基于改进旋转森林算法的窃电检测研究
CSTR:
作者:
作者单位:

(上海电力大学电气工程学院,上海 200090)

通讯作者:

梅智聪(1998—),男,硕士研究生,主要从事电力系统窃电检测方向的研究;E?mail:1762860273@qq.com

中图分类号:

TM731

基金项目:

国家自然科学基金(61873159)


Research on electricity theft detection based on improved rotation forest algorithm
Author:
Affiliation:

(College of Electrical Engineering, Shanghai University of Electric Power, Shanghai 200090, China)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [31]
  • | | | |
  • 文章评论
    摘要:

    如何准确检测出用户侧窃电行为是长期存在于各供电公司一个难点,传统的窃电检测方案均存在一定的局限性。针对窃电检测领域正负类样本高度不平衡,且单一分类模型表现不佳的问题,提出一种基于改进旋转森林算法的窃电检测方法。旋转森林算法采用主成分分析(principal component analysis,PCA)进行特征提取,利用原始训练集的所有主成分训练每个基分类器。在经典的旋转森林算法基础上,使用改进合成少数类过采样(synthetic minority oversampling technique,SMOTE)算法平衡样本子集中的正负类样本;使用Bagging算法中的Bootstrap抽样对训练子集进一步抽样;按准确率对基分类器进行选择性集成等3个方面的改进。算例使用华东某地区实际用户数据,结果表明所提窃电检测方法对比单一分类模型和现有集成学习策略,在多项评价指标下均取得更好的效果。

    Abstract:

    Detecting user-side electricity theft accurately has long been a challenge for power supply companies, with traditional theft detection methods having certain limitations. Addressing the highly imbalanced positive and negative samples in the field of theft detection, and the poor performance of single classification models, this study proposes a theft detection method based on an improved Rotation Forest algorithm. The Rotation Forest algorithm uses Principal Component Analysis (PCA) for feature extraction, training each base classifier with all principal components of the original training set. Building upon the classical Rotation Forest algorithm, improvements are made in three aspects: balancing the positive and negative samples in the subset using the Synthetic Minority Oversampling Technique (SMOTE) algorithm, further sampling the training subset using Bootstrap sampling in the Bagging algorithm, and selectively integrating base classifiers based on accuracy. A case study using actual user data from a region in East China demonstrates that the proposed theft detection method achieves better results in multiple evaluation metrics compared to single classification models and existing ensemble learning strategies.

