Research On Reactive Power Optimization Of Distribution Network Based On Improved Bee Colony Algorithm

With the gradual expansion of the power grid,the power load is growing rapidly,the voltage level is getting higher and higher,and the economic operation of the power grid is receiving increasing attention.However,the lack of reactive power in the distribution network system and the unreasonable distribution will cause problems such as high line network loss and voltage fluctuation.Reactive power optimization is not only an effective means to ensure the safe and economic operation of the system,but also one of the effective measures to improve voltage stability.Therefore,how to rationally utilize and optimize reactive power resources,further reduce the loss of the distribution network,and improve the voltage qualification rate is of great significance.Firstly,according to the domestic and foreign research on reactive power optimization in distribution network,a mathematical model of reactive power optimization for distribution network is established,which takes the minimum active power loss in the distribution network system and the voltage constraint penalty function as the optimization objective.The advantages and disadvantages of Newton-Raphson and forward-backward sweep power flow calculation are analyzed.Aiming at these problems,the second-order cone programming algorithm of power flow calculation is introduced.Three power flow calculation algorithms are simulated and compared in IEEE-28 and IEEE-33 node systems of distribution network to verify the effectiveness and rapidity of this algorithm.The calculation ability of second-order cone programming power flow calculation in reactive power optimization of distribution network with distributed generation is verified by simulation examples,which provides a basis for reactive power optimization scheme in the following.Secondly,the bee colony algorithm is used as the main algorithm by using the characteristics of less adjustment parameters and high solution accuracy.Considering the slow convergence speed and poor robustness of the algorithm,it is proposed to introduce reverse learning and Gaussian mutation strategy in the initial population stage of the algorithm.The mutation strategy in the differential algorithm is introduced into the bee colony algorithm to perform neighborhood search to speed up the convergence speed and local search ability of the algorithm.Through six groups of common test function simulation experiments,the accuracy and operability of the proposed improved bee colony algorithm are verified by comparing the improved bee colony algorithm,bee colony algorithm and ant colony algorithm.Finally,the reactive power margin analysis method is used to determine the location of the compensation node in the distribution network system.Taking the IEEE-33 node system and the American PG & E69 node system as an example for simulation,the improved bee colony algorithm is compared with the original bee colony algorithm and ant colony algorithm to deal with the reactive power optimization problem of the distribution network.The simulation results show that the improved bee colony algorithm has faster convergence speed and better optimization results than the original bee colony algorithm,and the algorithm is more efficient in reducing the active network loss of the system.The voltage fluctuation amplitude is significantly reduced and the voltage is more stable.