基于改进Canopy-FCM的光伏电站动态特性聚类算法  

A CLUSTERING ANALYSIS ALGORITHM OF DYNAMIC CHARACTERISTICS OF PHOTOVOLTAIC POWER PLANT BASED ON IMPROVED CANOPY FUZZY C-MEANS

作　　者：郭易鑫 韩霞[1] 倪光捷 贺鹏远 孙惠娟[4] Guo Yixin;Han Xia;Ni Guangjie;He Pengyuan;Sun Huijuan(State Grid Shanxi Electric Power Company,Taiyuan 030021,Shanxi,China;State Grid Yili Technology Co.,Ltd.,Fuzhou 350003,Fujian,China;Fujian Wangneng Technology Development Co.,Ltd.,Fuzhou 350003,Fujian,China;East China Jiaotong University,Nanchang 330000,Jiangxi,China)

机构地区：[1]国网山西省电力公司,山西太原030021 [2]国网信通亿力科技有限责任公司,福建福州350003 [3]福建网能科技开发有限责任公司,福建福州350003 [4]华东交通大学,江西南昌330000

出　　处：《计算机应用与软件》2022年第4期287-293,331,共8页Computer Applications and Software

基　　金：国家自然科学基金项目(51867008)。

摘　　要：为了准确分析并网光伏电站的动态特性,提出一种基于Canopy-模糊C-means(fuzzy C-means, FCM)聚类算法的等效建模方法。在传统聚类指标选取方面,通过详细的数学模型,推导并网逆变器控制系统的传递函数,并以闭环传递函数的零极点表达式为基础,提出一种新的聚类指标。为改进FCM聚类对初始聚类中心和外部聚类中心的敏感性,采用Canopy算法确定初始聚类中心和聚类个数,提高聚类算法的准确性和效率。算例研究验证了所提方法的有效性。In order to accurately analyze the dynamic characteristics of grid connected photovoltaic power station, an equivalent modeling method based on Canopy-FCM(fuzzy C-means) clustering algorithm is proposed. In the aspect of traditional clustering index selection, the transfer function of grid connected inverter control system was derived through detailed mathematical model, and a new clustering index was proposed based on the expression of zero pole of closed-loop transfer function. In order to improve the sensitivity of FCM clustering to the initial cluster center and the external cluster center, Canopy algorithm was used to determine the initial cluster center and the number of clusters, which improved the accuracy and efficiency of the clustering algorithm. The effectiveness of the proposed method was verified by an example.

关 键 词：并网光伏电站 改进FCM 聚类 Canopy算法 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术] TM6151[电气工程—电力系统及自动化]