An active early warning method for abnormal electricity load consumption based on data multi-dimensional feature

Jia Cui, Tianhe Fu, Junyou Yang, Shunjiang Wang, Chaoran Li, Ni Han and Ximing Zhang

Energy, 2025, vol. 314, issue C

Abstract: The vast amount of data in the power system is an essential foundation for maintaining grid stability and ensuring the normal operation of businesses and residents in their daily activities. However, some consumers reduce their electricity bills by using electricity in an irregular manner. This behavior significantly reduces the security of the power grid and the reliability of load power data in the power system. In this paper, a multi-dimensional feature-based method for detecting electricity load anomalies is proposed. Firstly, deep denoising recurrent neural network (DDRNN) and gated recurrent unit (GRU) are utilized to improve the generator and discriminator of auxiliary classifier generating adversarial network (ACGAN). Compared with the data injection method, the data generated by Model exhibits a high degree of similarity to the data from actual abnormal electricity load. Secondly, the data is analyzed and mined by deep convolutional neural networks (DCNN) from multiple perspectives to improve detection method. Information about the data and the connections between the data are revealed in detail. Thirdly, reactive power is used to extend the data dimensions and optimize the training process of the machine learning model. The input information of the learner is augmented to improve the adaptability of the detection method. Numerical examples show that the proposed method is superior to the existing technologies in terms of detection results.

Keywords: Energy management; Energy security; Electricity theft; Deep learning; Generative adversarial network (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544224039859
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:314:y:2025:i:c:s0360544224039859

DOI: 10.1016/j.energy.2024.134207

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().