智能电网海量信息处理中的关联规则与数据流挖掘研究  

Research on Association Rules and Data Flow Mining in Massive Information Processing of Smart Grid

作　　者：阮诗迪 刘欣然 张雄宝 谢虎 RUAN Shidi;LIU Xinran;ZHANG Xiongbao;XIE Hu(Power Dispatch and Control Center of Guangxi Power Grid Co.,Ltd.,Nanning 530013,China;Digital Grid Research Institute,China Southern Power Grid,Guangzhou 510000,China)

机构地区：[1]广西电网电力调度控制中心,广西南宁530013 [2]南方电网数字电网研究院有限公司,广东广州510000

出　　处：《微型电脑应用》2023年第7期61-64,共4页Microcomputer Applications

基　　金：南方电网公司重点科技项目资助(GXKJXM20190619)。

摘　　要：通过对智能电网中的数据挖掘,用关联规则实现对未来电力负荷的预测。分析现有的各类数据挖掘方法和关联规则;在PFP-tree(Parallel Feature Pyramid-tree)算法和FUP(Fast UPdate)算法的基础上,提出可以实现关联规则增量实时更新的SPUFP(Spark Parallel Updated Frequent Pattern)算法。该算法的优点是可以有效地减小空间和时间的复杂度,并对计算分组策略和模式树结构进行优化。以SPUFP算法为核心建立预测模型,并和回归方程的预测模型进行数据对比分析。This paper mainly studies the prediction of power load through association rules by data mining in smart grid.Various existing data mining methods and association rules are analyzed.On the basis of two algorithm of PFP-Tree(Parallel Feature pyramid-Tree)algorithm and FUP(Fast UP date)algorithm,the Spark Parallel Updated Frequent Pattern(SPUFP)algorithm is proposed to implement incremental real-time updating of association rules.The SPUFP algorithm can effectively reduce the complexity of space and time,and optimize the computing grouping strategy and pattern tree structure.The SPUFP algorithm is used as the core to establish a prediction model,and the data are compared with the prediction model of regression equation.

关 键 词：数据挖掘 关联规则 电力预测 稳定性 准确性 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]