Paper published in a book (Scientific congresses, symposiums and conference proceedings)
Towards a peer-to-peer residential short-term load forecasting with federated learning
DELGADO FERNANDEZ, Joaquin; POTENCIANO MENCI, Sergio; PAVIĆ, Ivan
2023In Proceedings of the 2023 IEEE Belgrade PowerTech
Peer reviewed
 

Files


Full Text
Towards_a_peer-to-peer_residential_short-term_load_forecasting_with_federated_learning.pdf
Publisher postprint (782.61 kB)
Request a copy

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Federated Learning; Peer-to-Peer; Clustering; K-means; Agent-Based Modelling; Short-Term Load Forecasting
Abstract :
[en] The inclusion of intermittent and renewable energy sources has increased the importance of demand forecasting in the power systems. Smart meters play a critical role in modern load forecasting due to the high granularity of the measurement data. Federated Learning can enable accurate residential load forecasting in a distributed manner. In this regard, to compensate for the variability of households, clustering them in groups with similar patterns can lead to more accurate forecasts. Usually, clustering requires a central server that has access to the entire dataset, which collides with the decentralized nature of federated learning. In order to complement federated learning, this study proposes a decentralized Peer-to-Peer strategy that employs agent-based modeling. We evaluate it in comparison to a typical centralized k-means clustering. To create clusters, we compare Euclidian and Dynamic time warping distances. We employ these clusters to build short-term load forecasting models using federated learning. Our results reveal the possibility of using Peer-to-Peer clustering along with simple Euclidean distances and Federated Learning to obtain highly performant load forecasting models in a fully decentralized setting.
Research center :
Interdisciplinary Centre for Security, Reliability and Trust (SnT) > FINATRAX - Digital Financial Services and Cross-organizational Digital Transformations
ULHPC - University of Luxembourg: High Performance Computing
Disciplines :
Computer science
Electrical & electronics engineering
Management information systems
Author, co-author :
DELGADO FERNANDEZ, Joaquin  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > FINATRAX
POTENCIANO MENCI, Sergio  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > FINATRAX
PAVIĆ, Ivan  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > FINATRAX
External co-authors :
no
Language :
English
Title :
Towards a peer-to-peer residential short-term load forecasting with federated learning
Publication date :
09 August 2023
Event name :
2023 IEEE Belgrade PowerTech
Event organizer :
IEEE
Event place :
Belgrade, Serbia
Event date :
25-29 June 2023
Main work title :
Proceedings of the 2023 IEEE Belgrade PowerTech
Publisher :
IEEE
ISBN/EAN :
978-1-6654-8778-8
Pages :
6
Peer reviewed :
Peer reviewed
Focus Area :
Security, Reliability and Trust
European Projects :
H2020 - 814654 - MDOT - Medical Device Obligations Taskforce
FnR Project :
FNR13342933 - Paypal-fnr Pearl Chair In Digital Financial Services, 2019 (01/01/2020-31/12/2024) - Gilbert Fridgen
Name of the research project :
Medical Device Obligations Taskforce
Funders :
CE - Commission Européenne
FNR - Fonds National de la Recherche
Union Européenne
Available on ORBilu :
since 14 August 2023

Statistics


Number of views
129 (29 by Unilu)
Number of downloads
9 (9 by Unilu)

OpenAlex citations
 
0

publications
0
supporting
0
mentioning
0
contrasting
0
Smart Citations
0
0
0
0
Citing PublicationsSupportingMentioningContrasting
View Citations

See how this article has been cited at scite.ai

scite shows how a scientific paper has been cited by providing the context of the citation, a classification describing whether it supports, mentions, or contrasts the cited claim, and a label indicating in which section the citation was made.

Bibliography


Similar publications



Sorry the service is unavailable at the moment. Please try again later.
Contact ORBilu