摘要
为进一步提高数据挖掘算法的处理速度和计算精确度,提出一种基于电力信息数据聚类分析的数据挖掘算法设计。该算法依据聚类分析原理,采用基于密度的聚类方法和相异度矩阵对数据和数据类型进行筛选和相异度计算,并基于聚类分析框架设计数据挖掘算法流程。在数据挖掘算法基本策略下对输入的数据采用SLIO算法处理离散字段,输出需要的数据结果。仿真结果表明,相比其他配网自动化系统数据挖掘算法,所设计算法在数据挖掘速度和准确度上均体现出较好的优势,具有良好的可信度。
In order to further improve the processing speed and calculation accuracy of data mining algorithm,a data mining algorithm design based on power information data clustering analysis was proposed.Based on the principle of clustering analysis,the algorithm used density-based clustering method and dissimilarity matrix to filter and calculate the dissimilarity of data and data types,and designed the data mining algorithm process based on the framework of clustering analysis.Under the basic strategy of data mining algorithm,the input data was processed by SLIO algorithm to deal with discrete fields,and the required data results were output.The simulation results showed that compared with other data mining algorithms of distribution network automation system,the designed algorithm had better advantages in data mining speed and accuracy,and had good credibility.
作者
陈子健
CHEN Zijian(Foshan Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Foshan 528000,Guangdong Chian)
出处
《粘接》
CAS
2024年第1期189-192,共4页
Adhesion
基金
广东电网有限责任公司佛山供电局资金资助项目(项目编号:030600KK52190214)。
关键词
聚类分析法
相异度矩阵
数据挖掘
算法设计
cluster analysis
dissimilarity matrix
data mining
algorithm design
