基于改进深度自编码网络的异常用电行为辨识

doi:10.11930/j.issn.1004-9649.201910005

中国电力 ›› 2020, Vol. 53 ›› Issue (6): 18-26.DOI: 10.11930/j.issn.1004-9649.201910005

• 人工智能在电力系统的应用 • 上一篇下一篇

基于改进深度自编码网络的异常用电行为辨识

林女贵¹, 洪兰秀², 黄道姗³, 易扬², 刘智煖³, 徐启峰⁴

1. 国网福建省电力有限公司，福建福州 350003;
2. 国网福建省电力有限公司经济技术研究院，福建福州 350012;
3. 国网福建省电力有限公司电力科学研究院，福建福州 350007;
4. 福州大学电气工程与自动化学院，福建福州 350116

收稿日期:2019-10-10 修回日期:2020-02-13 发布日期:2020-06-05
作者简介:林女贵(1974-),女,硕士,高级经济师,从事电力营销方面研究,E-mail:13805063176@139.com;徐启峰(1959-),男,通信作者,博士,教授,从事智能电网测量新技术研究,E-mail:ranger123098@163.com
基金资助:
国家自然科学基金资助项目(51977038)；国网福建省电力有限公司科技项目(52130419000Y)

Abnormal Electricity Consumption Behaviors Detection Based on Improved Deep Auto-Encoder

LIN Nvgui¹, HONG Lanxiu², HUANG Daoshan³, YI Yang², LIU Zhixuan³, XU Qifeng⁴

1. State Grid Fujian Electric Power Company Limited, Fuzhou 350003, China;
2. State Grid Fujian Economics and Technology Institute, Fuzhou 350012, China;
3. State Grid Fujian Electric Power Research Institute, Fuzhou 350007, China;
4. College of Electric Engineering and Automation, Fuzhou University, Fuzhou 350116, China

Received:2019-10-10 Revised:2020-02-13 Published:2020-06-05
Supported by:
This work is supported by National Natural Science Foundation of China(No.51977038), Science and Technology Project of State Grid Fujian Electric Power Company (No.52130419000Y)

摘要/Abstract

摘要： 为准确检测异常用电行为以降低电力公司的运营成本，提出一种基于改进深度自编码网络的异常用电行为辨识方法。首先将正常用户的用电数据作为训练样本，自编码网络逐层学习数据的有效特征；然后重构输入数据以计算检测阈值，而由于异常用电行为破坏数据的特征规则，再通过对比重构误差与检测阈值的差异即可实现异常用电行为辨识。为了改善自编码网络的特征提取能力与鲁棒性，分别引入了稀疏约束和噪声编码，并利用粒子群算法优化网络的超参数以提高模型的学习效率和泛化能力。选用福建省某地区居民用电和商业用电数据集进行了验证，这一模型的异常行为检测的准确率高于92%。实验表明所提方法具有优异的特征提取能力和异常用电行为辨识能力。

关键词: 异常用电, 自编码网络, 稀疏约束, 噪声, 特征提取, 数据重构

Abstract: In order to accurately detect the abnormal electricity consumption behaviors for reducing the operating costs of power companies, a detection method of abnormal electricity consumption behaviors is proposed based on the improved deep auto-encoder (DAE). Firstly, the data of normal electricity users are employed as training samples, and the effective features of the data are automatically extracted by AE; and then the data is reconstructed to calculate the detection threshold. Because the effective data characteristics are destroyed by the abnormal behaviors, the abnormal behaviors can be detected through comparing the difference between the reconstruction error and the detection threshold. To improve the feature extraction ability and the robustness of AE network, the sparse restrictions and the noise coding are introduced into the auto-encoder, and the hyper-parameters of AE network are optimized through the particle swarm optimization algorithm to improve the learning efficiency and generalization ability. The proposed model is validated by the electricity consumption dataset of domestic and business users of a region in Fujian province, and the abnormal detection accuracy is higher than 92%, which indicates that the proposed method has a powerful ability in feature extraction and abnormal behavior detection.

Key words: abnormal electricity consumption behavior, auto-encoder, sparse restriction, noise, feature extraction, data reconstruction

林女贵, 洪兰秀, 黄道姗, 易扬, 刘智煖, 徐启峰. 基于改进深度自编码网络的异常用电行为辨识[J]. 中国电力, 2020, 53(6): 18-26.

