基于混合神经网络的配电网用户窃电检测方法A detection method for electricity theft by distribution network users based on a hybrid neural network
成跃宇,成国锋
CHENG Yueyu,CHENG Guofeng
摘要(Abstract):
针对传统的基于一维用电量数据挖掘分析的用户窃电检测方法检测精度低的问题,提出了一种基于混合神经网络的配电网用户窃电检测方法。首先,为了增强正常用户与窃电用户用电量的特征差异性,采用MTF(马尔可夫变迁场)对一维用电量数据进行图变换,实现用电数据的二维化;同时,为提高模型的准确性及泛化性,引入了用户用电量档案数据。然后,采用混合神经网络分别对预处理后的二维用电图像、档案数据进行特征量提取及融合,以实现配电网用户窃电检测。最后,通过两组对比实验,验证所提方法的有效性和精确性。实验结果表明:与其他模型相比,基于混合神经网络在窃电识别的准确率、查全率及ROC(接受者操作特征)曲线下面积均有较大的提升,具有较好的识别性能。
Given the low accuracy of the traditional electricity theft detection method based on one-dimensional electricity consumption data mining and analysis, a detection method for electricity theft by distribution network users based on a hybrid neural network is proposed. Firstly, to enhance the characteristic difference between the electricity consumption of normal users and that of power theft users, the Markov transition field(MTF) is used to transform one-dimensional electricity consumption data into two-dimensional graphs. Moreover, to improve the accuracy and generalization of the model, profile data of users' electricity consumption is introduced. Then, the hybrid neural network is used to extract and fuse the feature quantities of the preprocessed two-dimensional electricity consumption graphs and profile data respectively to detect electricity theft by distribution network users. Finally, the effectiveness and accuracy of the proposed method are verified through two sets of comparison experiments. The experimental results show that the method based on a hybrid neural network is superior to other models in detection accuracy of electricity theft, recall rate, and AUROC(area under the receiver operating characteristics), and has higher detection performance.
关键词(KeyWords):
配电网;用户窃电检测;马尔可夫变迁场;混合神经网络
distribution network;electricity theft detection;MTF;hybrid neural network
基金项目(Foundation): 国网江苏省电力有限公司扬州供电分公司科技项目(63106022005)
作者(Author):
成跃宇,成国锋
CHENG Yueyu,CHENG Guofeng
DOI: 10.19585/j.zjdl.202311012
参考文献(References):
- [1]胡一伟,刘珊,黄浩.基于ARIMA和递归贝叶斯的窃电用户识别算法[J].电测与仪表,2022,59(6):196-200.HU Yiwei,LIU Shan,HUANG Hao. An energy theft identification approach based on ARIMA and recursive Bayesian[J].Electrical Measurement&Instrumentation,2022,59(6):196-200.
- [2]冯小峰,姚诚智,杨俊华.基于Stacking集成学习的窃电检测研究[J/OL].电测与仪表:1-12[2022-12-06].http://kns. cnki. net/kcms/detail/23.1202. th. 20221125.1422.002.html.Feng Xiaofeng,Yao Chengzhi,Yang Junhua.Research on Electricity Theft Detection Based on Stacking Ensemble Learning[J/OL].Electrical Measurement&Instrumentation:1-12[2022-12-06].http://kns.cnki.net/kcms/detail/23.1202.th.20221125.1422.002.html.
- [3]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.
- [4]殷锋,周绍军,漆翔宇,等.基于改进LOF算法的窃电检测方法研究[J].中南民族大学学报(自然科学版),2022,41(5):579-585.YIN Feng,ZHOU Shaojun,QI Xiangyu,et al. Research on detection method of electricity stealing based on improved LOF algorithm[J].Journal of South-Central Minzu University(Natural Science Edition),2022,41(5):579-585.
- [5]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.
- [6]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.
- [7]高昂,郑建勇,梅飞,等.基于三元组孪生网络的窃电检测算法[J].中国电机工程学报,2022,42(11):3975-3986.GAO Ang,ZHENG Jianyong,MEI Fei,et al.Electricity theft detection algorithm based on triplet network[J].Proceedings of the CSEE,2022,42(11):3975-3986.