    参考文献
    [1] 杨志东,丁建武,陈广久,等.基于LightGBM和LSTM模型的电力大数据异常用电检测方法研究[J/OL].电测与仪表:1-7[2024-04-03].http://kns.cnki.net/kcms/detail/23.1202.TH.20220713.1958.004.html. YANG Zhidong,DING Jianwu,CHEN Guangjiu,et al.Research on abnormal electricity consumption detection method for power big data based on lightGBM and LSTM models[J].Electrical Measurement & Instrumentation:1-7[2023-04-03].http://kns.cnki.net/kcms/detail/23.1202.TH.20220713.1958.004.html.
    [2] 俞林刚,李铭,伍栋文,等.配电网拓扑参数未知场景下中压用户窃电检测方法[J].电网与清洁能源,2023,39(9):91-100. YU Lingang,LI Ming,WU Dongwen,et al.An electricity theft detection method of medium voltage users in the case of unknown for distribution network topology parameters[J].Power System and Clean Energy,2023,39(9):91-100.
    [3] 李双伟,付慧,史明明,等.基于堆叠稀疏自编码器和深度森林的窃电检测模型[J].供用电,2023,40(5):77-83+99. LI Shuangwei,FU Hui,SHI Mingming,et al.Electricity theft detection model based on stacked sparse auto encoder and deep forest[J].Distribution & Utilization,2023,40(5):77-83+99.
    [4] 徐明杰,赵健,王小宇,等.基于多任务联合模型的居民用电模式分类方法[J].电工技术学报,2022,37(21):5490-5502. XU Mingjie,ZHAO Jian,WANG Xiaoyu,et al.Residential electricity consumption pattern classification method based on multi-task joint model[J].Transactions of China Electrotechnical Society,2022,37(21):5490-5502.
    [5] 鄢仁武,郑杨,林志雄,等.基于改进ECC的含电动汽车用户负荷信息保护方案[J].中国电力,2023,56(1):150-157. YAN Renwu,ZHENG Yang,LIN Zhixiong,et al.ECC based load information protection scheme for electric vehicle users[J].Electric Power,2023,56(1):150-157.
    [6] 张衡,张沈习,程浩忠,等.Stackelberg博弈在电力市场中的应用研究综述[J].电工技术学报,2022,37(13):3250-3262. ZHANG Heng,ZHANG Shenxi,CHENG Haozhong,et al.A state-of-the-art review on Stackelberg game and its applications in power market[J].Transactions of China Electrotechnical,2022,37 (13):3250-3262.
    [7] 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. 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.
    [8] 陈光宇,徐嘉杰,卢兆军,等.基于相关性度量算法的台区线损异常判断及精准定位[J].电力工程技术,2022,41(4):67-74. CHEN Guangyu,XU Jiajie,LU Zhaojun,et al.Judgment and precise location of abnormal line loss in station area based on correlation measurement algorithm[J].Electric Power Engineering Technology,2022,41(4):67-74.
    [9] 薛阳,张蓬鹤,杨艺宁,等.基于线损协方差分析的群体性固定比例窃电行为检测方法[J].电力系统自动化,2022,46(13):112-120. XUE Yang,ZHANG Penghe,Yang Yining,et al.Detection method for group fixed-ratio electricity theft based on covariance analysis of line loss[J].Automation of Electric Power Systems,2022,46(13):112-120.
    [10] 武超飞,孙冲,刘厦,等.基于改进FCM聚类的窃电行为检测[J].电力科学与技术学报,2021,36(6):164-170. WU Chaofei,SUN Chong,LIU Sha,et al.Detection of stealing electricity energy based on improved fuzzy C-means clustering[J].Journal of Electric Power Science and Technology,2021,36(6):164-170.
    [11] 邹念,魏梅芳,苏盛,等.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. ZOU Nian,WEI Meifang,SU Sheng,et al.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.
    [12] 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. 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.
    [13] ABDULRAHMAN T,MUHAMMAD I,USMAN Z,et al.Robust electricity theft detection against data poisoning attacks in smart grids[J].IEEE Transactions on Smart Grid,2021,12(3):2675-2684.
    [14] 伍栋文,于艾清,俞林刚,等.基于ICS-K-means聚类算法和WNN的有源低压台区线损估算方法[J].智慧电力,2022,50(4):8-14. WU Dongwen,YU Aiqing,YU Lingang,et al.Line loss estimation method based on ICS-K-means clustering algorithm and WNN for transformer district with DGs[J].Smart Power,2022,50(4):8-14.
    [15] 李景歌,荣娜,陈庆超.基于生成对抗网络的分布式光伏窃电数据增强方法[J].电力科学与技术学报,2022,37(5):181-190. LI Jingge,RONG Na,CHEN Qingchao.A data augmentation method for distributed photovoltaic electricity theft using generative adversarial network[J].Journal of Electric Power Science and Technology,2022,37(5):181-190.
