A hybrid model for smart grid theft detection based on deep learning

Yinling LIAO; Jincan LI; Bing WANG; Jun ZHANG; Yaoyuan LIANG

doi:10.11959/j.issn.1000-0801.2024027

您当前的位置：

首页 >

文章列表页 >

A hybrid model for smart grid theft detection based on deep learning

Research and Development | 更新时间：2024-06-05

- A hybrid model for smart grid theft detection based on deep learning
- Telecommunications Science Vol. 40, Issue 2, Pages: 72-82(2024)
- 作者机构：
  
  1. 广西电网有限责任公司，广西南宁 530023
  2. 广西电网有限责任公司梧州供电局，广西梧州 543002
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2024027
  CLC： TM72
- Published Online：2024-02，
  
  Published：20 February 2024
- 稿件说明：
移动端阅览
Yinling LIAO, Jincan LI, Bing WANG, et al. A hybrid model for smart grid theft detection based on deep learning[J]. Telecommunications science, 2024, 40(2): 72-82.
DOI：

Yinling LIAO, Jincan LI, Bing WANG, et al. A hybrid model for smart grid theft detection based on deep learning[J]. Telecommunications science, 2024, 40(2): 72-82. DOI： 10.11959/j.issn.1000-0801.2024027.

摘要

针对传统窃电检测模型受维度诅咒、类不平衡等问题，提出一种能有效检测智能电网窃电行为的混合深度学习模型，利用深度学习卷积神经网络（AlexNet）处理维度诅咒问题，显著提升数据处理的准确性；通过自适应增强（AdaBoost）对正常和异常用电行为分类，进一步提高分类精度；使用欠采样技术解决类不平衡问题，确保模型在各类数据的均衡性能；利用人工蜂群算法对AdaBoost和AlexNet的超参数进行优化，有效提高整体模型性能。使用真实智能电表数据集评估混合模型的有效性，与同类模型相比，提出的混合深度学习模型在准确率、精确度、召回率、F1 分数、马修斯相关系数（MCC）和曲线下面积-接收者操作特征曲线（AUC-ROC）分数上分别达到了 88%、86%、84%、85%、78%和 91%，不仅提高了用电行为监测的准确性，也为电力系统的智能分析提供了新视角。

Abstract

A hybrid deep learning model was proposed to effectively detect electricity theft in smart grids.The hybrid model employed a deep learning convolutional neural network (AlexNet) to tackle the curse of dimensionality

significantly enhancing data processing accuracy and efficiency.It further improved classification accuracy by differentiating between normal and abnormal electricity usage using adaptive boosting (AdaBoost).To resolve the issue of class imbalance

undersampling techniques were utilized

ensuring balanced performance across various data classes.Additionally

the artificial bee colony algorithm was used to optimize hyperparameters for both AdaBoost and AlexNet

effectively boosting overall model performance.The effectiveness of this hybrid model was evaluated using real smart meter datasets from an electricity company.Compared to similar models

this hybrid model achieves accuracy

precision

recall

F1-score

Matthews correlation coefficient (MCC)

and area under the curve-receiver operating characteristic curve (AUC-ROC) scores of 88%

86%

84%

85%

78%

and 91%

respectively.The proposed model not only increases the accuracy of electricity usage monitoring

but also offers a new perspective for intelligent analysis in power systems.

关键词

Keywords

references

刘炜 , 王朝亮 , 肖涛 , 等 . AMI数据驱动的电动汽车充电设施计量运行误差状态评估方法 [J ] . 电力自动化设备 , 2022 , 42 ( 10 ): 70 - 76 .

LIU W , WANG C L , XIAO T , et al . AMI data-driven state evaluation method for measurement operation error of electric vehicle charging facilities [J ] . Electric Power Automation Equipment , 2022 , 42 ( 10 ): 70 - 76 .

白志霞 , 刘馨卉 , 索思远 , 等 . 基于集群智能的智能电能表异常检测技术 [J ] . 电测与仪表 , 2022 , 59 ( 1 ): 195 - 200 .

BAI Z X , LIU X H , SUO S Y , et al . Anomaly detection technology based on swarm intelligence for smart meters [J ] . Electrical Measurement ＆ Instrumentation , 2022 , 59 ( 1 ): 195 - 200 .

曹德发 , 马明 , 汤平瑜 , 等 . 多源量测环境下基于深度强化学习的配电网分布式状态估计方法 [J ] . 可再生能源 , 2021 , 39 ( 10 ): 1414 - 1420 .

CAO D F , MA M , TANG P Y , et al . Distributed state estimation method of distribution network based on deep reinforcement learning in multi source measurement environment [J ] . Renewable Energy Resources , 2021 , 39 ( 10 ): 1414 - 1420 .

