基于水电关联信息的零电量低压用户窃电检测

RichHTML

PDF (PC)

可视化

RichHTML

PDF (PC)

可视化

基于水电关联信息的零电量低压用户窃电检测

Detection of Electricity Theft by Low Voltage Users with Zero Power Consumption Based on Water-Electricity Correlation Information

基于水电关联信息的零电量低压用户窃电检测

Detection of Electricity Theft by Low Voltage Users with Zero Power Consumption Based on Water-Electricity Correlation Information

基于水电关联信息的零电量低压用户窃电检测

Detection of Electricity Theft by Low Voltage Users with Zero Power Consumption Based on Water-Electricity Correlation Information

摘要/Abstract

图/表 15

参考文献 26

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 26

相关文章 4

编辑推荐

Metrics

模态框（Modal）标题

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 26

相关文章 4

编辑推荐

Metrics

doi:10.11930/j.issn.1004-9649.202211098

中国电力 ›› 2023, Vol. 56 ›› Issue (12): 206-216.DOI: 10.11930/j.issn.1004-9649.202211098

邹念¹(), 魏梅芳²(), 苏盛¹(), 郑应俊¹(), 周文晴¹

1. 长沙理工大学电气与信息工程学院，湖南长沙　410014
2. 国网湖南省电力有限公司技术技能培训中心，湖南长沙　410131

收稿日期:2022-11-28 接受日期:2023-04-27 出版日期:2023-12-28 发布日期:2023-12-28
作者简介:邹念（1999—），男，硕士研究生，从事用电数据挖掘分析利用，E-mail： 873685903@qq.com
魏梅芳（1982—），女，硕士，副教授，从事电力营销服务研究，E-mail： weimf@cseptc.net
苏盛（1975—），男，通信作者，博士，教授，从事配用电大数据应用、电力气象灾害和电力系统网络安全防护，E-mail： eessheng@163.com
郑应俊（1997—）男，硕士研究生，从事用电数据挖掘分析利用研究，E-mail： 5226495@qq.com
基金资助:
国家自然科学基金资助项目（51777015）；湖南省自然科学基金资助项目（2022JJ60089）。

Nian ZOU¹(), Meifang WEI²(), Sheng SU¹(), Yingjun ZHENG¹(), Wenqing ZHOU¹

1. School of Electrical & Information Engineering, Changsha University of Science and Technology, Changsha 410114, China
2. Technical Skills Training Center of State Grid Hunan Electric Power Co., Ltd., Changsha 410131, China

Received:2022-11-28 Accepted:2023-04-27 Online:2023-12-28 Published:2023-12-28
Supported by:
This work is supported by National Natural Science Foundation of China (No.51777015) and Natural Science Foundation of Hunan Province (No.2022JJ60089).

摘要：

低压居民窃电用户中的零电量用户，不能提供用户用电行为的有效信息，又容易与空置房用户混淆，是窃电检测中的一类特殊问题。利用居民用户用水、用电数据之间具有强关联性的特点，提出基于水、电关联信息的零电量用户窃电检测。首先，分析居民用户用电、用水数据间的关联关系；然后，构建用户日用电量与日用水量的最大互信息模型，计算不同时间尺度下的最大互信息系数来衡量其信息相关度；接着，对用户的最大互信息系数进行聚类，将显著偏离类簇的样本识别为水电量具有弱相关性的窃电嫌疑用户，当窃电嫌疑用户的用电量为零时即为零电量窃电用户；2个台区的测试算例表明：所提方法可有效检出零电量窃电用户，指导现场窃电检测工作。

关键词: 多表合一, 零电量用户, 窃电, 水电使用行为, 最大互信息系数, 小波聚类

Abstract:

It is hard to obtain the effective information on electricity consumption behaviors of zero-power consumption users among low-voltage residential electricity theft users who are easily confused with vacant house owners, which is a special problem for electricity theft detection. Based on the strong correlation between water and electricity data of residential users, a zero-power user electricity theft detection method is proposed based on water-electricity correlation information. Firstly, the relationship between electricity and water consumption data of residential users is analyzed. Then, a maximal information model of users' daily electricity consumption and daily water consumption is constructed, and the maximum information coefficient (MIC) at different time scales is calculated to measure the information correlation. Thirdly, the user's maximal information coefficient is clustered, and the samples that deviate significantly from the cluster are identified as the suspected electricity theft users with weak water-electricity correlation. When the electricity consumption of the suspected electricity theft users is zero, they are identified as the electricity theft users with zero power consumption. The test examples of two distribution areas show that the proposed method can effectively detect the electricity theft users with zero power consumption and guide the on-site electricity theft detection work.