LIN Nvgui, HONG Lanxiu, HUANG Daoshan, YI Yang, LIU Zhixuan, XU Qifeng. Abnormal Electricity Consumption Behaviors Detection Based on Improved Deep Auto-Encoder[J]. Electric Power, 2020, 53(6): 18-26.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201910005

https://www.electricpower.com.cn/CN/Y2020/V53/I6/18

参考文献

[1] LEITE J B, SANCHES MANTOVANI J R. Detecting and locating non-technical losses in modern distribution networks[J]. IEEE Transactions on Smart Grid, 2018, 9(2): 1023-1032.
[2] NAGI J, YAP K S, TIONG S K, et al. Improving SVM-based nontechnical loss detection in power utility using the fuzzy inference system[J]. IEEE Transactions on Power Delivery, 2011, 26(2): 1284-1285.
[3] AHMAD T, CHEN H X, WANG J Y, et al. Review of various modeling techniques for the detection of electricity theft in smart grid environment[J]. Renewable and Sustainable Energy Reviews, 2018, 82(3): 2916-2933.
[4] 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
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
[5] 许刚, 谈元鹏, 戴腾辉. 稀疏随机森林下的用电侧异常行为模式检测[J]. 电网技术, 2017, 41(6): 1973-1982
XU Gang, TAN Yuanpeng, DAI Tenghui. Sparse random forest-based abnormal behavior pattern detection of electric power user side[J]. Power System Technology, 2017, 41(6): 1973-1982
[6] 胡殿刚, 李韶瑜, 楼俏, 等. ELM算法在用户用电行为分析中的应用[J]. 计算机系统应用, 2016, 25(8): 155-161
HU Diangang, LI Shaoyu, LOU Qiao, et al. Application of ELM algorithm in the analysis of customer electrical behavior[J]. Computer Systems & Applications, 2016, 25(8): 155-161
[7] 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.
[8] 杨斌. 基于聚类的异常检测技术的研究[D]. 长沙: 中南大学, 2008.
YANG Bin. Research on cluster-based anomaly detection technology[D]. Changsha: Central South University, 2008.
[9] PASSOS JÚNIOR L A, OBA RAMOS C C, RODRIGUES D, et al. Unsupervised non-technical losses identification through optimum-path forest[J]. Electric Power Systems Research, 2016, 140: 413-423.
[10] 王桂兰, 周国亮, 赵洪山, 等. 大规模用电数据流的快速聚类和异常检测技术[J]. 电力系统自动化, 2016, 40(24): 27-33
WANG Guilan, ZHOU Guoliang, ZHAO Hongshan, et al. Fast clustering and anomaly detection technique for large-scale power data stream[J]. Automation of Electric Power Systems, 2016, 40(24): 27-33
[11] 庄池杰, 张斌, 胡军, 等. 基于无监督学习的电力用户异常用电模式检测[J]. 中国电机工程学报, 2016, 36(2): 379-387
ZHUANG Chijie, ZHANG Bin, HU Jun, et al. Anomaly detection for power consumption patterns based on unsupervised learning[J]. Proceedings of the CSEE, 2016, 36(2): 379-387
[12] 田力, 向敏. 基于密度聚类技术的电力系统用电量异常分析算法[J]. 电力系统自动化, 2017, 41(5): 64-70
TIAN Li, XIANG Min. Abnormal power consumption analysis based on density-based spatial clustering of applications with noise in power systems[J]. Automation of Electric Power Systems, 2017, 41(5): 64-70
[13] 李娜, 张文月, 陈国平, 等. 基于用户负荷特性的电价交叉补贴测算模型[J]. 中国电力, 2019, 52(5): 148-154
LI Na, ZHANG Wenyue, CHEN Guoping, et al. Electricity price cross subsidy calculation model considering load characteristics of electricity consumers[J]. Electric Power, 2019, 52(5): 148-154
[14] 孙毅, 李世豪, 崔灿, 等. 基于高斯核函数改进的电力用户用电数据离群点检测方法[J]. 电网技术, 2018, 42(5): 1595-1606
SUN Yi, LI Shihao, CUI Can, et al. Improved outlier detection method of power consumer data based on Gaussian kernel function[J]. Power System Technology, 2018, 42(5): 1595-1606
[15] 胡天宇, 郭庆来, 孙宏斌. 基于堆叠去相关自编码器和支持向量机的窃电检测[J]. 电力系统自动化, 2019, 43(1): 119-125
HU Tianyu, GUO Qinglai, SUN Hongbin. Nontechnical loss detection based on stacked uncorrelating autoencoder and support vector machine[J]. Automation of Electric Power Systems, 2019, 43(1): 119-125
