Power Stealing Identification And Loss Assessment Method Based On Analysis Of Marketing And Distribution Data

In recent years,the phenomenon of stealing electricity is frequent,which brings serious challenges to the security of power grid and economic losses to the country.The investigation and punishment of stealing electricity only rely on manpower,which has many problems,such as heavy workload,lack of pertinence,difficulty in obtaining evidence,and difficulty in power loss recovery calculation.The integration of power marketing and distribution breaks the business boundary and information island phenomenon of both sides of power distribution and consumption management departments.The data analysis and application under the integration provide conditions for the identification and verification of power theft.Based on the integrated data,this paper digs the attribute characteristics behind the power stealing,forms the identification criteria of power stealing,and explores an effective method to distinguish power stealing.The main work and innovation of this paper are as follows.Based on the algorithms of distance and density,the outlier detection method has well recognition effect on the detection for two kinds of electricity theft,loss of voltage method and loss of current method.Aiming at the problem of false identification in power factor,Pearson coefficient and grey correlation analysis are also used to analyze the variation degree of load characteristic curve after the occurrence of power stealing.For the low voltage users,a monitoring model of line loss and abnormal power consumption is then established to realize the identification of power stealing relying on the integrated power consumption and line loss management system from two dimensions:the power stealing occurs in the monitoring period and before the monitoring period.For the users who steal electricity in small quantity,the recognition model is established based on the time sequence analysis of characteristic events of intelligent power meter.The means mentioned above still have some problems,such as the inability to do anything about the power stealing without meter method,and the difficulty in the loss evaluation of the power stealing with total loss of voltage method.According to the changes of electrical parameters of the network before and after the occurrence of the power stealing event,a method of identifying the electric larceny without meter is deduced theoretically,proposing the two criterions of branch loss and node voltage,which lays the foundation for the subsequent calculation of power recovery.On the basis of the theoretical analysis,a branch stealing model is established based on IEEE 33 node and the actual distribution network.The simulation experiment is carried out on Matlab platform,and the variation law of each branch loss and intermediate node voltage is analyzed.The experimental results verify the correctness and effectiveness of the methods proposed in this paper.Lastly,based on the realization of power stealing identification,this paper puts forward the evaluation methods of power stealing loss,which are correction coefficient,dichotomy iteration of stealing load and data fusion means of consumption and distribution,etc.Combined with relevant cases and simulation models,the effectiveness of the loss assessment methods is verified.