Key words: multi-meter unification, zero power consumption users, electricity theft, water and electricity consumption behavior, maximal information coefficient, wavelet clustering

邹念, 魏梅芳, 苏盛, 郑应俊, 周文晴. 基于水电关联信息的零电量低压用户窃电检测[J]. 中国电力, 2023, 56(12): 206-216.

Nian ZOU, Meifang WEI, Sheng SU, Yingjun ZHENG, Wenqing ZHOU. Detection of Electricity Theft by Low Voltage Users with Zero Power Consumption Based on Water-Electricity Correlation Information[J]. Electric Power, 2023, 56(12): 206-216.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202211098

https://www.electricpower.com.cn/CN/Y2023/V56/I12/206

图 1 典型居民用户不同时间尺度水电消费散点图

Fig.1 Scatter diagram of water and electricity consumption of typical residential users at different time scales

图 2 典型居民用户日水电消费示意

Fig.2 Diagram of daily water and electricity consumption of typical residential users

图 3 典型居民用户每月水电消费Pearson值箱线图

Fig.3 Box diagram of Pearson value of monthly water and electricity consumption of typical residential users

图 4 二维空间生成的MIC值

Fig.4 MIC generated by two-dimensional space

图 5 小波聚类流程

Fig.5 Flowchart of wavelet clustering

图 6 基于用户日水电数据的窃电检测流程

Fig.6 Flowchart of electricity theft detection based on users' daily water and electricity consumption data

图 7 当Q=20%、T=25天时日用水电量的MIC曲线

Fig.7 MIC curve of daily water and electricity consumption when Q=20% and T=25 days

图 8 当Q=20%、T=25天时小波聚类可视化结果

Fig.8 Visualization result of wavelet clustering when Q=20% and T=25 days

表 1 Q=20%时各种滑窗区间下的小波聚类结果

Table 1 Wavelet clustering results under various sliding window intervals when Q=20%

表 2 仅有用电数据的小波聚类结果

Table 2 Wavelet clustering results with only electricity consumption data

Q/%	检测结果（异常用户1~5的个数）	误检个数	Q/%	检测结果（异常用户1~5的个数）	误检个数
20	2	2	60	5	0
30	4	5	80	5	0
40	5	0	100	5	0

表 3 各种零电量时长比例下的检测结果

Table 3 Test results under various zero-power duration ratios

检测时长T/天	检测结果（异常用户1~5的个数）
检测时长T/天	Q=20%	Q=40%	Q=60%	Q=80%	Q=100%
15	3	5	5	5	5
20	5	5	5	5	5
25	5	5	5	5	5
30	5	5	5	5	5
35	5	5	5	5	5
40	4	5	5	5	5

图 9 高损台区2日用水电量的最大信息系数曲线

Fig.9 Maximum information coefficient curve of daily water and electricity consumption in 2# high power-loss distribution area

图 10 高损台区2小波聚类可视化结果

Fig.10 Visualization result of wavelet clustering in 2# high power-loss distribution area

表 4 高损台区2线损电量与用电量的格兰杰检验结果

Table 4 Granger test results of line loss and electricity consumption in 2# high power-loss distribution area

假设		显著性水平
线损电量不是用户7的格兰杰原因		0.2414
用户7不是线损电量的格兰杰原因		0.3008
线损电量不是用户29的格兰杰原因		0.3559
用户29不是线损电量的格兰杰原因		0.0385

图 11 高损台区2线损电量与用电量的MIC值结果

Fig.11 MIC value of line loss and electricity consumption in 2# high power-loss distribution area

