Research On Fault Recovery Reconstruction Of Distribution Network Based On GA-BFGS Algorithm

With the global depletion of non-renewable energy as well as the increasingly serious environmental damage and pollution caused by the exploitation and utilization of these energy,governments of all countries are increasingly mature in the research and utilization of new energy,utilising renewable energy including wind,tidal and solar energy to generate electricity.The emergence and large-scale application of these distributed power generation can significantly reduce people’s dependence of fossil energy and effectively protect environment,but with a large number of DG intergrated into the power system,the traditional power system is impacted.The access of DG changes the network topology of the distribution system,and affects the power flow distribution in the system,and if not contained,can seriously undermine the stability of the power system.In order to solve this problem,using optimization algorithm to reconstruct the distribution network is proposed,and has attracted the attention of many researchers.Distribution network reconfiguration can be divided into two kinds,optimal reconfiguration in normal operation and fault recovery reconfiguration in case of system failure.Distribution network reconfiguration and fault recovery reconfiguration can effectively reduce the active power loss and improve the power quality of the system,and can also improve the power supply reliability of the system.The main research of this paper is as follows:(1)The power flow distribution of the distribution system needs to be solved in the normal optimization reconstruction and fault recovery reconstruction.When a large number of DG are connected to the grid,the power flow distribution of the system will change significantly,which makes it more difficult to solve the power flow distribution.According to the different characteristics of different types of DG,this paper models them respectively,and divides them into four node types,namely PQ,PV,PI and PQ(V).It also analyzes several commonly used load models.Finally,taking ieee33 node distribution network system as an example,it solves several load models by using forward and backward method.The experimental results show that the ZIP load,which takes the characteristics of static voltage into account,has better calculation results in solving the power flow distribution of distribution network with DG.(2)After verifying the superiority of ZIP load model in solving power flow distribution,this paper proposes to use GA-BFGS algorithm to optimize and reconstruct the distribution network with DG under normal operation.This paper sets two convergence accuracy indexes primairily,intergrates genetic algorithm and quasi Newton method BFGS algorithm,combines the excellent global search ability of genetic algorithm with powerful local search ability and high convergence speed of BFGS algorithm,finally achieves the goal to speed up the algorithm and improve the global and local search ability of the algorithm.In the end,taking ieee33 node distribution system as an example,the paper uses constant power load model and ZIP load model,as well as a variety of algorithms to carry out a full range of comparative experiments,which proves the effectiveness of the algorithm in this paper.(3)In the process of fault recovery reconstruction of distribution system with DG,an improved Kruskal algorithm is proposed to divide the isolated islands of DG with independent power supply capability optimaly.This paper measures the weights in terms of the perceived importance of the load,and takes the electrical distance from the load to a certain DG as an index to determine whether to divide the load into an isolated island or not,finally find an optimal solution of island partition on meeting the division criteria and the constraint condition of islanded partition.At the same time,the GA-BFGS algorithm is used to reconstruct the remaining network outside the island,and the ieee33 node distribution network system is still used as an example to prove that the method is affective,which can significantly improve the reliability of power supply and power quality,and reduce the active power loss of the system.There are 15 figures,21 tables and 76 references in this paper.