    [16] SINGH S K,BOSE R,JOSHI A.Entropy-based electricity theft detection in AMI network[J].IET Cyber-Physical Systems:Theory & Applications,2018,3(2):99-105.
    [17] BUZAU M M,TEJEDOR-AGUILERA J,CRUZROMERO P,et al.Detection of non-technical losses using smart meter data and supervised learning[J].IEEE Transactions on Smart Grid,2019,10(3):2661-2670.
    [18] MUHAMMAD I,SHAABAN M F,MAHESH N,et al.Deep learning detection of electricity theft cyber-attacks in renewable distributed generation[J].IEEE Transactions on Smart Grid,2020,11(4):3428-3437.
    [19] 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 Kang,LIU Xin,ZHANG Penghe,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.
    [20] 程超鹏,彭显刚,曾勇斌,等.相异模型下Stacking集成结构的异常用电用户识别方法[J].电网技术,2021,45(12):4828-4836. CHENG Chaopeng,PENG Xiangang,ZENG Yongbin,et al.An abnormal power user recognition method for stacking integrated structures with different models[J].Power System Technology,2021,45(12):4828-4836.
    [21] 刘钊瑞,高云鹏,郭建波,等.基于深度自编码器高斯混合模型的窃电行为检测[J].电力系统保护与控制,2022,50(18):92-102. LIU Zhaorui,GAO Yunpeng,GUO Jianbo,et al.Abnormal detection of electricity theft using a deep auto-encoder Gaussian mixture model[J].Power System Protection and Control,2022,50(18):92-102.
    [22] 游文霞,李清清,杨楠,等.基于多异学习器融合Stacking集成学习的窃电检测[J].电力系统自动化,2022,46(24):178-186. YOU Wenxia,LI Qingqing,YANG Nan,et al.Electricity theft detection based on multiple different learners fusion by stacking ensemble learning[J].Automation of Electric Power Systems,2022,46(24):178-186.
    [23] 李国成,陆俊,王赟,等.基于Bagging二次加权集成的孤立森林窃电检测算法[J].电力系统自动化,2022,46(2):92-100. LI Guocheng,LU Jun,WANG Yun.Electricity theft detection algorithm of isolation forest based on bagging secondary weighted ensemble[J].Automation of Electric Power Systems,2022,46(2):92-100.
    [24] RODRIGUEZ J J,KUNCHEVA L I,ALONSO C J.Rotation forest:A new classifier ensemble method[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(10):1619-1630.
    [25] YAO D,WEN M,LIANG X,et al.Energy theft detection with energy privacy preservation in the smart grid[J].IEEE Internet of Things Journal,2019,6(5):7659-7669.
    [26] 夏云舒,王勇,周林,等.基于改进生成对抗网络的虚假数据注入攻击检测方法[J].电力建设,2022,43(3):58-65. XIA Yunshu,WANG Yong,ZHOU Lin,et al.False data injection attack detection method based on improved generative adversarial network[J].Electric Power Construction,2022,43(3):58-65.
    [27] MART NEZ-MU OZ G,HERN NDEZ-LOBATO D,SU REZ A.An analysis of ensemble pruning techniques based on ordered aggregation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(2):245-259.
    [28] MARCELO Z,EDGARD J,MARCELO P,et al.A tunable fraud detection system for advanced metering infrastructure using short-lived patterns[J].IEEE Transactions on Smart Grid,2019,10(1):830-840.
    [29] 潘骏,夏祥武,李梁,等.基于关联潮流感知与高斯混合模型的异常用电检测[J].电力建设,2023,44(11):138-148. PAN Jun,XIA Xiangwu,LI Liang,et al.Abnormal power consumption detection based on associative power flow sensing and Gaussian mixture model[J].Electric Power Construction,2023,44(11):138-148.
    [30] ZHENG Z,YANG Y,NIU X,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.
    [31] 徐耀松,李佳旺,段彦强.基于相似度机制AdaBoost-DBN的变压器故障层级诊断[J].高压电器,2023,59(6):154-164. XU Yaosong,LI Jiawang,DUAN Yanqiang.Fault Hierarchical Diagnosis of Transformer Based on AdaBoost-DBN Similarity Mechanism[J].High Voltage Apparatus,2023,59(6):154-164.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

刘建锋,梅智聪,刘梦琪,等.基于改进旋转森林算法的窃电检测研究[J].电力科学与技术学报,2024,(1):93-104.
LIU Jianfeng, MEI Zhicong, LIU Mengqi, et al. Research on electricity theft detection based on improved rotation forest algorithm[J]. Journal of Electric Power Science and Technology,2024,(1):93-104.

复制
分享
文章指标
  • 点击次数:160
  • 下载次数: 580
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 在线发布日期: 2024-04-22
文章二维码