基于BRB和LSTM网络的电力大数据用电异常检测方法

doi:10.12204/j.issn.1000-7229.2021.08.005

电力建设 ›› 2021, Vol. 42 ›› Issue (8): 38-45.doi: 10.12204/j.issn.1000-7229.2021.08.005

• 能源互联网人工智能关键技术及其应用·栏目主持刘友波副教授、胡伟副教授、王迎新主任、顾雨嘉高级工程师· • 上一篇下一篇

基于BRB和LSTM网络的电力大数据用电异常检测方法

万磊¹, 陈成², 黄文杰³, 卢涛², 刘威²

1.国网湖北省电力有限公司,武汉市 430077
2.武汉工程大学计算机科学与工程学院,武汉市 430205
3.湖北华中电力科技开发有限责任公司,武汉市 430077

收稿日期:2020-11-06 出版日期:2021-08-01 发布日期:2021-07-30
作者简介:万磊(1964),男,硕士,高级工程师,主要研究方向为大规模流数据集成与分析等;
陈成(1998),男,硕士研究生,主要研究方向为深度学习及大数据分析等;
黄文杰(1987),男,高级项目经理,主要研究方向为服务计算、大规模流数据集成与分析等;
卢涛(1980),男,博士,教授,主要研究方向为图像处理和深度学习等;
刘威(1987),男,博士,讲师,通信作者,主要研究领域为深度学习、风险评价及模式识别等。
基金资助:
国网湖北省电力有限公司科技项目“湖北电网运营监控实用化关键技术研究”(XM052014111);湖北省教育厅科学技术研究项目(Q20201507)

Power Abnormity Detection Method Based on Power Big Data Applying BRB and LSTM Network

WAN Lei¹, CHEN Cheng², HUANG Wenjie³, LU Tao², LIU Wei²

1. State Grid Hubei Electric Power Co., Ltd., Wuhan 430077, China
2. School of Computer Science and Engineering, Wuhan Institute of Technology, Wuhan 430205, China
3. Hubei Huazhong Electric Power Technology Development Co., Ltd., Wuhan 430077, China

Received:2020-11-06 Online:2021-08-01 Published:2021-07-30
Supported by:
Science and Technology Project of State Grid Hubei Electric Power Co., Ltd.(XM052014111);Science and Technology Research Project of the Hubei Provincial Department of Education(Q20201507)

摘要/Abstract

摘要：

为了降低电力公司的运营成本,针对非技术性损失(non-technical loss, NTL),提出一种基于置信规则推理(belief rule-based,BRB)和长短记忆网络模型(long short-term memory,LSTM)的用户窃电行为诊断方法。该方法首先从用电量大数据中提取电量波动系数和用电量曲线的毛刺宽度两种用电异常特征,制定BRB异常特征输入前置属性转换,并通过证据推理(evidential reasoning,ER)方法输出最终的置信度,建立适用于NTL异常检测的置信规则库,从而自动获取具有高鲁棒性的有标签正负样本训练数据集。接着,以此为基础,提出构建一种多LSTM网络检测模型,实现对异常用电特征的有效提取与检测。实验结果表明,与现有的主流网络故障检测模型相比,所提方法能够更准确地从电力大数据中诊断出用户的异常用电行为。

关键词: 电力大数据, 非技术性损失(NTL)用电异常, 长短记忆网络模型(LSTM), 置信规则推理(BRB), 证据推理(ER)方法

Abstract:

In order to reduce the operating cost of power companies, for non-technical loss (NTL), this paper proposes a belief rule-based (BRB) and long short-term memory (LSTM) based method for diagnosing the user’s electricity theft behavior. This method firstly extracts the power fluctuation coefficient and the burr width of the power consumption curve from the electric power dataset, and then the input pre-attribute conversion method for BRB anomaly feature is established. The final confidence is output by evidential reasoning (ER) method, and the confidence-rule base for NTL anomaly detection is established, thus, the labeled positive and negative training data sets with high robustness can be obtained automatically. Then, on this basis, a multi-LSTM network detection model is proposed to realize the effective extraction and detection of abnormal electrical features. The experimental results show that, compared with existing mainstream fault detection networks, the proposed method can better accurately diagnose the abnormal electrical behavior of users from the power big data.

Key words: power big data, abnormal non-technical loss (NTL) electricity behavior, long short-term memory (LSTM), belief rule-based (BRB), evidential reasoning (ER) method

中图分类号:

TM73
TP277

万磊, 陈成, 黄文杰, 卢涛, 刘威. 基于BRB和LSTM网络的电力大数据用电异常检测方法[J]. 电力建设, 2021, 42(8): 38-45.

WAN Lei, CHEN Cheng, HUANG Wenjie, LU Tao, LIU Wei. Power Abnormity Detection Method Based on Power Big Data Applying BRB and LSTM Network[J]. ELECTRIC POWER CONSTRUCTION, 2021, 42(8): 38-45.

