中文摘要：

为建立合适的变电站负荷模型,将聚类方法引入到变电站负荷特性分析,提出了一种基于蚁群优化K-medoids的综合聚类算法。该综合算法是K-medoids算法对蚁群的历史最优位置进行聚类分析,蚁群算法全局搜索能力强,克服了K-medoids算法易陷入局部最优的缺点,提高了聚类的准确率。最后通过变电站特性聚类实例,验证了综合算法在变电站特性聚类的可行性和有效性。

英文摘要：

To establish a proper substation load model, this paper applies clustering method into load characteristics analysis, and proposes a synthetic clustering algorithm of based on ant colony optimization K-medoids. The synthesis algorithm is makes clustering analysis for the history optimal position of ant colony using K-medoids algorithm, ant colony optimization algorithm has higher global search capability, overcoming the K-medoids algorithm easily trapped into local optimal shortcomings, improved clustering accuracy. Finally, it analyzes transformersubstation the characteristics clustering examples and results verify the feasibility and effectiveness of the proposed synthetic clustering method of substation characteristics.