Research On Multi-Wind Farms Output Characteristics And Scenario Generation Method Considering The Spatial-Temporal Correlation

With the steady growth of wind power installed capacity in China,regional wind power clusters are increasing,and several large wind power bases are under construction.Centralized development,and long-distance delivery has become one of the main ways of wind power development in China.Generally,the development scale of wind power clusters is greater than 1 million kW.The fluctuation of wind power output will grow with the increase of installed capacity of wind power clusters.In addition,regional wind power clusters occupy a large area with complex terrain condition.Therefore,the large-scale grid connection of wind power clusters brings uncertainty risk to the operation safety and economic dispatch of power grids.This paper first studies the spatial-temporal correlation of regional wind power clusters.In terms of spatial correlation,the correlation coefficient is used to analyze the factors affecting the spatial correlation from the distance and time.For temporal correlation,the multi-fractal detrended fluctuation analysis method is employed to analyze long-term correlation and multi-fractal property.The method is capable of exploiting nonlinear characteristic and adaptive for non-stationary of wind power series.Aiming at the problem of multi-wind farms output scenario generation,a two-stage wind power scenario generation method considering spatial-temporal correlation is established.Through the spatial correlation modeling of multi-wind farms output,stochastic modeling of wind power fluctuation and the reconstruction process,the generated scenario can preserves the temporal and spatial correlation of the original scenario.In addition,an index system for evaluating the effectiveness of the generated scenarios is built.The multi-fractal detrended fluctuation analysis method is employed to calculate multi-dimensional index and describe the autocorrelation and dynamic fluctuation characteristics of wind power series.Finally,the power flow probability characteristics of power systems with multiple wind farms are analyzed.The IEEE57 node test system is taken as an example to verify that the spatial-temporal correlation of multi-wind farm output is an important feature to be considered in the random modeling of wind power cluster output.