EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Asynchronous adaptive federated learning for distributed load forecasting with smart meter data

Mohammad Navid Fekri, Katarina Grolinger, Syed Mir

International Journal of Electrical Power & Energy Systems · 2023-11 · 153 · 109285

DOI: 10.1016/j.ijepes.2023.109285 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Electrical Power & Energy Systems, 2023-11. Verfasst von Mohammad Navid Fekri und 2 weiteren Autor:innen. Die Arbeit wurde bisher 42-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Alliance de recherche numérique du Canada, Natural Sciences and Engineering Research Council of Canada.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Load forecasting is essential for the operation and planning of a utility company. Recent large-scale smart meter deployments enabled the collection of fine-grained load data and created opportunities for sensor-based load forecasting. Machine learning has achieved great successes in load forecasting; however, conventional machine learning techniques require data transfer to the cloud or another centralized location for model training. This not only exposes data to privacy and security risks but also, with a large number of smart meters, increases network traffic. Federated Learning (FL) has a potential to alleviate mentioned concerns by training a single ML model in a distributed manner without requiring participants to share their data. Consequently, this paper proposes FedNorm, a novel asynchronous FL approach for load forecasting with smart meter data. While most FL strategies are synchronous and require all clients to complete local training in each round of aggregation, FedNorm is asynchronous and aggregates updates without waiting for lagging clients. To achieve this, FedNorm measures the clients contributions considering similarities of local and global models as well as the loss function magnitudes. The experiments demonstrate that FedNorm achieves higher accuracy than seven state-of-the-art FL techniques. Moreover, experiments show that FedNorm converges in fewer communication rounds compared to other FL models.

Förderung

  • Alliance de recherche numérique du Canada
  • Natural Sciences and Engineering Research Council of Canada

Zitieren

APA

Fekri, M. N., Grolinger, K., & Mir, S. (2023). Asynchronous adaptive federated learning for distributed load forecasting with smart meter data. International Journal of Electrical Power & Energy Systems, 153, 109285. https://doi.org/10.1016/j.ijepes.2023.109285

Vancouver

Fekri MN, Grolinger K, Mir S. Asynchronous adaptive federated learning for distributed load forecasting with smart meter data. International Journal of Electrical Power & Energy Systems. 2023;153:109285. doi:10.1016/j.ijepes.2023.109285

BibTeX
@article{Fekri2023,
  author = {Fekri, Mohammad Navid and Grolinger, Katarina and Mir, Syed},
  title = {Asynchronous adaptive federated learning for distributed load forecasting with smart meter data},
  journal = {International Journal of Electrical Power & Energy Systems},
  year = {2023},
  volume = {153},
  pages = {109285},
  doi = {10.1016/j.ijepes.2023.109285}
}
Kostenlos starten →