Privacy-preserving energy theft detection based on federated learning
File(s)
Author(s)
Hua, Junxi
Kesici, Mert
Pal, Bikash
Type
Conference Paper
Abstract
This paper proposes a secure and privacy-preserving framework for detecting unauthorized energy usage by leveraging consumer energy consumption data in smart grids. Addressing the limitations of traditional centralized schemes, a secure privacy-preserving federated learning framework, named Paillier-Encrypted Federated Learning-based Detection (PE-FLD), is adopted. This architecture consists of a global centre and multiple local detection centers, which interact solely with local consumer data and subsequently communicate aggregated parameters to the global center, thereby safeguarding user privacy. Furthermore, a modified Transformer Neural Network is employed for energy theft monitoring in smart meters. Experimental validation is conducted using real energy consumption data from the Irish Electricity Metering Dataset.
Date Issued
2025-02-01
Date Acceptance
2024-11-01
Citation
IET Conference Proceedings, 2025, 2024 (29), pp.269-274
ISSN
2732-4494
Publisher
The Institution of Engineering and Technology.
Start Page
269
End Page
274
Journal / Book Title
IET Conference Proceedings
Volume
2024
Issue
29
Copyright Statement
© [2025] The Institution of Engineering and Technology. This is the author’s accepted manuscript made available under a CC-BY licence in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy)
License URL
Source
14th Mediterranean Conference on Power Generation Transmission, Distribution and Energy Conversion (MEDPOWER 2024)
Publication Status
Published
Start Date
2024-11-03
Finish Date
2024-11-06
Coverage Spatial
Athens, Greece
Date Publish Online
2025-01-24