基于实值深度置信网络的用户侧窃电行为检测
|
摘要
用户侧窃电行为造成的非技术性损失对电网企业危害重大,不仅会影响电力系统的供电质量,还会增加电网的运营成本。为了辅助电网公司提高用电稽查效率、管理用户规范化用电,提出了基于实值深度置信网络的用户侧窃电行为检测模型。实值深度置信网络具有提取抽象特征的功能,并通过前馈神经网络微调后可实现较高分类精度。为了优化实值深度置信网络因随机初始化产生的局部最优化问题,该模型通过萤火虫算法对网络参数全局寻优。针对用户窃电行为检测,该模型利用因子分析进行数据降维,利用随机欠采样和套索算法应对数据不平衡问题,并利用ROC(receiver operatingcharacteristiccurve)曲线选取该模型的检测阈值。最后仿真实验验证了所提出模型的有效性和精确性。
Non-technical losses cause great harm to power utility due to electricity theft, affecting power supply quality and increasing operating cost. In order to assist power utility to improve efficiency of electric inspection and normalize power consumption, an electricity theft detection model for customers is proposed in this paper based on real-valued deep belief network(RDBN). RDBN has the function of extracting abstract features, and high classification accuracy after fine-tuning by back propagation network. Facing to the local optimization problem in RDBN caused by random initialization of network parameters, the proposed model optimizes the network parameters with firefly algorithm. Aiming at electricity theft detection, factor analysis is used in the proposed model for data dimension reduction. Random under-sampling and LASSO are employed to deal with data imbalance, and the detection threshold is chosen through ROC curve. Simulink experiment verified effectiveness and accuracy of the proposed model.
Non-technical losses cause great harm to power utility due to electricity theft, affecting power supply quality and increasing operating cost. In order to assist power utility to improve efficiency of electric inspection and normalize power consumption, an electricity theft detection model for customers is proposed in this paper based on real-valued deep belief network(RDBN). RDBN has the function of extracting abstract features, and high classification accuracy after fine-tuning by back propagation network. Facing to the local optimization problem in RDBN caused by random initialization of network parameters, the proposed model optimizes the network parameters with firefly algorithm. Aiming at electricity theft detection, factor analysis is used in the proposed model for data dimension reduction. Random under-sampling and LASSO are employed to deal with data imbalance, and the detection threshold is chosen through ROC curve. Simulink experiment verified effectiveness and accuracy of the proposed model.
引文
[1]León C,Carlos N,Biscarri F,et al.Integrated expert system applied to the analysis of non-technical losses in power utilities[J].Expert Systems With Applications,2011,38(8):10274-10285.
[2]Depuru S,Wang L,Devabhaktuni V.Electricity theft:overview,issues,prevention and a smart meter based approach to control theft[J].Energy Policy,2011,39(2):1007-1015.
[3]Smith T B.Electricity theft:a comparative analysis[J].Energy Policy,2004,32(18):2067-2076.
[4]Han W,Xiao Y.NFD:A practical scheme to detect non-technical loss fraud in smart grid[C]//IEEE International Conference on Communications.2014:605-609.
[5]彭显刚,郑伟钦,林利祥,等.基于密度聚类和Fréchet判别分析的电价执行稽查方法[J].电网技术,2015,39(11):3195-3201.Peng Xiangang,Zheng Weiqin,Lin Lixiang,et al.A method to inspect the implementation of electricity price based on density clustering analysis and Fréchet discriminant analysis[J].Power System Technology,2015,39(11):3195-3201(in Chinese).
[6]郝然,艾芊,肖斐.基于多元大数据平台的用电行为分析构架研究[J].电力自动化设备,2017,37(8):20-27.Hao Ran,Ai Qian,Xiao Fei.Architecture based on multivariate big data platform for analyzing electricity consumption behavior[J].Electric Power Automation Equipment,2017,37(8):20-27(in Chinese).
[7]杨劲锋,肖勇,马千里,等.基于过程回声状态网络的电力用户用电模式分类研究[J].电网技术,2015,39(12):3563-3569.Yang Jinfeng,Xiao Yong,Ma Qianli,et al.Process echo state network for pattern classification of electricity customers[J].Power System Technology,2015,39(12):3563-3569(in Chinese).
[8]Cavdar I H.A solution to remote detection of illegal electricity usage via power line communications[J].IEEE Transactions on Power Delivery,2004,19(4):1663-1667.
