Reactive Power Optimization Based On Dynamic Cloud Evolution Particle Swarm Optimization Algorithm In Distribution Network With Wind Farm

As a green energy in the new century, wind power has broadprospects in the future. However, because of the uncertainty of the wind,a large scale of fans interconnecting the grid will affect the power quality,even tide disorders and grid collapses. To this end, for reactive powerplanning of power grid with wind farm, reasonably arranging the reactivepower has important significance in the safe and stable operation of thegrid.This thesis firstly introduces the working principle of the fan, therelationship between the fan output power and the size of wind speed,then analyzes the effect of reactive power current which shows theimportance of reactive power compensation. After that, use P-Q iterationmodel to simulate the wind power system flow calculation. Finally, makethe active network loss minimum as the optimization goal, establish themathematical model of reactive power compensation in powerdistribution network with wind farm.According to the optimization problem, the thesis adopts Particle Swarm Optimization (PSO) algorithm. Because of the uncertainty of thefans’ output resulting in the worse environment of particle optimization,the thesis designs PSO based on changing the flight strategy method.Change the conventional linear adjustment strategy of PSO totwo-dimensional search strategy, make flying particles double affected bythe weight coefficient and flight step length, it can balance theconventional global and local search ability of PSO algorithm, so betteroptimal solution can be obtained. This method has good effect onparticles of lower fitness. Finally, the simulation proves the effectivenessof the proposed algorithm.On this basis, to cope with the complex and changeable windenvironment, improve the efficiency of seeking the optimal value, makebetter use of higher fitness particles, this thesis designs a new PSOalgorithm based on dynamic cloud evolution. Select the excellentindividuals to evolve, so as to reduce the scale of pessimum particles,which enhances the search speed, improves the distribution of theparticles. On this basis, the conventional particles dynamically changespeed according to PSO based on changing the flight strategy method.Finally, the simulation proves the effectiveness of the proposed algorithm.At last, based on the proposed dynamic cloud evolutionary PSOestablish reactive power optimization software of wind power system.Taking consideration of the grid voltage, generator, the software makes the minimal active network loss as the goal which can optimize theparameters and intelligent data storage to improve the operationefficiency in the wind environment. The system software is based on theprinciple of modular design, thus increasing the openness and simplicityof software. The realization of the software not only greatly improves thevoltage level of wind environment, reduces the network loss, but alsoscientifically makes use of reactive power equipment to avoid onlinereactive power flowing on the road which reflects the principlehierarchical partial pressure, balance on the spots. Therefore, it has agood practicability and broad application prospects.