[16] 张承智, 肖先勇, 郑子萱. 基于实值深度置信网络的用户侧窃电行为检测[J]. 电网技术, 2019, 43(3): 1083-1091
ZHANG Chengzhi, XIAO Xianyong, ZHENG Zixuan. Electricity theft detection for customers in power utility based on real-valued deep belief network[J]. Power System Technology, 2019, 43(3): 1083-1091
[17] 张小斐, 耿俊成, 孙玉宝. 图正则非线性岭回归模型的异常用电行为识别[J]. 计算机工程, 2018, 44(6): 8-12
ZHANG Xiaofei, GENG Juncheng, SUN Yubao. Abnormal electricity behavior recognition of graph regularization nonlinear ridge regression model[J]. Computer Engineering, 2018, 44(6): 8-12
[18] 郭志民, 袁少光, 孙玉宝. 基于L0稀疏超图半监督学习的异常用电行为识别[J]. 计算机应用与软件, 2018, 35(2): 54-59
GUO Zhiminx, YUAN Shaoguang, SUN Yubao. Abnormal electricity power consumption recognition based on L0 sparse hypergraph semi-supervised learning[J]. Computer Applications and Software, 2018, 35(2): 54-59
[19] 袁静, 章毓晋. 融合梯度差信息的稀疏去噪自编码网络在异常行为检测中的应用[J]. 自动化学报, 2017, 43(4): 604-610
YUAN Jing, ZHANG Yujin. App1ication of sparse denoising auto encoder network with gradient difference information for abnormal action detection[J]. Acta Automatica Sinica, 2017, 43(4): 604-610
[20] 赵洪山, 刘辉海, 刘宏杨, 等. 基于堆叠自编码网络的风电机组发电机状态监测与故障诊断[J]. 电力系统自动化, 2018, 42(11): 102-108
ZHAO Hongshan, LIU Huihai, LIU Hongyang, et al. Condition monitoring and fault diagnosis of wind turbine generator based onStacked autoencoder network[J]. Automation of Electric Power Systems, 2018, 42(11): 102-108
[21] NG A. Sparse autoencoder[J]. CS294A Lecture Notes, 2011, 72: 1-19.
[22] VINCENT P, LAROCHELLE H, LAJOIE I, et al. Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion[J]. Journal of Machine Learning Research, 2010, 11(12): 3371-3408.
[23] 姜锐, 滕伟, 刘潇波, 等. 风电机组发电机轴承电腐蚀故障的分析诊断[J]. 中国电力, 2019, 52(6): 128-133
JIANG Rui, TENG Wei, LIU Xiaobo, et al. Diagnosis of electrical corrosion fault in wind turbine generator bearing based on vibration signal analysis[J]. Electric Power, 2019, 52(6): 128-133
[24] 王鹏, 张朋宇, 高亚静, 等. 监管视角下的电力市场用户分类指标体系及算法研究[J]. 中国电力, 2018, 51(12): 139-148
WANG Peng, ZHANG Pengyu, GAO Yajing, et al. Research on index system and algorithm of customer classification in electricity market from the regulatory perspective[J]. Electric Power, 2018, 51(12): 139-148
[25] 林顺富, 黄娜娜, 赵伦加, 等. 基于用户行为的家庭日负荷曲线模型[J]. 电力建设, 2016, 37(10): 114-121
LIN Shunfu, HUANG Nana, ZHAO Lunjia, et al. Domestic daily load curve modeling based on user behavior[J]. Electric Power Construction, 2016, 37(10): 114-121
[26] 姚历毅, 罗萍萍, 项胤兴, 等. 具有抗逆序及权重自适应的黑启动方案评估方法[J]. 中国电力, 2019, 52(3): 92-99
YAO Liyi, LUO Pingping, XIANG Yinxing, et al. Evaluation method of black start scheme with anti-reverse order and weight adaptive[J]. Electric Power, 2019, 52(3): 92-99
[27] GOODFELLOW I, BENGIO Y, AARON C. Deep learning[M]. Cambridge, MA, USA: MIT Press, 2016.
[28] LAROCHELLE H, BENGIO Y, LOURADOUR J, et al. Exploring strategies for training deep neural networks[J]. Journal of Machine Learning Research, 2009, 1(10): 1-40.
[29] 蔡坤宝, 王成良, 陈曾汉. 产生标准高斯白噪声序列的方法[J]. 中国电机工程学报, 2004, 24(12): 207-211
CAI Kunbao, WANG Chengliang, CHEN Zenghan. A method for generating standard gaussian white noise sequences[J]. Proceedings of the CSEE, 2004, 24(12): 207-211
[30] HU T Y, GUO Q L, SHEN X W, et al. Utilizing unlabeled data to detect electricity fraud in AMI: a semisupervised deep learning approach[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(11): 3287-3299.
[31] 朱栋. 典型负荷用电行为模式分析方法及其应用研究[D]. 南京: 东南大学, 2017.
ZHU Dong. Analysis and application of typical load electricity behavior model[D]. Nanjing: Southeast University, 2017.

