基于IBBO和K-means++融合的多属性台区聚类研究

Research on Multi-attribute Station Clustering Based on Fusion of IBBO and K-means++

为更全面、精细地研究配网侧台区特性,提出一种基于改进的生物地理学优化算法(IBBO)和K-means++融合的多属性台区聚类方法。首先,采用低方差滤波方法筛选台区电气特征参数;其次,引入IBBO,求得最优解作为K-means++的初始聚类中心;最后,对海量台区用电数据进行聚类分析,得到不同类别的台区特性,完成多属性台区的聚类。算例以某市台区真实数据进行仿真分析。结果表明,所提方法相比经典聚类算法能更加快速且准确地实现台区的有效分类,为分析复杂台区的用电行为提供了支撑。

In order to study the characteristics of the side station area of the distribution network more comprehensively and precisely,a multi-attribute station area clustering method based on the fusion of improved biogeographic optimization algorithm(IBBO)and K-means++was proposed.The whole idea was as follows:Firstly,use the low variance filtering(LVF)method to filter the electrical characteristic parameters of the station area;secondly,introduce IBBO to obtain the optimal solution as the initial clustering center of K-means++;finally,perform clustering analysis on the massive power consumption data of the station area to obtain different types of station area characteristics,and complete the clustering of multi-attribute station areas.The calculation example adopted the real data of a certain city to perform a simulation analysis.The results show that the proposed method can realize the effective classification of the station areas more quickly and accurately than the classic clustering algorithm,which provides support for the analysis of the electricity consumption behavior of the complex station area.