1	ZHENG Z B, YANG Y T, NIU X D, et al. Wide and deep convolutional neural networks for electricity-theft detection to secure smart grids[J]. IEEE Transactions on Industrial Informatics, 2018, 14 (4): 1606- 1615. DOI
2	殷涛, 薛阳, 杨艺宁, 等. 基于向量自回归模型的高损线路窃电检测[J]. 中国电机工程学报, 2022, 42 (3): 1015- 1024. DOI
	YIN Tao, XUE Yang, YANG Yining, et al. Electricity theft detection of high-loss line with vector autoregression[J]. Proceedings of the CSEE, 2022, 42 (3): 1015- 1024. DOI
3	段俊峰, 李宁, 唐求, 等. 基于CK-GPR的多应力环境智能电表剩余寿命预测[J]. 仪器仪表学报, 2021, 42 (4): 102- 110. DOI
	DUAN Junfeng, LI Ning, TANG Qiu, et al. Remaining useful life prediction of smart meter based on CK-GPR in multi-stress environment[J]. Chinese Journal of Scientific Instrument, 2021, 42 (4): 102- 110. DOI
4	滕伟, 黄乙珂, 吴仕明, 等. 基于XGBoost与LSTM的风力发电机绕组温度预测[J]. 中国电力, 2021, 54 (6): 95- 103. DOI
	TENG Wei, HUANG Yike, WU Shiming, et al. Wind turbine generator winding temperature prediction based on XGBoost and LSTM[J]. Electric Power, 2021, 54 (6): 95- 103. DOI
5	陈启鑫, 郑可迪, 康重庆, 等. 异常用电的检测方法: 评述与展望[J]. 电力系统自动化, 2018, 42 (17): 189- 199. DOI
	CHEN Qixin, ZHENG Kedi, KANG Chongqing, et al. Detection methods of abnormal electricity consumption behaviors: review and prospect[J]. Automation of Electric Power Systems, 2018, 42 (17): 189- 199. DOI
6	杨艺宁, 薛阳, 徐英辉, 等. 基于行业特性的路灯用电异常检测方法[J]. 电力科学与技术学报, 2021, 36 (3): 165- 173. DOI
	YANG Yining, XUE Yang, XU Yinghui, et al. Sector electricity consumption behavior features based abnormal electricity consumption detection method for street lamps[J]. Journal of Electric Power Science and Technology, 2021, 36 (3): 165- 173. DOI
7	金晟, 苏盛, 曹一家, 等. 基于格兰杰归因分析的高损台区窃电检测[J]. 电力系统自动化, 2020, 44 (23): 82- 89. DOI
	JIN Sheng, SU Sheng, CAO Yijia, et al. Electricity-theft detection for high-loss distribution area based on granger causality analysis[J]. Automation of Electric Power Systems, 2020, 44 (23): 82- 89. DOI
8	刘康, 李彬, 薛阳, 等. 基于传递熵密度聚类的用户窃电识别方法[J]. 中国电机工程学报, 2022, 42 (20): 7535- 7546. DOI
	LIU Kang, LI Bin, XUE Yang, et al. User electric theft detection method based on transfer entropy density clustering[J]. Proceedings of the CSEE, 2022, 42 (20): 7535- 7546. DOI
9	唐冬来, 刘友波, 熊智临, 等. 基于时空关联矩阵的配电台区反窃电预警方法[J]. 电力系统自动化, 2020, 44 (19): 168- 176. DOI
	TANG Donglai, LIU Youbo, XIONG Zhilin, et al. Early warning method of electricity anti-theft in distribution station area based on spatiotemporal correlation matrix[J]. Automation of Electric Power Systems, 2020, 44 (19): 168- 176. DOI
10	吴迪. 基于曲线相似性分析的窃电用户判断[J]. 中国电力, 2017, 50 (2): 181- 184. DOI
	WU Di. Electricity theft identification method based on curve similarity[J]. Electric Power, 2017, 50 (2): 181- 184. DOI
11	ZHENG K D, CHEN Q X, WANG Y, et al. A novel combined data-driven approach for electricity theft detection[J]. IEEE Transactions on Industrial Informatics, 2019, 15 (3): 1809- 1819. DOI
12	章博, 卢峰, 董寒宇, 等. 基于决策树和数据驱动的零电量用户筛选方法[J]. 山东大学学报(工学版), 2019, 49 (5): 29- 36.
	ZHANG Bo, LU Feng, DONG Hanyu, et al. None-consumption users filtering algorithm based on decision tree and data-driven methods[J]. Journal of Shandong University (Engineering Science), 2019, 49 (5): 29- 36.
13	金晟, 苏盛, 薛阳, 等. 数据驱动窃电检测方法综述与低误报率研究展望[J]. 电力系统自动化, 2022, 46 (1): 3- 14.
	JIN Sheng, SU Sheng, XUE Yang, et al. Review on data-driven based electricity theft detection method and research prospect for low false positive rate[J]. Automation of Electric Power Systems, 2022, 46 (1): 3- 14.
