Abstract：Aiming at the shortcomings of traditional power theft detection methods, which only use one-dimensional power, rely on manual features, and have low detection accuracy, an eletricity theft detection model based on multiscale residual attention network is proposed. The model is based on pyramidal convolution to fully extract multi-scale detail features, and introduces hybrid dilated convolutional attention residual network to improve the detection performance. In this paper, the proposed method is experimentally validated using the public dataset of the State Grid, and the results show that compared with the traditional logistic regression, support vector machine, random forest, and other models, the AUC, MAP, and F1 score indexes of the proposed model have achieved effective improvement.
[1] CHAUHAN A, RAJVANSHI S. "Non-Technical Losses in power system: a review" [C]//2013 International Conference on Power, Energy and Control (ICPEC). Dindigul, India, 2013:558-561. [2] 肖宇,叶志,黄瑞,等.窃电行为检测方法研究综述[J].电力科学与技术学报,2023,38(4):1-14. XIAO Y, YE Z, HUANG R, et al. Research review on detection methods of electric power theft behavior [J]. Journal of Electric Power Science and Technology,2023,38(4):1-14. [3] 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 Q X, ZHENG K D, KANG C Q, et al. Detection methods of abnormal power consumption: review and prospect [J]. Automation of Electric Power Systems,2018,42(17):189-199. [4] 刘林,祁兵,李彬,等.面向电力物联网新业务的电力通信网需求及发展趋势[J].电网技术,2020,44(8):3114-3130. LIU L, QI B, LI B,et al. Demand and development trend of electric power communication network for new services of electric power internet of things [J]. Power Grid Technology,2020,44(8):3114-3130. [5] 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 C J, ZHANG B, HU J, et al. Abnormal consumption pattern detection of power users based on unsupervised learning [J]. Proceedings of the CSEE,2016,36(2):379-387. [6] JUNIOR L A P, RAMOS C C O, RODRIGUES D, et al. Unsupervised non-technical losses identification through optimum-path forest[J]. Electric Power Systems Research, 2016, 140: 413-423. [7] ANGELOS E W S, SAAVEDRA O R, CORTES O A C, et al. Detection and identification of abnormalities in customer consumptions in power distribution systems[J]. IEEE Transactions on Power Delivery, 2011, 26(4): 2436-2442. [8] 卿柏元,陈珏羽,李金瑾,等.基于CNN-LG模型的窃电行为检测方法研究[J].湖南大学学报(自然科学版),2022,49(8):138-148. QING B Y, CHEN J Y, LI J J et al. Research on electric theft behavior detection method based on CNN-LG model [J]. Journal of Hunan University (Natural Science Edition),2022,49(8):138-148. [9] MONEDERO I, BISCARRI F, LEÓN C, et al. Detection of frauds and other non-technical losses in a power utility using Pearson coefficient, Bayesian networks and decision trees[J]. International Journal of Electrical Power & Energy Systems, 2012, 34(1): 90-98. [10] JOKAR P, ARIANPOO N, LEUNG V C M. Electricity theft detection in AMI using customers’ consumption patterns[J]. IEEE Transactions on Smart Grid, 2015, 7(1): 216-226. [11] 游文霞,申坤,杨楠,等.基于AdaBoost集成学习的窃电检测研究[J].电力系统保护与控制,2020,48(19):151-159. YOU W X, SHEN K, YANG N, et al. Research on electric theft detection based on AdaBoost ensemble learning [J]. Power System Protection and Control, 2019,48(19):151-159. [12] 游文霞,申坤,杨楠,等.基于Bagging异质集成学习的窃电检测[J].电力系统自动化,2021,45(2):105-113. YOU W X, SHEN K, YANG N, et al. Electric theft detection based on bagging heterogeneous ensemble learning [J]. Automation of Electric Power Systems, 2019,45(2):105-113. [13] 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 C P, Peng X G, ZENG Y B, et al. Identification method of abnormal power users in the stacking integrated structure under different models [J]. Power Grid Technology, 2021,45(12):4828-4836. [14] 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):92-97,429. ZHANG MN, Li HJ. Electric theft detection based on DCNN and SVC [J]. Computer Simulation,202,39(6):92-97,429. [15] 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. [16] FINARD P, CAMPIOTTI I, PLENSACK G, et al. Electricity theft detection with self-attention [DB/OL]. [2023-08-06].http://arxiv.org/pdf/2002.06219.