[9]Rengaraju P,Pandian S R,Lung C H.Communication networks and non-technical energy loss control system for smart grid networks[C]//IEEE Innovative Smart Grid Technologies-Asia(ISGT ASIA2014).Kuala Lumpur,Malaysia:IEEE,2014:418-423.
[10]Nagi J,Yap K S,Tiong S K,et al.Nontechnical loss detection for metered customers in power utility using support vector machines[J].IEEE Transactions on Power Delivery,2010,25(2):1162-1171.
[11]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.
[12]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(in Chinese).
[13]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):1964-1971.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):1964-1971(in Chinese).
[14]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(in Chinese).
[15]Zhang X,Wang D.Deep learning based binaural speech separation in reverberant environments[J].IEEE/ACM Transactions on Audio,Speech&Language Processing,2017,25(5):1075-1084.
[16]Delfarah M,Wang D.Features for masking-based monaural speech separation in reverberant conditions[J].IEEE/ACM Transactions on Audio,Speech&Language Processing,2017,25(5):1085-1094,[17]蒋树强,闵巍庆,王树徽.面向智能交互的图像识别技术综述与展望[J].计算机研究与发展,2016,53(1):113-122.Jiang Shuqiang,Min Weiqin,Wang Shuhui.Survey and prospect of intelligent interaction-oriented image recognition techniques[J].Journal of Computer Research and Development,2016,53(1):113-122(in Chinese).
[18]Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2006,18(7):1527.
[19]代杰杰,宋辉,杨祎,等.基于深度信念网络的变压器油中溶解气体浓度预测方法[J].电网技术,2017,41(8):2737-2742.Dai Jiejie,Song Hui,Yang Yi,et al.Concentration prediction of dissolved gases in transformer oil based on deep belief networks[J].Power System Technology,2017,41(8):2737-2742(in Chinese).
[20]张新伯,唐炬,潘成,等.用于局部放电模式识别的深度置信网络方法[J],电网技术,2016,40(10):3272-3278.Zhang Xinbo,Tang Ju,Pan Cheng,et al.Research of partial discharge recognition based on deep belief nets[J].Power System Technology,2016,40(10):3272-3278(in Chinese).
[21]Taylor G W,Hinton G E,Roweis S T.Two distributed-state models for generating high-dimensional time series[J].Journal of Machine Learning Research,2011,12(2):1025-1068.
[22]Hinton G E.Training products of experts by minimizing contrastive divergence[J].Neural Computation,2014,14(8):1771-1800.
[23]Yang X S.Stochastic algorithms:foundations and applications[M].Springer Berlin Heidelberg,2009.
[24]杨宗霄,毛智杰,杨本渤,等.基于因子分析与神经网络的输电网安全评价[J].电网技术,2009,33(14):26-30.Yang Zongxiao,Mao Zhijie,Yang Benbo,et al.Safety assessments of transmission grid based on factor analysis and neural network[J].Power System Technology,2009,33(14):26-30(in Chinese).
[25]Fluss R,Faraggi D,Reiser B.Estimation of the Youden Index and its associated cutoff point[J].Biometrical Journal,2005,47(4):458-472.
[26]He H,Garcia E A.Learning from imbalanced data[J].IEEETransactions on Knowledge&Data Engineering,2009,21(9):1263-1284.
[27]Office of Energy Efficiency&Renewable Energy(EERE).Commercial and residential hourly load profiles for all TMY3 Locations in the United States[DB/OL].(2013-07-02)[2017-09-17].http://en.openei.org/doe-opendata/dataset/commercialand-residential-hourly-load-profiles-for-all-tmy3-locations-in-the-unit ed-states.
[2]Depuru S,Wang L,Devabhaktuni V.Electricity theft:overview,issues,prevention and a smart meter based approach to control theft[J].Energy Policy,2011,39(2):1007-1015.
[3]Smith T B.Electricity theft:a comparative analysis[J].Energy Policy,2004,32(18):2067-2076.
[4]Han W,Xiao Y.NFD:A practical scheme to detect non-technical loss fraud in smart grid[C]//IEEE International Conference on Communications.2014:605-609.
[5]彭显刚,郑伟钦,林利祥,等.基于密度聚类和Fréchet判别分析的电价执行稽查方法[J].电网技术,2015,39(11):3195-3201.Peng Xiangang,Zheng Weiqin,Lin Lixiang,et al.A method to inspect the implementation of electricity price based on density clustering analysis and Fréchet discriminant analysis[J].Power System Technology,2015,39(11):3195-3201(in Chinese).
