基于改进鸟群算法优化聚类的风电场等值建模

EQUIVALENT MODELING OF WIND FARM BASED ON IMPROVED BIRD SWARM OPTIMIZATION CLUSTERING

为了提高风电场动态等值建模的精确度,采用风力发电机组的12个状态变量作为分群指标,使用改进鸟群算法(Improve Bird Swarm Algorithm,IBSA)搜索最佳聚类中心,通过K-means算法对风电场进行分群处理。在MATLAB/Simulink中搭建详细模型与等值模型,并与传统聚类算法进行对比。实验结果表明,该方法对等值建模的准确度有很大的提高,可以精确地表征风电场的对外特性。

In order to improve the accuracy of dynamic Equivalent modeling of wind farm,12 state variables of wind turbines were used as clustering indicators,and the improved bird swarm algorithm(IBSA)was used to search the best clustering center.Then,the K-means algorithm was used to cluster the wind farms.Finally,we built detailed model and equivalent model in MATLAB/Simulink,and compared them with traditional clustering algorithms.The experimental results show that our method improves the accuracy of equivalent modeling greatly,and the external characteristics of wind farms can be accurately represented.