User-side Abnormal Power Consumption Detection Method Based on High-dimensional Random Matrix

To accurately grasp the abnormal electricity usage behavior of users on the distribution network side, the application of user electricity risk hazard identification feature extraction and recognition technology is used to achieve a sense of security in electricity usage and a strategy for intellige...

Full description

Saved in:
Bibliographic Details
Published in2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) pp. 782 - 788
Main Authors Gao, Jiaming, Kong, Xiangyu, Zhang, Delong, Xu, Xihao
Format Conference Proceeding
LanguageEnglish
Published IEEE 21.07.2024
Subjects
customer-side abnormal electricity use
Data visualization
distribution of eigenvalues
Electricity
highdimensional random matrix
identification electricity theft identification
Maintenance
Meters
Power demand
Power grids
Reliability theory
Security
Smart meters
Systems simulation
Online AccessGet full text

Cover

More Information
Summary:To accurately grasp the abnormal electricity usage behavior of users on the distribution network side, the application of user electricity risk hazard identification feature extraction and recognition technology is used to achieve a sense of security in electricity usage and a strategy for intelligent, digital user electricity risk early warning push services is proposed. The collected power grid operation data is analysed and modeled based on the random matrix theory. By characterizing the statistical properties of the matrix, the power consumption information is visualized, and a method for identifying abnormal power consumption and meter faults is proposed. This method analyses the fluctuation characteristics of the eigenvalues and gives the specific steps for applying the random matrix theory to the location of abnormal areas of user power consumption. It can also realize the feature discovery of abnormal power consumption points on the user side and then determine abnormal power consumption behavior. The case study shows that this method is highly reliable and accurate.
DOI:10.1109/CCSSTA62096.2024.10691714