运用数据挖掘中的聚类技术对电力系统日负荷曲线进行分析,提出一种基于特性指标降维的日负荷曲线聚类方法——特性指标聚类(pattern index clustering,PIC),通过负荷率、日峰谷差率等6个日负荷特性指标对日负荷曲线进行降维处理,利用基...运用数据挖掘中的聚类技术对电力系统日负荷曲线进行分析,提出一种基于特性指标降维的日负荷曲线聚类方法——特性指标聚类(pattern index clustering,PIC),通过负荷率、日峰谷差率等6个日负荷特性指标对日负荷曲线进行降维处理,利用基于聚类有效性修正的德尔菲方法配置各指标权重,以加权欧式距离作为相似性判据,对日负荷曲线进行聚类。算例结果表明所提方法运行时间短,鲁棒性好,提高了负荷曲线聚类质量,能直观反映典型负荷曲线的特点。展开更多
The summarization and evaluation of the advances in fuzzy clustering theory are made in the aspects including the criterion functions, algorithm implementations, validity measurements and applications. Several importa...The summarization and evaluation of the advances in fuzzy clustering theory are made in the aspects including the criterion functions, algorithm implementations, validity measurements and applications. Several important directions for a further study and the application prospects are also pointed out.展开更多
文摘运用数据挖掘中的聚类技术对电力系统日负荷曲线进行分析,提出一种基于特性指标降维的日负荷曲线聚类方法——特性指标聚类(pattern index clustering,PIC),通过负荷率、日峰谷差率等6个日负荷特性指标对日负荷曲线进行降维处理,利用基于聚类有效性修正的德尔菲方法配置各指标权重,以加权欧式距离作为相似性判据,对日负荷曲线进行聚类。算例结果表明所提方法运行时间短,鲁棒性好,提高了负荷曲线聚类质量,能直观反映典型负荷曲线的特点。
文摘The summarization and evaluation of the advances in fuzzy clustering theory are made in the aspects including the criterion functions, algorithm implementations, validity measurements and applications. Several important directions for a further study and the application prospects are also pointed out.