AMIN S , SCHWARTZ G A , TEMBINE H . Incentives and security in electricity distribution networks [C ] // International Conference on Decision and Game Theory for Security . Berlin,Heidelberg:Springer , 2012 : 264 - 280 .

刘文浩 , 冯玥 , 姜东良 . 基于AMI数据驱动的窃电用户识别研究 [J ] . 制造业自动化 , 2022 , 44 ( 11 ): 5 - 8 .

LIU W H , FENG Y , JIANG D L . Research on identification of electric theft customers based on AMI data drive [J ] . Manufacturing Automation , 2022 , 44 ( 11 ): 5 - 8 .

PEREIRA J , SARAIVA F . Convolutional neural network applied to detect electricity theft:a comparative study on unbalanced data handling techniques [J ] . International Journal of Electrical Power ＆ Energy Systems , 2021 ( 131 ): 107085 .

刘智慧 . 基于并行CNN-GRU的高级量测体系网络入侵检测研究 [D ] . 大连:大连理工大学 , 2022 .

LIU Z H . Research on network intrusion detection of advanced metering infrastructure based on parallel CNN-GRU [D ] . Dalian:Dalian University of Technology , 2022 .

姚冬缓 . 智能电网中的盗电检测技术研究 [D ] . 上海:上海电力大学 , 2019 .

YAO D H . Research on detection detection technology of energy theft in smart grid [D ] . Shanghai:Shanghai University of Electric Power , 2019 .

汪涛 , 梁瑞宇 , 黄虎 , 等 . 基于多层多核卷积神经网络的非侵入式负荷监测方法研究 [J ] . 电子器件 , 2021 , 44 ( 6 ): 1429 - 1435 .

WANG T , LIANG R Y , HUANG H , et al . Research on non-intrusive load monitoring method based on multi-layer and multi-kernel convolution neural network [J ] . Chinese Journal of Electron Devices , 2021 , 44 ( 6 ): 1429 - 1435 .

张玉天 , 邓春宇 , 刘沅昆 , 等 . 基于卷积神经网络的非侵入负荷辨识算法 [J ] . 电网技术 , 2020 , 44 ( 6 ): 2038 - 2044 .

ZHANG Y T , DENG C Y , LIU Y K , et al . Non-intrusive load identification algorithm based on convolution neural network [J ] . Power System Technology , 2020 , 44 ( 6 ): 2038 - 2044 .

徐先峰 , 赵依 , 刘状壮 , 等 . 用于短期电力负荷预测的日负荷特性分类及特征集重构策略 [J ] . 电网技术 , 2022 , 46 ( 4 ): 1548 - 1556 .

XU X F , ZHAO Y , LIU Z Z , et al . Daily load characteristic classification and feature set reconstruction strategy for short-term power load forecasting [J ] . Power System Technology , 2022 , 46 ( 4 ): 1548 - 1556 .

张梦楠 , 李红娇 . 基于DCNN和SVC的窃电检测 [J ] . 计算机仿真 , 2022 , 39 ( 6 ): 92 - 97 .

ZHANG M N , LI H J . Electricity theft detection based on deep convolutional neural network and support vector classification [J ] . Computer Simulation , 2022 , 39 ( 6 ): 92 - 97 .

齐言强 . 基于蚁群算法伺服系统控制参数的研究应用 [D ] . 西安:西安工程大学 , 2018 .

QI Y Q . Research and application of servo system control parameters based on ant colony algorithm [D ] . Xi’an:Xi’an Polytechnic University , 2018 .

陈春玲 , 许童羽 , 郑伟 , 等 . 多类分类SVM在电能质量扰动识别中的应用 [J ] . 电力系统保护与控制 , 2010 , 38 ( 13 ): 74 - 78 .

CHEN C L , XU T Y , ZHENG W , et al . Application of multi-class classification SVM in power quality disturbances classification [J ] . Power System Protection and Control , 2010 , 38 ( 13 ): 74 - 78 .

周高涵 . 基于机器学习算法的电力系统暂态稳定性评估研究 [D ] . 武汉:武汉理工大学 , 2022 .

ZHOU G H . Research on Power system transient stability assessment based on machine learning algorithms [D ] . Wuhan:Wuhan University of Technology , 2022 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Optimal active defense decision model of wireless communication network under APT attack

Low complexity radar signal classification based on spectrum shape

Face recognition algorithm based on attractive local second gradient contours

Internet bad information detection based on Bert model

Smart prediction of the complaint hotspot problem in mobile network

Related Author

Danni WANG

Jun LYU

Fuliang ZHANG

Meng MENG

Yuan HE

Lin ZHOU

Yuliang YAO

Xiaolei WANG

Related Institution

School of Communication and Information Engineering, Beijing University of Posts and Telecommunications

National Innovation Institute of Defense Technology, Academy of Military Science

College of Information Engineering,China Jiliang University

College of Mechanical and Electrical Engineering,China Jiliang University

⁰