Anomaly Detection Method of Smart Meter Based on Tikhonov Regularization and Generalized Cross-Validation

Lidan CHEN , Yongliang MA , Zhe ZHANG , Zibin KE

›› 2023, Vol. 17 ›› Issue (5) : 125 -133.

PDF
›› 2023, Vol. 17 ›› Issue (5) : 125 -133. DOI: 10.13648/j.cnki.issn1674-0629.2023.05.014
research-article

Anomaly Detection Method of Smart Meter Based on Tikhonov Regularization and Generalized Cross-Validation

Author information +
History +
PDF

Abstract

At present, the power grid companies of China basically adopt expired replacement and batch sampling rotation for smart meters, which causes a waste of human and material resources. This paper proposes a method for identifying and recommending that the operating error of a smart meter is out of tolerance based on the electricity consumption data of the metering automation system. Firstly, the topology and mathematical model for analyzing the operating error of the smart meters under the distribution transformer are present. Secondly, considering the ill-conditioned problem, a solution method based on Tikhonov regularization (Tik) and generalized cross-validation (GCV) to optimize the regularization parameters is suggested. Then, a sorting recommendation mechanism for identifying out-of-tolerance smart meters is proposed. Finally, it is verified with actual distribution grid data in a certain district and compared with other methods. The results show that the proposed method is practical and effective.

Keywords

smart meter / generalized cross-validation (GCV) / Tikhonov regularization (Tik) / anomaly detection

Cite this article

Download citation ▾
Lidan CHEN,Yongliang MA,Zhe ZHANG,Zibin KE. Anomaly Detection Method of Smart Meter Based on Tikhonov Regularization and Generalized Cross-Validation. 2023, 17(5): 125-133 DOI:10.13648/j.cnki.issn1674-0629.2023.05.014

登录浏览全文

4963

注册一个新账户 忘记密码

References

Funding

the International Cooperation and Exchange program of National Natural Science Foundation of China(51761145106)

the Open Fund Project of Intelligent Electric Power Grid Key Laboratory of Sichuan Province(2021-IEPGKLSP-KFZA01)

PDF

8

Accesses

0

Citation

Detail

Sections
Recommended

AI思维导图

/