基于聚类算法的风电场动态等值  

Dynamic Equivalence for Wind Farms Based on Clustering Algorithm

作　　者：陈树勇[1] 王聪[2] 申洪[1] 高宁超[3] 朱琳[3] 兰华[2] 

机构地区：[1]中国电力科学研究院,北京市海淀区100192 [2]东北电力大学,吉林省吉林市132012 [3]华北电力大学,北京市昌平区102206

出　　处：《中国电机工程学报》2012年第4期11-19,24,共9页Proceedings of the CSEE

摘　　要：风电机组在实际运行时,受尾流效应等因素影响,运行状态并不相同。为提高风电场实际运行模型的精度,提出了一种适用于双馈式风力发电机的动态等值建模方法。它将风电机组的状态变量矩阵作为分群指标,利用聚类算法将矩阵中的数据进行分群,将同群的风电机组等值成为一台风力发电机,实现了风电场的动态等值。利用PSD/BPA平台,对系统侧故障与风速变化2种情况仿真,并与传统等值方法及风电场详细模型对比。仿真结果表明,采用仿真过程中的状态变量作为分群指标是合理的,该模型与详细模型的动态特性基本一致,可以用来描述风电场的实际运行状态。The running states of wind generators are not identical due to the influence of some factors（such as wake effect）.In order to improve the precision of the wind farm model for the actual operation,a dynamic equivalence modeling method that is suitable for doubly-fed wind generators was proposed.The state variables of matrix of the wind generators were adopted as cluster-dependent index.By using clustering algorithms,the data of matrix were divided into groups and the wind turbine generators in the same group were equivalent as a wind generator,therefore,the dynamic equivalence of the wind farm was realized.The faults at power system side and the change of wind velocity were simulated on PSD/BPA power system analysis platform and the results were compared to that of the traditional method.The simulation results show that it is reasonable that the state variables of simulation process were considered as cluster-dependent index.The dynamic characteristic of the new model is the same to that of the detailed model and it can be applied to describe the actual states of a wind farm.

关 键 词：双馈式风力发电机 聚类算法 状态变量 动态等值 电力系统分析 

分 类 号：TM74[电气工程—电力系统及自动化]