Research On Operation Mode And Risk Assessment Issues Of Distribution Network Based On Big Data Analysis

With the continuous development of social economy and people's living standard,people pay more attention to power supply reliability,security,power quality and other aspects of the distribution network.How to make accurate evaluation of the operation mode of the distribution network becomes one of the most popular issues of power system.On the other hand,the power outages happened in recent years made people worry about the risk of distribution network.With the continuous improvement of power system risk theory,the risk assessment of distribution network has become a growing concern of experts and scholars.In addition,a variety of big data analysis methods provide possibility for the correlation analysis between the distribution network operation mode and risk indexes in today's big data environment.This paper firstly researches on the traditional analysis method of distribution network operation mode index,and proposes a distribution network operation mode index system which is suitable for urban distribution network.The operation mode indexes of distribution network are analyzed by least squares method,and operation mode indexes are divided into several levels according to the calculated sensitivity and deviation,which can provide reference for the optimization of operation mode indexes.Secondly,this paper researches on the traditional weight-determining method of the distribution network operation mode indexes,and proposes a weight-determining method of the indexes based on the index sensitivity and the gray relationality.This method takes the difference of each index's influence on the distribution network operation mode level into account.In this method,the operation mode of the distribution network is pre-evaluated using principal component analysis method,and the sensitivity weights of the indexes are obtained by further analysis according to the pre-evaluation result.Besides,the gray relational weights of the indexes are calculated using gray relational analysis method.Finally,the comprehensive weights of the indexes are calculated using the variable weight formula,which makes the evaluation result more reasonable.In order to solve the problem of the application of the traditional power system risk assessment methods in distribution network,this paper concentrates on the existing risk assessment methods of distribution network,and proposes a risk assessment method considering the real-time operation risk and the grid structure risk.The risk assessment index system is established from two aspects of real-time operation risk and grid structure risk.Besides,the risk assessment result is quantized and the risk level is divided into grades,which can make comprehensive risk assessment of the distribution network risk.What's more,a correlation analysis method based on Apriori algorithm is proposed to analyze the relationship between the operation mode and the risk level of the distribution network.The relationship between the operation mode indexes and the risk indexes of the distribution network is analyzed by means of big data analysis methods,and the association rules are mined,which can provide reference for the improvement of the risk level of distribution network.Finally,practical data of an urban distribution network is analyzed,which proves the effectiveness and practicality of the proposed index sensitivity analysis method,comprehensive evaluation method,risk assessment method and the index correlation analysis method.The analysis results can provide reference for the planning,transformation and dispatching of distribution network.