Application Of Improved Particle Swarm Optimization In Fault Location Of Distribution Network In Yishui County

In the traditional power system theory,the whole process of power system is divided into five parts: generation,transformation,transmission,distribution and utilization.Therefore,the most important point of power distribution system is to find out the fault location of power supply system.However,most of the algorithms applied to the actual distribution network fault location have the defect of local convergence,and how to improve the poor local convergence and fault tolerance of the existing intelligent algorithms is the focus of research.Therefore,the main task of this paper is to improve the particle swarm optimization algorithm and apply it to the actual distribution network,and analyze its convergence and fault tolerance,and finally realize the fast and accurate fault location of distribution network.In order to simulate the fault location of distribution network with distributed generation,and explore its feasibility in the actual distribution network,this paper mainly does the following work:first,analyzes the current situation of Yishui County distribution network.It includes the distribution of distribution network lines in Yishui County,the construction of distribution automation system in Yishui County,and the current methods of fault location in Yishui County.After the analysis,the existing problems of fault location in Yishui County are listed.Secondly,the common algorithms of distribution network fault location are introduced.In this paper,six kinds of widely used algorithms are selected to introduce the algorithm process,which are divided into two categories: indirect and direct.The advantages and disadvantages of different algorithms are analyzed.Through comparison,the particle swarm optimization algorithm with excellent convergence and optimization performance is selected.Thirdly,the traditional binary particle swarm optimization algorithm is improved.The improvement includes: establishing compression factor to limit the value of self learning factor C1 and social learning factor C2,limiting the emergence of local convergence results from the root;introducing evolution factor,using evolution factor to determine whether the population is trapped in local convergence,analyzing the convergence results,avoiding the output result of local minimum convergence Optimal solution.Finally,the improved particle swarm optimization algorithm is applied to the actual distribution network.After introducing the coding method,switch function and fitness function of particle swarm optimization algorithm,this paper selects two classic lines in the urban distribution network of Yishui County:the traditional radial distribution network-Bashan line 2;the distribution network with distributed generation-Industrial Park Line 1.The improved particle swarm optimization algorithm and MATLAB are used to simulate the two actual lines.Compared with the simulation results of the improved algorithm,it is found that the improved PSO algorithm is more accurate and faster than the traditional PSO algorithm,and the improved PSO algorithm also has a certain fault tolerance for the encoding information uploaded by FTU.Therefore,the improved particle swarm optimization algorithm can basically meet the requirements of distribution network fault location in practical application,whether facing the complex distribution network with distributed generation or the more complex information after the distortion of coding information.