Research On Uninterrupted Maintenance Of Distribution Network Based On Edge Computing

With the accelerating process of social development,people’s requirements for power quality are constantly improving.As the last link of power system transmission,it is very important to establish an economic and secure uninterrupted maintenance of power grid.Uninterrupted maintenance of power grid is to arrange the maintenance plan to a reason-able period of time,transfer the load affected by the maintenance as much as possible,and ensure the safety of the power grid.Due to the complex structure of the distribution network and the large amount of data,it brings pressure on data storage,cloud computing and transmission bandwidth.At present,the research of non outage maintenance based on edge computing is still in its infancy.In view of the above problems and needs,this thesis carries out the research of uninterrupted maintenance based on edge computing,and completes the following work.(1)This thesis conducts in-depth analysis and research on the domestic and foreign literature on uninterrupted maintenance of distribution network,summarizes the problems of uninterrupted maintenance,and analyzes the relevant ideas and algorithms to deal with the problem.(2)The model of uninterrupted maintenance of distribution network is established.Based on the analysis of maintenance status,the optimization problem of uninterrupted maintenance is decomposed into maintenance planning problem and load transfer prob-lem.(3)The cellular genetic algorithm is used to solve the maintenance scheduling prob-lem.In the distribution network maintenance plan preparation model,the impact of arrang-ing the maintenance plan on holidays or weekends is considered,and the economically optimal model of the electricity sales company is established.During the process of solv-ing the problem of preparing the maintenance plan of the distribution network,the cellular genetic algorithm is adopted to solve the problem.The performance of the algorithm is analyzed through calculation examples and compared with the results of the basic genetic algorithm.(4)The problem of load transfer in distribution network is addressed by using the tree cutting algorithm to be restored and the reliability evaluation of load transfer.In solv-ing the problem of load transfer,considering that the load of the equipment is constantly changing during the maintenance time,and there are errors in the prediction algorithm,which will bring the risk of overload failure of load transfer.This paper analyzes the reliability probability of the prediction algorithm,combined with the power flow calcu-lation method,puts forward the reliability evaluation method of the transfer scheme,and combines this method with the tree cutting algorithm to be restored transfer load.The optimization method is verified by an example.(5)The function module of power on maintenance of distribution network based on edge computing is designed.According to the model and algorithm of this paper,a vi-sual uninterrupted maintenance function module is developed,which integrates automatic generation of maintenance plan,transfer path and most of the data processing on the edge side.