图/表 11

图1 用户异常用电曲线

Fig.1 Abnormal power consumption curve

图2 总体流程

Fig.2 Overall flow chart

图3 LSTM结构

Fig.3 Structure of LSTM

图4 基于BRB和LSTM网络的NTL异常用电检测总体流程

Fig.4 Overall flow chart of NTL abnormal power consumption detection based on BRB and LSTM network

表1 用电训练样本集

Table 1 Training sample set of electric power consumptionkW·h

表2 用电测试样本集

Table 2 Test sample set of electric power consumption

图5 测试样本每月电量波动系数

Fig.5 Monthly power fluctuation coefficient of test samples

图6 测试样本每月毛刺宽度系数

Fig.6 Monthly burr width coefficient of test sample

表3 训练后的置信规则库

Table 3 Confidence-rule base after training

表4 本文所提方法与其他方法的检测准确率对比

Table 4 Accuracy of the proposed method compared with other methods

表5 2016年上半年国家电网公司用户用电数据[4]异常检测实验结果

Table 5 Experimental results of abnormality detection of power consumption data of state grid users in the first half of 2016[4]

参考文献 18

[1]	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.
[2]	于小青, 齐林海. 基于流数据聚类算法的电力大数据异常检测[J]. 电力信息与通信技术, 2020, 18(3): 8-14.
	YU Xiaoqing, QI Linhai. Power big data anomaly detection based on stream data clustering algorithm[J]. Electric Power Information and Communication Technology, 2020, 18(3): 8-14.
[3]	卢峰, 丁学峰, 尹小明, 等. 基于样本优化选取的支持向量机窃电辨识方法[J]. 计算机测量与控制, 2018, 26(6): 223-226.
	LU Feng, DING Xuefeng, YI Xiaoming, et al. Electricity theft identification using support vector machine based on sample optimization selection[J]. Computer Measurement & Control, 2018, 26(6): 223-226.
[4]	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: 10.1109/TII.2017.2785963 URL
[5]	沈海涛, 秦靖雅, 陈浩, 等. 电力用户用电数据的异常数据审查和分类[J]. 电力与能源, 2016, 37(1): 17-22.
	SHEN Haitao, QIN Jingya, CHEN Hao, et al. Anomaly detection and category of electrical utilization data[J]. Power & Energy, 2016, 37(1): 17-22.
[6]	李彬, 王红卫, 杨剑波, 等. 基于置信规则推理的库存控制方法[J]. 华中科技大学学报(自然科学版), 2011, 39(7): 76-79.
	LI Bin, WANG Hongwei, YANG Jianbo, et al. Belief-rule-based inference method for inventory control[J]. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2011, 39(7): 76-79.
[7]	YANG J B, LIU J, WANG J, et al. Belief rule-base inference methodology using the evidential reasoning Approach-RIMER[J]. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 2006, 36(2): 266-285. doi: 10.1109/TSMCA.2005.851270 URL
[8]	YANG J B, XU D L. Evidential reasoning rule for evidence combination[J]. Artificial Intelligence, 2013, 205: 1-29. doi: 10.1016/j.artint.2013.09.003 URL
[9]	MIKOLOV T, KOMBRINK S, BURGET L, et al. Extensions of recurrent neural network language model[C]// 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Prague, Czech Republic: IEEE, 2011: 5528-5531.
[10]	HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. doi: 10.1162/neco.1997.9.8.1735 URL
[11]	耿坤, 吕枫. 基于PCA/LSTM的海底观测网电力系统供电海缆故障定位[J]. 海洋技术学报, 2020, 39(3): 22-29.
	GENG Kun, LYU Feng. PCA/LSTM-based submarine cable fault location approach for seafloor observatory network power systems[J]. Journal of Ocean Technology, 2020, 39(3): 22-29.
[12]	ZHAO Z, CHEN W H, WU X M, et al. LSTM network: A deep learning approach for short-term traffic forecast[J]. IET Intelligent Transport Systems, 2017, 11(2): 68-75. doi: 10.1049/iet-its.2016.0208 URL
[13]	DE BRUIN T, VERBERT K, BABUŠKA R. Railway track circuit fault diagnosis using recurrent neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(3): 523-533. doi: 10.1109/TNNLS.2016.2551940 URL
[14]	KINGMA D P, BA J. Adam: A method for stochastic optimization[C]// Proceedings of the 3th International Conference on Learning Representations. San Diego, CA: Elsevier, 2015:1-15
[15]	QI Y M, SHEN C Q, WANG D, et al. Stacked sparse autoencoder-based deep network for fault diagnosis of rotating machinery[J]. IEEE Access, 2017, 5: 15066-15079. doi: 10.1109/ACCESS.2017.2728010 URL
[16]	XU C, WANG G, LIU X G, et al. Health status assessment and failure prediction for hard drives with recurrent neural networks[J]. IEEE Transactions on Computers, 2016, 65(11): 3502-3508. doi: 10.1109/TC.2016.2538237 URL
[17]	CHEN Z Q, LI C, SÁNCHEZ R V. Multi-layer neural network with deep belief network for gearbox fault diagnosis[J]. Journal of Vibroengineering, 2015, 17(5): 2379-2392.
[18]	XIANG H, WANG D S. The application of genetic BP neural network and D-S evidence theory in the complex system fault diagnosis[J]. Lecture Notes in Electrical Engineering, 2012, 124:219-224.

基于BRB和LSTM网络的电力大数据用电异常检测方法

Power Abnormity Detection Method Based on Power Big Data Applying BRB and LSTM Network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 18

相关文章 4

编辑推荐

Metrics

本文评价

[1]	张磐, 郑悦, 李海龙, 刘航旭, 李国栋, 葛磊蛟. 配电网电力设备缺陷文本智能辨识运维综述[J]. 电力建设, 2022, 43(5): 90-99.
[2]	刘琪琛, 雷景生, 郝珈玮,黄燕刚,李强,罗海波. 基于Spark平台和并行随机森林回归算法的短期电力负荷预测[J]. 电力建设, 2017, 38(10): 84-.
[3]	尹晓华，胡楠，刘为，王萌，祁兵，李彬. OpenFlow在电力大数据平台中的应用[J]. 电力建设, 2015, 36(3): 43-48.
[4]	赵春晖，姜大为，崔灿，李彬. 大数据时代下的新型电网规划体系[J]. 电力建设, 2014, 35(10): 17-20.