Research On Distributed Power Configuration Optimization Based On Improved Three-point Power Flow Calculation

In recent years,with the extensive use of fossil fuels in power plants,the problems caused by centralized power supply mechanism dominated by conventional grid have gradually attracted the attention of countries all over the world.As a new way of power generation,distributed generation plays an important role in dealing with energy shortage and environmental pollution.However,the grid connection of distributed power supply and electric vehicles also poses new challenges to the reliability of power system.The influence of distributed power supply on power system mainly depends on its location and installed capacity.Uncertainty exists between distributed power and electric vehicles.This paper aims to make distributed power generation run more intelligently,safely and reliably.In this paper,an improved three-point estimation method based on Halton sequence sampling is proposed to estimate probabilistic power flow.This method adds a pair of estimation points on the basis of the first three orders of input variables to complete the calculation,so as to improve the accuracy of the moment estimation of output variables.It avoids the calculation of higher order central moments and the non-real solutions that may appear in the normalized distance center.At the same time,the correlation between variables is also considered in the calculation.The performance of the improved three-point estimation method is verified in IEEE 14 node system and IEEE 33 node system.The accuracy of I3 PEM was verified by Monte Carlo simulation results under different coefficient of variation(CV).A multi-objective optimization function model is established,which includes annual investment cost,power purchase cost and active power loss cost.To solve this model,based on the parameter adaptive differential evolution algorithm based on the successful history,this paper proposes a new mutation strategy,a fitting-induction parental selection scheme for binomial crossing.In the modified recombination framework,by making each mutant only perform binomial crossing with the highest ranking individual in the current population,In order to improve the convergence speed of the algorithm.In addition,when the algorithm is applied in practice,the uncertainty of distributed power supply is taken into account,and the relevant models of wind power,photovoltaic,load and electric vehicle are solved by the improved three-point estimation method.Finally,the probabilistic power flow model is globally optimized by the improved recorded parameter adaptive differential evolution algorithm to obtain the final optimal configuration scheme.Finally,the improved three-point estimation method is verified in IEEE 14-node system and IEEE 33-node system.According to standard numerical datum of CEC 2005 and CEC 2010,the improved adaptive differential evolution algorithm is compared with other algorithms,and its convergence performance is verified.Using IEEE 33 bus as the platform,the optimal algorithm configuration scheme of distributed power supply in the optimization process is solved,and the convergence characteristics and voltage distribution characteristics of the optimized nodes are verified.Simulation results show that the algorithm has high convergence accuracy,short computation time and high efficiency,which provides a new solution for distributed power supply optimization configuration.