With the continuous development of the economy,the demand for the reliability and quality of power supply is increasing.However,due to the lag of the traditional distribution network management mode,it can’t meet the increasing demand for electricity,which leads to many defects in asset management,operation and maintenance.The household transformation relationship in distribution network is the basis of line loss calculation,fault location and combination of operation and distribution,and the parameters of distribution line are the important support of power flow calculation,fault analysis and loss calculation.It is helpful to carry out the work of lean management and active operation and maintenance of distribution network system in electric power enterprises to obtain accurate parameters.Under the background of the popularization and application of intelligent electric meter,it is the mainstream of distribution network development to use measurement data to do relevant research.In this paper,based on the research of the identification of the household variation relation and the line parameter identification,the data of the low voltage distribution station is mined deeply,analyzed and applied synthetically,and the household variation relation and the line parameter in the low voltage distribution station is identified by the way of data analysis,which is of great significance to the improvement of the information and automation level of the distribution network and realization of the situation awareness and active operation and maintenance of the distribution network.The specific research contents are divided into the following three parts:Firstly,it introduces the main wiring ways and characteristics of low-voltage distribution station area,and expounds the structure and functions of the power consumption information collection system as well as the meter reading ways of low-voltage meter users in detail.This paper analyzes the structure of collecting device in distribution station area and the process of information interaction between terminal collecting device and main station system in low-voltage station area during the construction of smart grid.Then,the pretreatment method of measuring data is studied,including outlier identification and missing data filling,to provide data basis for the follow-up research.Secondly,an improved SVDD algorithm based on the low-voltage station area household variable relationship identification method is proposed.The original voltage data of all users in the station area to be analyzed are obtained by using the electric information collection system,and the original voltage data are extracted by KPCA,the SVDD multi-class model is constructed by using the user voltage data set which retains the main features of the daily voltage curve as the input,second recognition using KNN algorithm.The results show that the proposed method has a high recognition rate and can be obtained by comparing with other methods.The method proposed in this paper has a higher ability to identify the cross-platform misfiled users in the actual distribution network,and can provide a reference for the actual investigation on the spot.Finally,on the basis of the identification of low-voltage station area,a method of on-line identification of low-voltage station area line parameters based on power balance is proposed.Combining the topology of low-voltage distribution station with the multi-time measurement data of nodes in the station,the parameter identification equation of power balance is established,and the parameter equation is solved by weighted least squares,the optimal solution of line parameters is obtained,and the parameter identification of low-voltage distribution line is realized.An example is given to verify the method.The result shows that the method proposed in this paper has good recognition accuracy,it can provide effective reference for electric power enterprises to carry out active operation and maintenance in intelligent station area and on-line monitoring of transmission lines,and has certain engineering value. |