14	中国电力企业联合会. 电、水、气、热能源计量管理系统第1部分: 总则: T/CEC 122.1—2016[S]. 2017.
15	王新刚, 赵舫, 朱文君. 基于综合能源计量数据的区域用能特性分析[J]. 中国电力, 2022, 55 (9): 140- 145.
	WANG Xingang, ZHAO Fang, ZHU Wenjun. Pattern analysis of regional energy consumption based on integrated energy measurement data[J]. Electric Power, 2022, 55 (9): 140- 145.
16	黄瑞, 刘谋海, 陈向群, 等. 基于递阶结构的多表合一能源计量系统运行状态综合评价方法[J]. 电测与仪表, 2023, 60 (8): 187- 193.
	HUANG Rui, LIU Mouhai, CHEN Xiangqun, et al.. Comprehensive evaluation method of multi-meter integrated energy metering operating state based on hierarchical structure[J]. Electrical Measurement & Instrumentation, 2023, 60 (8): 187- 193.
17	孟燕霞. 最大信息系数算法研究[D]. 太原: 太原理工大学, 2019.
	MENG Yanxia. The maximal information coefficient algorithm research[D]. Taiyuan: Taiyuan University of Technology, 2019.
18	王月. 最大信息系数的算法分析及改进[D]. 西安: 西安电子科技大学, 2019.
	WANG Yue. Algorithm analysis and improvement of the maximal information coefficient[D]. Xi'an: Xidian University, 2019.
19	赵雅雪, 王旭, 蒋传文, 等. 基于最大信息系数相关性分析和改进多层级门控LSTM的短期电价预测方法[J]. 中国电机工程学报, 2021, 41 (1): 135- 146, 404. DOI
	ZHAO Yaxue, WANG Xu, JIANG Chuanwen, et al. A novel short-term electricity price forecasting method based on correlation analysis with the maximal information coefficient and modified multi-hierachy gated LSTM[J]. Proceedings of the CSEE, 2021, 41 (1): 135- 146, 404. DOI
20	RESHEF D N, RESHEF Y A, FINUCANE H K, et al. Detecting novel associations in large data sets[J]. Science, 2011, 334 (6062): 1518- 1524. DOI
21	薛阳, 张宁, 吴海东, 等. 基于UTCI-MIC与振幅压缩灰色模型的用户侧微电网短期负荷预测方法[J]. 电网技术, 2020, 44 (2): 556- 563. DOI
	XUE Yang, ZHANG Ning, WU Haidong, et al. Short-term load forecasting method for user side microgrid based on UTCI-MIC and amplitude compression grey model[J]. Power System Technology, 2020, 44 (2): 556- 563. DOI
22	孙广路, 宋智超, 刘金来, 等. 基于最大信息系数和近似马尔科夫毯的特征选择方法[J]. 自动化学报, 2017, 43 (5): 795- 805. DOI
	SUN Guanglu, SONG Zhichao, LIU Jinlai, et al. Feature selection method based on maximum information coefficient and approximate Markov blanket[J]. Acta Automatica Sinica, 2017, 43 (5): 795- 805. DOI
23	刘晓波, 邵伟芹, 张明明, 等. 基于双网格校正小波聚类的转子故障诊断[J]. 计算机集成制造系统, 2017, 23 (9): 1883- 1890. DOI
	LIU Xiaobo, SHAO Weiqin, ZHANG Mingming, et al. Rotor fault diagnosis based on double grid corrected wavecluster algorithm[J]. Computer Integrated Manufacturing Systems, 2017, 23 (9): 1883- 1890. DOI
24	沈沉, 秦建, 盛万兴, 等. 基于小波聚类的配变短期负荷预测方法研究[J]. 电网技术, 2016, 40 (2): 521- 526. DOI
	SHEN Chen, QIN Jian, SHENG Wanxing, et al. Study on short-term forecasting of distribution transformer load using wavelet and clustering method[J]. Power System Technology, 2016, 40 (2): 521- 526. DOI
25	张明明. 基于双网格校正小波聚类的航空发动机转子系统振动故障诊断[D]. 南昌: 南昌航空大学, 2016.
	ZHANG Mingming. The aero engine rotor system's vibration fault diagnosis based on the double grid corrected wave cluster algorithm[D]. Nanchang: Nanchang Hangkong University, 2016.
26	左红艳, 刘晓波, 洪连环. 用于转子故障诊断的改进小波聚类算法[J]. 振动测试与诊断, 2018, 38 (2): 320- 326, 420.
	ZUO Hongyan, LIU Xiaobo, HONG Lianhuan. Improved wave cluster method for rotor fault diagnosis[J]. Journal of Vibration, Measurement & Diagnosis, 2018, 38 (2): 320- 326, 420.

[1]	陶磊, 罗萍萍, 林济铿. 基于深度学习的直流微电网虚假数据注入攻击二阶段检测方法[J]. 中国电力, 2024, 57(9): 11-19.
[2]	王新刚, 赵舫, 朱文君. 基于综合能源计量数据的区域用能特性分析[J]. 中国电力, 2022, 55(9): 140-145.
[3]	吴迪. 基于曲线相似性分析的窃电用户判断[J]. 中国电力, 2017, 50(2): 181-184.
[4]	刘盛, 朱翠艳. 应用数据挖掘技术构建反窃电管理系统的研究[J]. 中国电力, 2017, 50(10): 181-184.

检测结果（异常
用户1~5的个数）

检测结果（异常
用户1~5的个数）