Research On A Hybrid Algorithm For Reconfiguration Of Distribution Network

Distribution network is an important part of power system and an important part to ensure the quality of electricity for users.With the development of economy,distribution network automation becomes more and more important,and distribution network reconfiguration is an effective means of distribution network automation,which can improve the reliability and quality of power supply by changing the topology of distribution network,and finally reducing network loss which is of great significance to the improvement of social and economic benefits.Network reconfiguration is a multi-objective nonlinear combinatorial optimization problem.The current research focuses mainly on reconstruction optimization by using intelligent optimization algorithms.In this paper,a hybrid algorithm is proposed to reduce network loss,which can effectively optimize distribution network reconfiguration.Firstly,this paper analyzed the topology of power distribution network,and improved traditional method of forward and back generation power flow calculation.When analyzed the topolpgy of power distribution network,the size of the node "degree" is used to determine whether the network is connected and forms flap type.This judgment can remove the invalid solution in the iterative process of the subsequent algorithm and improve the computational efficiency of the algorithm.On the basis of the traditional method of forward and back generation power flow calculation,the tree type distribution network is layered,and the same layer nodes are used to parallel forward and back generation power flow calculation,which provides the basis for the intelligent optimization algorithm.Secondly,based on introduction of two intelligent optimization algorithms,a hybrid intelligent optimization is proposed.With the analysis of intelligent optimization algorithm introduced in this paper,it is found that particle swarm optimization algorithm can quickly converge to the optimal solution in the optimization process of distribution network reconfiguration,but when none of the population particles is near the best,the process may fall into the local optimal solution.The simulated annealing algorithm can escape from the local optimal solution and can find the global optimal solution in theory,but it is not a parallel algorithm,so sometimes the operation time is too long.Based on the properties of the two algorithms,a hybrid intelligent particle swarm optimization algorithm and simulated annealing algorithm is proposed.The basic idea of the algorithm is to calculate the fitness of each particle after the particle swarm optimization algorithm updates the velocity and position of the particle.The method of roulette is used to select the particle according to the probability,and then annealing the particle,the two algorithms alternately can achieve the purpose of optimization.Finally,this paper analyzed the application of hybrid intelligent algorithm in the reconfiguration of distribution network in theory and two examples are used to carry out the empirical research.when the hybrid algorithm is applied to the distribution network reconfiguration,on account of the basic particle swarm optimization algorithm fails to solve the discrete problem,the particle swarm optimization algorithm should be discretized for the distribution network reconfiguration.In order to enrich the diversity of operation mode,the open loop is initialized by random selection of switch.the simulation results of IEEE33 node and PG&E69 node system show that the algorithm can effectively and stably solve the optimization problem of distribution network reconfiguration.