Research Of Fault Location For Distribution Network With Distributed Generation Based On Improved Gray Wolf Algorithm

With the extensive insert of distributed generation,the distribution network changes from single power mode to multi-power mode,and the bidirectional power flow appears in the system,which lets the exactness of the indigenous fault location method in the distribution network decline.In order to make better use of the distributed generation and let the operation of distribution network more safety and reliable,It is essential to study the fault location method considering the access characteristics of distributed generation.Therefore,based on the existing literature,this thesis focuses on the gray wolf algorithm,and proposes a fault section location method of distributed generation in distribution network based on the improved gray wolf algorithm.Firstly,this thesis introduces the feeder automation technology,expounds the classification of main fault location methods,and introduces in detail the advantages and disadvantages of matrix algorithm,overheated arc search method,linked list method,neural network,expert system and other intelligent algorithms in the application of fault location.Secondly,the influence of the access location of distributed generation on the operation mode and fault location of distribution network is discussed,the fault location method model considering the access characteristics of distributed generation is constructed,and the corresponding coding mode,switching function and objective function are established,which are solved by improving the grey wolf algorithm.Aiming at the shortcomings of gray wolf algorithm,such as insufficient global search ability and easy premature convergence,five improved strategies are proposed,including good point set initialization,dynamic range search,Levy flight,local search and nonlinear convergence factor,so as to improve the optimization performance of the algorithm in solution space.Through MATLAB simulation experiments and comparison with other algorithms,it can be seen that the fault location method based on the improved gray wolf algorithm proposed in this shesis has high fault location accuracy and fault tolerance.Finally,aiming at the problems of difficult fault location and long running time in complex distribution network,a hierarchical location method combining regional location and section location is proposed,which is solved by Improved Gray Wolf algorithm and implicit enumeration method.Aiming at the problem that the layered location method relies too much on the information uploaded by the regional end nodes,this thesis proposes a solution of fault section scanning,which further improves the fault tolerance of the layered location method.The simulation experiment is carried out in the large-scale complex distribution network with distributed generation,and compared with the improved gray wolf algorithm,gray wolf algorithm,ant colony algorithm and sparrow search algorithm.The results show that the hierarchical location method proposed in this thesis has fast operation speed,high location accuracy and high error tolerance ability,and is suitable for large-scale complex distribution network.