摘要
针对传统多源异构时序数据集成方法执行和抽取时间较长、筛选准确率较低的问题,设计基于时空聚类算法的多源异构时序数据集成方法。协同处理多源异构时序数据,根据协同处理结果与时空集合标定结果构建时空聚类矩阵,以时空聚类矩阵为基础计算多源异构时序数据集成需求,通过时空聚类算法转换数据,以此进行多源异构时序数据集成。实验结果表明,该文方法的多源异构时序数据执行和抽取时间在20 s以内,筛选准确率高于95%,集成效果更好。
Aiming at the problems of long execution and extraction time and low screening accuracy of traditional multi⁃source heterogeneous temporal data integration methods,a multi⁃source heterogeneous temporal data integration method based on spatio⁃temporal clustering algorithm is designed.Collaborative processing of multi⁃source heterogeneous time series data,building a spatiotemporal clustering matrix according to the collaborative processing results and spatiotemporal set calibration results,calculating the integration requirements of multi⁃source heterogeneous time series data based on the spatiotemporal clustering matrix,and converting data through spatiotemporal clustering algorithm to integrate multi⁃source heterogeneous time series data.The experimental results show that the execution and extraction time of multi⁃source heterogeneous temporal data in this method is less than 20 s,the screening accuracy is higher than 95%,and the integration effect is better.
作者
陈超
胡才亮
崔钰
谢芳
杨慧芳
王健
CHEN Chao;HU Cailiang;CUI Yu;XIE Fang;YANG Huifang;WANG Jian(AnHui Mingsheng Hengzhuo Technology Co.,Ltd.,Hefei 230000,China)
出处
《电子设计工程》
2023年第20期168-171,176,共5页
Electronic Design Engineering
基金
国家电网安徽电力有限公司科技项目(521207220002)。
关键词
时空聚类
多源异构
异构时序
数据集成
集成方法
spatiotemporal clustering
multi⁃source heterogeneity
heterogeneous time series
data int⁃egration
integration method
