future energy markets and mechanisms
• Energy system modelling and (open) energy system data
• Protocols and architectures for IT systems in the energy sector
• Data analytics and machine learning for smart energy systems and decentralised decision-making, as well as platforms for data analysis
• Open data and software for energy research
• Management of distributed generation and demand side management
• ICT for (multi-) energy networks and micro-grids
• Energy-efficient mobility, charging management for electric vehicles, energy-aware traffic control, and smart grid integration of mobile storage
• Smart buildings, digital metering, occupant comfort, and user interaction
• Adoption of ICT in the energy sector
• Cross-cutting issues including cyber security and privacy protection, interoperability, verification of networked smart grid systems

Posters, Demos and Workshops

Submissions for posters, demos, and workshop suggestions are welcome, too. The topics of interest are the same as indicated above. Posters and demos require the submission of an extended abstract, which will be peer-reviewed. If accepted, the abstract will appear in the conference proceedings. Further details can be found on the conference website.

Submission and Publication

Submitted papers will be reviewed in a double-blind process. Accepted and presented papers will be published in the Springer Open Journal Energy Informatics (https://energyinformatics.springeropen.com). The conference language is English, and papers must be written in English. We solicit full research papers (max. 18 pages of content plus 2 additional pages for references) as well as short papers (max. 10 pages of content plus 2 additional pages for references). Templates and instructions will be made available at http://www.energy-informatics.eu/. Further information on the submission of posters and demos is also available on the website. The Open Access fee for the journal article is included in the registration fee.

Important dates

Apr 09, 2023:                     Submission of papers

May 16, 2023:                     Decision acceptance (assignment of shepherds) / rejection

May – Aug                          Incremental revision process between author and shepherd

Jul 02, 2023:                        Camera-ready deadline for poster abstracts for accepted contributions

Oct 04 2023:                      14th Doctoral Workshop Energy Informatics

Oct 05-06, 2023:                12th DACH+ Conference on Energy Informatics