- [8] YAN Z Z,WEN H.Electricity theft detection base on extreme gradient boosting in AMI[C]//2020 IEEE International Instrumentation and Measurement Technology Conference(I2MTC),May 25-28,2020,Dubrovnik,Croatia.IEEE,2020:1-6.
- [9]县国成,王永攀,高俊,等.基于ELM-SVM模型与电能计量大数据的窃电识别技术研究[J].智慧电力,2022,50(9):82-89.XIAN Guocheng,WANG Yongpan,GAO Jun,et al.Research on electricity stealing identification technology based on ELM-SVM model and big data of electric energy measurement[J].Smart Power,2022,50(9):82-89.
- [10]孙红彬.基于卷积神经网络的φ-OTDR模式识别与频率测量[D].北京:北京交通大学,2021.SUN Hongbin.φ-OTDR pattern recognition and frequency measurement based on convolutional neural network[D].Beijing:Beijing Jiaotong University,2021.
- [11]周赣,华济民,李铭钧,等.基于图转换和混合卷积神经网络的窃电检测方法[J].电力系统自动化,2022,46(19):78-86.ZHOU Gan,HUA Jimin,LI Mingjun,et al. Electricity theft detection method based on graph transformation and hybrid convolutional neural network[J]. Automation of Electric Power Systems,2022,46(19):78-86.
- [12]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 Mengnan,LI Hongjiao.Electricity theft detection based on deep convolutional neural network and support vector classification[J]. Computer Simulation,2022,39(6):92-97.
- [13]刘康,刘鑫,张蓬鹤,等.基于负荷尖峰特征LSTM自编码器的窃电识别方法[J].电力系统自动化,2023,47(2):96-104.LIU K,LIU X,ZHANG P H,et al.Identification method of electricity theft based on long short-term memory autoencoder with load peak features[J].Automation of Electric Power Systems,2023,47(2):96-104.
- [14]李春芳,张锐峰,贾璐,等.基于改进k-means算法和大数据分析的新型窃电识别模型与仿真[J].电子设计工程,2022,30(22):84-88.LI Chunfang,ZHANG Ruifeng,JIA Lu,et al.New model and simulation of electricity stealing identification based on improved k-means algorithm and big data analysis[J].Electronic Design Engineering,2022,30(22):84-88.
- [15]姜家国,郭曼利.基于MTF和DenseNet的滚动轴承故障诊断方法[J].工矿自动化,2022,48(9):63-68.JIANG Jiaguo,GUO Manli. Fault diagnosis method of rolling bearing based on MTF and DenseNet[J]. Journal of Mine Automation,2022,48(9):63-68.
- [16]夏睿,高云鹏,朱彦卿,等.基于SE-CNN模型的窃电检测方法研究[J].电力系统保护与控制,2022,50(20):117-126.XIA Rui,GAO Yunpeng,ZHU Yanqing,et al. Research on detection method of stealing electricity based on SECNN model[J]. Power System Protection and Control,2022,50(20):117-126.
- [17]刘家瑞,杨国田,王孝伟.基于孪生深度神经网络的风电机组故障诊断方法[J].系统仿真学报,2022,34(11):2348-2358.LIU Jiarui,YANG Guotian,WANG Xiaowei.Fault diagnosis method of wind turbine based on twin depth neural network[J].Journal of System Simulation,2022,34(11):2348-2358.
- [18]尹文哲,夏虹,彭彬森,等.基于CNN-SVM的核电厂轴承故障诊断方法[J].哈尔滨工程大学学报,2023,44(3):410-417.YIN W Z,XIA H,PENG B S,et al.Bearing fault diagnosis method in nuclear power plants based on CNN-SVM[J]. Journal of Harbin Engineering University,2023,44(3):410-417.
- [19]刘浪,时宏伟.基于注意力机制的CNN-LSTM的ADS-B异常数据检测[J].计算机系统应用,2023,32(4):94-103.LIU L,SHI H W.ADS-B anomaly detection based on attention mechanism for CNN-LSTM[J]. Computer Systems&Applications,2023,32(4):94-103.