基于改进k-means算法的海量智能用电数据分析.pdfVIP

38 卷 10 期 电 网 技 术 Vol. 38 No. 10 20 14 年10 月 Power System Technology Oct. 2014 文章编号：1000-3673 （20 14）10-2715-06 中图分类号：TM 734 文献标志码：A 学科代码：470 4054 基于改进k-means 算法的海量智能用电数据分析 赵莉，候兴哲，胡君，傅宏，孙洪亮 庆市电力公司， 庆市 渝中区 400015） Improved K-Means Algorithm Based Analysis on Massive Data of Intelligent Power Utilization ZHAO Li, HOU Xingzhe, HU Jun, FU Hong, SUN Hongliang (Chongqing Electric Power Company, Yuzhong District, Chongqing 400015, China) ABSTRACT: In allusion to such difficulties as huge amount of 将簇间距离及簇内对象的分散程度作为聚类数目选择的重 data and low mining efficiency that the data mining of intelligent 要参 ，对k-means 算法进行改进；为提高数据处理效率， power utilization has to be faced with, on the basis of 进行Map-Reduce 处理模型下的海量家庭用户用电数据的并 Map-Reduce parallel processing model the improved k-means 行挖掘。通过在Hadoop 集群上进行实验，结果证明提出的 algorithm based analysis on massive power utilization data is 算法运行稳定、高效、可行，且具有良好的加速比。 performed. Taking family user as the example, a data dimension 关键词：智能用电；云计算；Map-Reduce 处理模型；k-means model of family user’s electricity utilization information 算法；并行挖掘 including residential user’s number, floor area of the housing, number of family member, daily electricity consumption, peak- DOI 10.13335/j.1000-3673.pst.2014.10.015 and valley-electricity consumption, number of electrical home 0 引言 appliances and so on is established. Utilizing the advantage of k-means algorithm such as simple clustering analysis and rapid 近年来，面对智能用电领域强劲增长的电力需 convergence and overcoming its defect of easily falling into local 求和日渐紧张的电力供应，以及不断提