Network Partitioning And Reactive Power Optimization Based On Scenario Reduction Technology In Wind Power System

With the continuous expansion of the scale of power system and the access of intermittent new energy such as wind power etc.,the uncertainties and complexity of power grid are increasing,and the reactive power optimization and voltage control of power system are becoming more and more complex.Therefore,it is of great significance to study reactive power voltage zoning and reactive power optimization in wind power system.Aiming at the problem that the voltage regulation strategy is not considered in the existing methods of reactive power and voltage partitioning,a reactive power margin index is proposed in this paper.The margin index takes into account the effect of the total reactive power reserve and the maximum voltage deviation of the PQ node in the region after the regional voltage regulation.An optimization model and algorithm for calculating the index are also proposed.Based on the reactive power margin index and calculation method,a two stage reactive power voltage zoning method is present.This method uses Latin hypercube sampling to sample wind speed and load to take account of the uncertainty of wind power and load.In the first stage of reactive power and voltage partitioning,reactive power flow tracing method is used to determine the supply-demand relationship between reactive power supply and load,and load nodes are merged into the reactive power supply area which provides the maximum reactive power.In the second stage of reactive power and voltage partitioning,based on the obtained pre-partitioning scheme,the proposed reactive power margin is taken as the objective function,and the combined scheme of reactive power and voltage pre-partitioning is solved by genetic algorithm.The proposed partitioning method is applied to IEEE 39-bus system and IEEE 118-bus system to verify the feasibility and effectiveness of the method.In the existing reactive power optimization methods for power systems with wind power,scenario reduction is mainly performed before optimization for multi-scenarios reflecting wind power and load uncertainty.A scenario clustering method based on solution similarity for reactive power optimization with wind power is proposed to reduce the computation of scenarios in this paper.Based on the partition scheme of reactive power and voltage,the eactive power betweenness of regional tie line and regional reactive power balance are introduced into the existing reactive power optimization model as objective functions.In the process of solving the model by genetic algorithm,the wind power output scenarios are analyzed by chromosomes of each generation of genetic algorithm.K-means clustering algorithm is used to cluster the objective functions of different wind power scenarios according to the distribution differences of the objective functions.In the process of genetic evolution,scene clustering and release are carried out.Simulation calculation is carried out on the IEEE 39 node system.The results show that the scene clustering method based on solution similarity can save computing time and deal with the contradiction between computational efficiency and fitting accuracy to a certain extent.