Research And Application Of Key Technologies For Fault Repair Of Distribution Network

Distribution network fault repair is a key management link of power distribution system,which directly affects the power supply quality of power companies,and then affects the customer satisfaction evaluation of power companies,so its importance is becoming more and more prominent.Traditional distribution network repair mainly relies on emergency repair commanders to arrange repair resources according to experience and incomplete information of emergency repair tasks.Today,the scale of the distribution network is becoming more and more complex,and its limitations are gradually becoming more prominent.In order to improve the fault repair efficiency of distribution network effectively,reduce customer complaints and improve service quality,it is necessary to study the related technologies of fault repair of distribution network.This paper mainly studies the key technologies for fault repair of distribution network,including the prediction of the number of distribution network faults,the fault repair stagnation point location and the optimal allocation of repair tasks in distribution network,optimal configuration of fault repair resources for distribution network and quality evaluation system for fault repair of distribution network.Through research,a reasonable distribution network fault repair resource allocation plan,a reasonable fault repair stagnation point location and evaluation system of quality and efficiency of Fault repair in distribution network can be got.Firstly,a genetic algorithm based on grey relational analysis is proposed to optimize BP neural network distribution network fault quantity prediction mode.The influencing factors of distribution network faults are analyzed from time and geographical perspectives,and the gray correlation analysis method is used to select the influencing factors with large correlation as the input data to establish the BP neural network prediction model,and genetic algorithm is used to optimize the weight and threshold of BP neural network,which improves the prediction accuracy of the number of distribution network faults.Secondly,the optimization strategy of fault repair stagnation point location and repair task distribution in distribution network are studied.Considering the limited resources of each emergency repair station,a fault location and fault allocation model of distribution network emergency repair station based on two-dimensional spatial geographic coordinates is established.Aiming at the established non-linear programming model,the immune algorithm is used to obtain the global optimal solution or the approximate global optimal solution,and at the same time solve the two problems of repairing stagnation location and fault task assignment.Thirdly,research on optimal allocation of fault repair resources in distribution network is studied.The types and quantities of repair resources required for each type of fault are investigated from the three dimensions of time,space and type,a resource allocation modelbased on emergency repair workload is established.According to the predicted number of fault types and the established resource allocation model,a minimum absolute difference allocation algorithm based on rational resource allocation is proposed to guide the rational allocation of repair resources.Finally,the quality evaluation of distribution network fault repair is studied.The monitoring indicators which can accurately reflect the quality and efficiency of the fault repair of the distribution network are extracted,and the quality evaluation system for fault repair of the distribution network is established.The expert knowledge optimization decision algorithm is used to score the indicators,the analytic hierarchy process and entropy weight method are used to determine the weight of each index,the weighted summation is the comprehensive score of the fault repair,according to the comprehensive score,the quality and efficiency of fault repair work of each responsible unit are evaluated.