Research On Optimization Configuration Of DG Distribution Network Based On Improving PSO

With the population base exceeding 1.4 billion in our country,the carbon emission of the power industry accounts for 2/5 of the total in our country.Influenced by factors such as ecological destruction brought by traditional fossil energy,clean distributed power has gradually developed and been added to the actual distribution network.But the access of distributed power has changed the flow direction and distribution of the original power distribution network.The system node voltage and network loss are affected.The amplitude of node voltage and network loss are closely related to the distributed power supply connected to the distribution network.Therefore,this paper studies the multi-objective optimization of distribution network with distributed power supply.Distributed generation technology and the classification of DG are introduced.The power generation principles and components of several common distributed power sources(wind power,photoelectric power,fuel cell and micro-gas turbine)are described.The specific node classification and corresponding node processing methods of DG grid-connected are analyzed.This paper classifies the power flow calculation method,introduces the advantages and disadvantages of the traditional forward and back substitution method and improves it.It expounds the influence caused by the access of DG to the distribution network,and mainly analyzes the influence of the access of DG to the distribution network on the network loss and voltage.A simple circuit model is established,and the analysis shows that the access location and capacity of DG are the main factors affecting the loss voltage of distribution network.In PSASP,the IEEE-17 node distribution network was taken as the model,and the single control variable method was used for simulation analysis,to obtain the change trend of the distribution network loss and node voltage after the access location or capacity of different DG.A multi-objective optimal configuration model is established with the minimum node voltage offset and network loss as the multi-objective function.Moreover,the annealing factor of simulated annealing method is introduced into the traditional PSO algorithm to avoid the traditional algorithm falling into the local optimal solution,and the traditional inertia weight is set as the adaptive weight and the learning factor is set as the synchronous learning factor,so as to obtain an improved multi-objective PSO algorithm.MATLAB simulation program was run to verify algorithm optimization with IEEE-33 node as the model.By comparing the node voltage per unit value under single objective and multi-objective,it is obvious that the node voltage offset after multi-objective optimization is smaller and the average voltage per unit value is higher.By comparing the node branch loss and the total system loss obtained by the traditional and improved PSO algorithm,it is found that the improved algorithm can greatly reduce the network loss.In the later stage,the iterative graph of the traditional PSO algorithm and the improved PSO algorithm is compared and verified.The improved algorithm has faster convergence speed,higher convergence accuracy and stronger optimization effect.