Distribution Network With Doubly-fed Wind Turbines Reactive Power Optimization Research

Wind energy has a central position in the development of today’s energy industry,with high commercial value and development potential.Its outstanding advantages include low pollution,low cost and high flexibility.Wind power generation is an important means to improve the stable operation of the power grid.When the wind turbine is connected to the power grid,it will cause certain changes in the grid voltage and power.The reactive power optimization of the power distribution system makes the power quality improve,and the active power loss is also reduced.Moreover,the voltage of each node is adjusted,so that the node voltage fluctuation range is small and tends to be stable.After the grid is merged into the doubly-fed induction generator,the purpose of reducing the system’s network loss,improving the voltage quality and supplying the reactive power to compensate is to provide an important research for reducing the cost and enhanced reliability of power transmission and distribution in the power industry in the future.basis.The research in this paper is carried out according to the determination of model and objective function,Latin hypercube sampling,scene reduction,and algorithm optimization.The main working steps and simulation results are as follows:(1)Through the equivalent circuit and power relationship of the doubly-fed induction generator,the ability of the doubly-fed induction generator to supply reactive power is derived,which can provide the basis for both the converter capacity and the wind farm in reactive voltage control.The reactive power control strategy of the wind farm.In calculating the power flow of distribution network with wind turbines,this paper introduces the power flow calculation of distribution network in detail,adopts the mathematical model of classical forward-return method,and proposes corresponding model for doubly-fed induction generator model of PQ node in distribution network.Solution.(2)The purpose of selecting Latin hypercube sampling in this paper is to avoid the problem of random fluctuation of wind speed;at the same time,the error under the same sampling times will be reduced.The number of scenes obtained by Latin Hypercube sampling is very large.If you want to speed up the work schedule,save computing time and improve efficiency within a certain period of time,you must reduce the number of scenes.This paper selects the scene reduction technology to simulate the effective samples.The precision retention is maximized.Substituting the samples obtained after the reduction into the trend,and selecting representative scenes for research,that is,turning the uncertainty trend optimization problem into a single scene deterministic trend optimization problem.(3)The ACO-PSO hybrid algorithm avoids the shortcomings of the early accumulation of information by the ACO algorithm,the slow process of finding the optimal solution,and the tendency of the PSO algorithm to fall into local optimum,and combines the advantages of the PSO algorithm and the ACO algorithm.,including less parameter setting,single frame and global search function,the analysis shows that the algorithm has certain practicability.Basedon the simulation of the IEEE-33 node in Matlab,the simulation results show that the three algorithms in the three scenarios are compared in the network loss and voltage.