Grid anomaly detection and localization based on recursive feature elimination and dynamic heterogeneous ensemble

XI Lei; LIU Wei; BAI Fangyan; GAO Haowen; LI Zongze

doi:10.19783/j.cnki.pspc.260213

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Grid anomaly detection and localization based on recursive feature elimination and dynamic heterogeneous ensemble

更新时间：2026-10-16

- Grid anomaly detection and localization based on recursive feature elimination and dynamic heterogeneous ensemble
- Power System Protection and Control Vol. 54, Issue 17, (2026)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：10.19783/j.cnki.pspc.260213
  CLC：
- Published：2026
- 稿件说明：
移动端阅览
XI Lei, LIU Wei, BAI Fangyan, et al. Grid anomaly detection and localization based on recursive feature elimination and dynamic heterogeneous ensemble[J]. Power System Protection and Control, 2026, 54(17).
DOI：

XI Lei, LIU Wei, BAI Fangyan, et al. Grid anomaly detection and localization based on recursive feature elimination and dynamic heterogeneous ensemble[J]. Power System Protection and Control, 2026, 54(17). DOI： 10.19783/j.cnki.pspc.260213.

摘要

针对故障和虚假数据注入攻击严重威胁电网稳定运行，且现有异常检测算法精度低、无法准确辨识系统状态并定位异常区域的问题，提出一种基于递归特征消除‑动态异构集成的电网异常状态检测与区域定位算法。首先，通过融入交叉验证的递归特征消除策略，为每个标签选择出拟合效果最佳的特征，以降低数据维度并提升分类器的响应效率。然后，运用动态异构集成算法对系统的故障、正常与攻击进行辨识，实现高精度的状态检测。最后，分别对故障和虚假数据注入攻击进行精确区域定位。在 IEEE14 与 30 节点系统上开展的大量对比仿真实验表明，所提算法在准确率及 F1 值均有明显优势，并且显著提升了区域定位的精度。

Abstract

System faults and false data injection attacks pose significant threats to the stable operation of power grids. Existing anomaly detection algorithms have low accuracy and cannot accurately identify system states and locate abnormal regions. To address these challenges

an abnormal state detection and regional localization algorithm for power grids based on recursive feature elimination‑dynamic heterogeneous ensemble is proposed. First

by utilizing the recursive feature elimination strategy integrated with cross‑validation

the features with the best fitting effect are selected for each label

thereby reducing the data dimensionality and improving the response efficiency of the classifier. Then

the dynamic heterogeneous ensemble algorithm is applied to distinguish among normal operation

system faults

and cyberattacks

enabling high‑accuracy system state detection. Finally

precise regional localization methods are performed for both faults and false data injection attacks. Extensive comparative simulations conducted on the IEEE14‑ and 30‑bus systems demonstrate that the proposed algorithm achieves superior performance in terms of accuracy and F1 score

while significantly enhancing the precision of regional localization.

关键词

Keywords

references

ASADI MAJD A, SAMET H, GHANBARI T. k‑NN based fault detection and classification methods for power transmission systems[J]. Protection and Control of Modern Power Systems, 2017, 2(1): 1‑11.

HOU W, YI L, MIAO H, et al. Robust kernel density estimation based data‑driven optimal scheduling for power systems considering data errors and uncertainties of renewable energy[J]. IEEE Access, 2024, 12: 81329‑81337.

ZHANG C, GUO Z, LI C. Unsupervised anomaly detection for gearboxes based on the deep convolutional support generative adversarial network[J]. Scientific Reports, 2025, 15(1).

JORJANI M, SEIFI H, VARJANI A Y. A graph theory‑ based approach to detect false data injection attacks in power system AC state estimation[J]. IEEE Transactions on Industrial Informatics, 2021, 17(4): 2465‑2475.

DRAYER E, ROUTTENBERG T. Detection of false data injection attacks in smart grids based on graph signal processing[J]. IEEE Systems Journal, 2020, 14(2): 1886‑1896.

LI S, YILMAZ Y, WANG X. Quickest detection of false data injection attack in wide‑area smart grids[J]. IEEE Transactions on Smart Grid, 2015, 6(6): 2725‑2735.

DING Y, MA K, PU T, et al. A deep learning‑based classification scheme for cyber‑attack detection in power system[J]. IET Energy Systems Integration, 2021, 3(3): 274‑284.

XIA W, HE D, YU L. Locational detection of false data injection attacks in smart grids: a graph convolutional attention network approach[J]. IEEE Internet of Things Journal, 2024, 11(6): 9324‑9337.

LOUK M H L, TAMA B A. Revisiting gradient boosting‑ based approaches for learning imbalanced data: a case of anomaly detection on power grids[J]. Big Data and Cognitive Computing, 2022, 6(2): 41‑49.

ZHOU M, WANG Y, SRIVASTAVA A K, et al. Ensemble‑ based algorithm for synchrophasor data anomaly detection[J]. IEEE Transactions on Smart Grid, 2019, 10(3): 2979‑2988.

YI N, WANG Q, YAN L, et al. A multi‑stage game model for the false data injection attack from attacker’s perspective[J]. Sustainable Energy, Grids and Networks, 2021, 28.

XI L, TIAN X, HE M, et al. False data injection detection in power system based on LOSSA‑AdaBoostDT[J]. Protection and Control of Modern Power Systems, 2025, 10(3): 55‑64.

YAN J, GUO F, WEN C. False data injection against state estimation in power systems with multiple cooperative attackers[J]. ISA Transactions, 2020, 101: 225‑233.

WU T, XUE W, WANG H, et al. Extreme learning machine‑based state reconstruction for automatic attack filtering in cyber physical power system[J]. IEEE Transactions on Industrial Informatics, 2021, 17(3): 1892‑1904.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A false data injection detection method for DC microgrids based on deep spatiotemporal feature learning

基于关联差异的电力信息物理系统虚假数据注入攻击检测

基于海马优化深层极限学习机的电力信息物理系统FDIA检测

基于混合黑猩猩优化极限学习机的电力信息物理系统虚假数据注入攻击定位检测

考虑负荷虚假数据注入攻击的电力信息物理系统防御策略

Related Author

ZHANG Shida

LI Zhongwen

TANG Liang

LUO Shengyao

WANG Yi

LI Bin

WANG Yue

GUO Xiaoli

Related Institution

郑州大学电气与信息工程学院

河南省电力电子与电能系统工程技术研究中心

国网镇江供电分公司

Bin(1. School of Computer Science, Northeast Electric Power University

Jilin Engineering Technology Research Center of Intelligent Electric Power Big Data Processing

AI问答

⁰