| The distribution network is an important link connecting the power generation,transmission grid and the majority of customers.The quality of the power delivered by the distribution network will have a direct impact on the working status of the entire power system.Distribution network reconfiguration is a non-linear and multi-objective combination optimization problem that optimizes power flow in the network by adjusting the state of various switches under constrained conditions to achieve the purpose of optimizing power loss,voltage quality and line load in the network.At present,the research of distribution network reconfiguration algorithms is mainly focused on intelligent optimization algorithms,so it is of great practical significance to find a reasonable and effective intelligent optimization algorithm to deal with reconfiguration optimization problems.The differential evolution algorithm is a bionic intelligent optimization algorithm based on group evolution in solving continuous domains.It has the advantages of simple structure,few control parameters and strong robustness.Since its birth,its applications in the continuous field have performed well and have been gradually recognized in the discrete field.In view of this,this paper improves the differential evolution algorithm and proposes a differential evolution algorithm based on the basis vector scaling strategy and the mirror crossover strategy.And on this basis,a discrete differential evolution algorithm is designed to study the single-objective optimization problem of power loss in distribution network reconstruction and the multi-objective problem of power loss and voltage quality.The specific research content is arranged as follows:Firstly,in order to solve the problems that the differential evolution algorithm is easy to fall into the local optimal solution and poor convergence,the basis vector is scaled and the crossover operator is redesigned.A differential evolution algorithm based on the basis vector scaling and mirror intersection is proposed.The algorithm adds a base vector scaling factor to avoid local optimal values;during the crossover process,mirror crossover is introduced to increase the probability that the better test vector enters the selection step.In order to detect the characteristics of the differential evolution algorithm with basis vector scaling and mirror crossing,simulation optimization is implemented for 10 benchmark test functions.The results show that the proposed algorithm has the ability to jump out of the local minimum and obtain better convergence performance.Secondly,this paper studies the optimization of active power loss in distribution network reconfiguration,and constructs a single objective optimization mathematical model for the purpose of minimizing active power loss.Then,the "and" and "or" in binary operation rules are introduced to redefine the differential mutation operator,and a discrete differential evolution algorithm is proposed.Discretization differential evolution algorithm is used to solve the optimization problem of minimum network loss in distribution network reconstruction,and IEEE33 node distribution network is used as an example for optimization.The results show that the network loss is effectively reduced compared with that before reconstruction,and the simulation results are consistent with those of other algorithms within the allowable error range.Finally,research on network loss and voltage quality optimization problems in distribution network reconstruction and build a multi-objective mathematical model.The discrete differential evolution algorithm is used to optimize the multi-objective model,and the IEEE33 node distribution network is used as an example for verification.The results of algorithm optimization show that the improved algorithm has better ability to reduce network loss and improve voltage quality.Compared with the basic differential evolution algorithm,the optimized result is improved and the number of iterations is reduced. |
PDF Full Text Request