[6]郝然,艾芊,肖斐.基于多元大数据平台的用电行为分析构架研究[J].电力自动化设备,2017,37(8):20-27.Hao Ran,Ai Qian,Xiao Fei.Architecture based on multivariate big data platform for analyzing electricity consumption behavior[J].Electric Power Automation Equipment,2017,37(8):20-27(in Chinese).
[7]杨劲锋,肖勇,马千里,等.基于过程回声状态网络的电力用户用电模式分类研究[J].电网技术,2015,39(12):3563-3569.Yang Jinfeng,Xiao Yong,Ma Qianli,et al.Process echo state network for pattern classification of electricity customers[J].Power System Technology,2015,39(12):3563-3569(in Chinese).
[8]Cavdar I H.A solution to remote detection of illegal electricity usage via power line communications[J].IEEE Transactions on Power Delivery,2004,19(4):1663-1667.
[9]Rengaraju P,Pandian S R,Lung C H.Communication networks and non-technical energy loss control system for smart grid networks[C]//IEEE Innovative Smart Grid Technologies-Asia(ISGT ASIA2014).Kuala Lumpur,Malaysia:IEEE,2014:418-423.
[10]Nagi J,Yap K S,Tiong S K,et al.Nontechnical loss detection for metered customers in power utility using support vector machines[J].IEEE Transactions on Power Delivery,2010,25(2):1162-1171.
[11]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.
[12]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(in Chinese).
[13]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):1964-1971.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):1964-1971(in Chinese).
[14]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(in Chinese).
[15]Zhang X,Wang D.Deep learning based binaural speech separation in reverberant environments[J].IEEE/ACM Transactions on Audio,Speech&Language Processing,2017,25(5):1075-1084.
[16]Delfarah M,Wang D.Features for masking-based monaural speech separation in reverberant conditions[J].IEEE/ACM Transactions on Audio,Speech&Language Processing,2017,25(5):1085-1094,[17]蒋树强,闵巍庆,王树徽.面向智能交互的图像识别技术综述与展望[J].计算机研究与发展,2016,53(1):113-122.Jiang Shuqiang,Min Weiqin,Wang Shuhui.Survey and prospect of intelligent interaction-oriented image recognition techniques[J].Journal of Computer Research and Development,2016,53(1):113-122(in Chinese).
[18]Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2006,18(7):1527.
[19]代杰杰,宋辉,杨祎,等.基于深度信念网络的变压器油中溶解气体浓度预测方法[J].电网技术,2017,41(8):2737-2742.Dai Jiejie,Song Hui,Yang Yi,et al.Concentration prediction of dissolved gases in transformer oil based on deep belief networks[J].Power System Technology,2017,41(8):2737-2742(in Chinese).
[20]张新伯,唐炬,潘成,等.用于局部放电模式识别的深度置信网络方法[J],电网技术,2016,40(10):3272-3278.Zhang Xinbo,Tang Ju,Pan Cheng,et al.Research of partial discharge recognition based on deep belief nets[J].Power System Technology,2016,40(10):3272-3278(in Chinese).
[21]Taylor G W,Hinton G E,Roweis S T.Two distributed-state models for generating high-dimensional time series[J].Journal of Machine Learning Research,2011,12(2):1025-1068.
[22]Hinton G E.Training products of experts by minimizing contrastive divergence[J].Neural Computation,2014,14(8):1771-1800.
[23]Yang X S.Stochastic algorithms:foundations and applications[M].Springer Berlin Heidelberg,2009.
[24]杨宗霄,毛智杰,杨本渤,等.基于因子分析与神经网络的输电网安全评价[J].电网技术,2009,33(14):26-30.Yang Zongxiao,Mao Zhijie,Yang Benbo,et al.Safety assessments of transmission grid based on factor analysis and neural network[J].Power System Technology,2009,33(14):26-30(in Chinese).
[25]Fluss R,Faraggi D,Reiser B.Estimation of the Youden Index and its associated cutoff point[J].Biometrical Journal,2005,47(4):458-472.
[26]He H,Garcia E A.Learning from imbalanced data[J].IEEETransactions on Knowledge&Data Engineering,2009,21(9):1263-1284.
[27]Office of Energy Efficiency&Renewable Energy(EERE).Commercial and residential hourly load profiles for all TMY3 Locations in the United States[DB/OL].(2013-07-02)[2017-09-17].http://en.openei.org/doe-opendata/dataset/commercialand-residential-hourly-load-profiles-for-all-tmy3-locations-in-the-unit ed-states.