Improved Outlier Detection Method of Power Consumer Data Based on Gaussian Kernel Function

被引：0

作者：

Sun Y. ^{[1
]}

Li S. ^{[1
]}

Cui C. ^{[1
]}

Li B. ^{[1
]}

Chen S. ^{[2
]}

Cui G. ^{[3
]}

机构：

[1] School of Electrical and Electronic Engineering, North China Electric Power University, Changping District, Beijing

[2] Beijing Key Laboratory of Demand Side Multi-Energy Carriers Optimization and Interaction Technique, China Electric Power Research Institute, Haidian District, Beijing

[3] State Grid Jiangsu Electric Power Research Institute, Nanjing, 210003, Jiangsu Province

来源：

Dianwang Jishu/Power System Technology | 2018年 / 42卷 / 05期

关键词：

Data mining; Gaussian kernel density local outlier factor; Outlier detection; Power big data; Power consumption behavior analysis;

D O I：

10.13335/j.1000-3673.pst.2017.1586

中图分类号：

学科分类号：

摘要：

In allusion to applicability of power consumer data outlier detection method in context of big data in smart power distribution and consumption systems, and high cost of obtaining abnormal samples for power consumption in actual data sets, an improved outlier detection method of power consumer data based on Gaussian kernel function was proposed. Firstly, the users were classified with fuzzy clustering method. Then various features of each type of users were extracted and PCA (principal components analysis) was used to reduce the dimension of feature vectors. Finally, Gaussian kernel function was used to improve local outlier factor (LOF) algorithm, and Gaussian kernel density local outlier factor (GKLOF) algorithm was proposed. Effectiveness of GKLOF algorithm was verified by combination of theoretical analysis and simulation. 5000 users' real power data were selected to perform the simulation, and simulation results proved that the proposed method had high detection accuracy and stable decision threshold. In addition, local data distribution had minor impact on this method. Therefor it is more suitable for outlier detection in the case that power consumption behavior is complex and type of power consumption behavior is unknown. © 2018, Power System Technology Press. All right reserved.

引用

页码：1595 / 1604

页数：9

共 25 条

[1] Miao X., Zhang D., Sun D., The opportunity and challenge of big data's application in power distribution networks, Power System Technology, 39, 11, pp. 3122-3127, (2015)
[2] Liu K., Sheng W., Zhang D., Et al., Big data application requirements and scenario analysis in smart distribution network, Proceedings of the CSEE, 35, 2, pp. 287-293, (2015)
[3] Peng X., Zheng W., Lin L., Et al., A method to inspect the implementation of electricity price based on density clustering analysis and Fréchet discriminant analysis, Power System Technology, 39, 11, pp. 3195-3201, (2015)
[4] Zhao T., Zhang Y., Zhang D., Application technology of big data in smart distribution grid and its prospect analysis, Power System Technology, 38, 12, pp. 3305-3312, (2014)
[5] Han S.J., Cho S.B., Et al., Evolutionary neural networks for anomaly detection based on the behavior of a program, IEEE Transactions on Systems, Man, and Cybernetics, Part B, 36, 3, pp. 559-570, (2005)
[6] Nizara H., Dong Z.Y., Wang Y., Power utility nontechnical loss analysis with extreme learning machine model, IEEE Transactions on Power Systems, 23, 3, pp. 946-955, (2008)
[7] Nagi J., Tiong K.S., Ahmed S.K., Et al., Non-technical loss detection for metered customers in power utility using support vector machines, IEEE Transactions on Power Delivery, 25, 2, pp. 1162-1171, (2010)
[8] Nagi J., Yap K.S., Et al., Detection of abnormalities and electricity theft using genetic support vector machines, TENCON-IEEE Region 10 Conference Proceedings, pp. 1-6, (2008)
[9] Nagi J., Yap K.S., Tiong S.K., Et al., Improving SVM-based nontechnical loss detection in power utility using the fuzzy inference system, IEEE Transactions on Power Delivery, 26, 2, pp. 1284-1285, (2011)
[10] Leon C., Variability and trend-generalized rule induction model to NTL detection in power companies, IEEE Transactions on Power Systems, 26, 4, pp. 1798-1807, (2011)

← 1 2 3 →