基于改进深度自编码网络的异常用电行为辨识

Abnormal Electricity Consumption Behaviors Detection Based on Improved Deep Auto-Encoder

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	张玉敏, 张涌琛, 叶平峰, 吉兴全, 石春友, 蔡富东, 李一宸. 基于集合经验模态分解的增强核岭回归配电系统状态估计[J]. 中国电力, 2024, 57(9): 156-168.
[2]	郑鹏, 韩鹏程, 王国栋, 娄颖. 中压配电线路断线高阻接地故障精细化诊断方法[J]. 中国电力, 2024, 57(4): 220-228.
[3]	张博智, 张茹, 焦东翔, 王龙宇, 周一凡, 周丽霞. 基于VMD-SAST的电能质量扰动分类识别方法[J]. 中国电力, 2024, 57(2): 34-40.
[4]	李丹, 梁云嫣, 缪书唯, 方泽仁, 胡越, 贺帅. 基于高斯混合聚类和改进条件变分自编码的多风电场功率日场景生成方法[J]. 中国电力, 2024, 57(12): 17-29.
[5]	马海飞, 滕伟, 彭迪康, 柳亦兵, 靳涛. 基于DRS与改进Autogram的风电齿轮箱复合故障特征提取[J]. 中国电力, 2023, 56(10): 71-79.
[6]	胡京, 邓颖, 蒋兴良, 曾蕴睿. 输电线路覆冰垭口微地形的特征提取与识别方法[J]. 中国电力, 2022, 55(8): 135-142.
[7]	王帅, 李琳, 崔建业, 赵寿生, 张鹏宁, 王亚琦, 何强. 滤波电容器心子中介质薄膜表面残余电荷对其振动影响的分析[J]. 中国电力, 2022, 55(2): 98-107,151.
[8]	樊江川, 于昊正, 刘慧婷, 杨丽君, 安佳坤. 基于多分支门控残差卷积神经网络的短期电力负荷预测[J]. 中国电力, 2022, 55(11): 155-162,174.
[9]	黄冬梅, 王玥琦, 胡安铎, 孙锦中, 时帅, 孙园, 房岭锋. 融合多维度特征的绝缘子状态边缘识别方法[J]. 中国电力, 2022, 55(1): 133-141.
[10]	周雪松, 王成龙, 马幼捷, 吴博宁. 抑制量测噪声的大功率异步电动机自抗扰矢量控制[J]. 中国电力, 2021, 54(9): 24-33.
[11]	陈智雄, 曾鸿海, 韩东升. 全双工和半双工自适应的电力线中继和功率优化分配算法[J]. 中国电力, 2021, 54(6): 199-207.
[12]	黄明祥, 郭志彬, 潘立志, 李学宝. 实验室内交流导线电晕噪声的频谱特性研究[J]. 中国电力, 2021, 54(5): 111-120.
[13]	徐艳春, 赵彩彩, 孙思涵, MI Lu. 基于改进LMD和能量相对熵的主动配电网故障定位方法[J]. 中国电力, 2021, 54(11): 133-143.
[14]	刘文慧, 徐遵义, 张旭冉, 张海燕. 基于互信息和PCA理论的湿法烟气脱硫工况特征提取方法[J]. 中国电力, 2020, 53(8): 158-163.
[15]	樊超, 聂京凯, 肖伟民, 王广周, 汲盛昌, 杨富尧, 刘洋, 杜鹏宇. 电力变压器铁心、油箱振动特性及与近场噪声的关联性研究[J]. 中国电力, 2020, 53(10): 34-41.

模态框（Modal）标题

基于改进深度自编码网络的异常用电行为辨识

Abnormal Electricity Consumption Behaviors Detection Based on Improved Deep Auto-Encoder

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics