改进的模糊C均值法在负荷特性统计数据聚类中的应用
摘要
电力负荷是整个电力系统的安全稳定运行中较活跃的一部分。建立符合实际的动态负荷模型对电力系统规划、设计和运行等诸方面均有十分重要现实意义。本文采用实用化负荷建模思想 ,对负荷特性进行聚类,从而为变电站建立合适的负荷模型打下基础。基于目前负荷建模方面存在的问题,使用模糊C均值法,对同一地域不同地点变电站的负荷统计数据进行聚类分析。针对湖南电网48个变电站,对模糊C均值法实施改进后对其进行聚类,并与未改进的模糊C均值法的聚类结果进行比较,以说明改进方案的有效性。
关键字:电力负荷;负荷特性;聚类;模糊C均值法
APPLICATION OF IMPROVED FCM TO ELECTRIC LOAD CHARACTERISTICS OF STATISTICAL DATA CLUSTERING
ABSTRACT
The power load is an active part in the security and stable operation of the entire electrical power system. It is significantly important to make suitable load model for the power system planning, design and operation. In this paper the practical load modeling method is employed, and the load characteristics is clustered to establish the actual load model for substations. Based on the current problems, FCM with hierarchical clustering is used to perform the clustering of the load characteristics data of the different substations on the same area, the improved method is applied for the clustering of Hunan grid substation. The clustering result shows that the improved method is effective comparing with the unimproved method.
Key Words: power load; load characteristic; cluster,FCM
目 录
第一章 绪论........................................................1
1.1 研究背景........................................................1
1.2发展及研究现状..................................................2
1.2.1 发展.......................................................2
1.2.2 研究现状...................................................4
1.2.2.1 电力负荷建模的总体原则.................................4
1.2.2.2 电力负荷建模的基本概念.................................4
1.2.2.3 分类...................................................5
1.2.3 实用化负荷建模思想......................................6
1.2.3.1 统计综合法............................................6
1.2.3.2 总体测辨法............................................7
1.3 聚类分析在负荷特性分析中的应用现状.............................8
1.4 本文主要研究内容...............................................9
第二章 聚类分析.................................................10
2.1 聚类分析的基本概念...........................................10
2.2 聚类方法.....................................................11
2.3 系统聚类法...................................................15
2.3.1 最小张树聚类法............................................16
2.3.2 基于密度的聚类算法........................................16
2.3.3 基于网络的聚类方法........................................16
2.3.4 基于模型的聚类算法........................................16
2.3.5 基于划分的聚类算法........................................16
2.4 各算法优缺点比较...........................................