基于自适应分段云模型的电力异构数据聚类研究  

Study of Power Heterogeneous Data Clustering Based on Adaptive Segmented Cloud Model

作　　者：孙妍 张俊超 马占海 严嘉正 SUN Yan;ZHANG Junchao;MA Zhanhai;YAN Jiazheng(Information and Commumication Company,State Grid Qinghai Electric Power Company,Xining 810008,China)

机构地区：[1]国网青海省电力公司信息通信公司,青海西宁810008

出　　处：《自动化仪表》2024年第2期69-73,共5页Process Automation Instrumentation

摘　　要：云空间中的电力数据种类繁多、数据维度大、结构复杂,容易引发电力系统运行安全问题。有效的数据聚类可以消除云空间中电力异构的无关、冗余数据。引入自适应分段云模型,提出一种全新的电力异构数据聚类方法。通过Tent映射预处理电力异构数据,使数据空间分布更加均匀,提高了数据聚类精度。利用自适应正态云模型建立云期望曲线方程,获取正态云滴,以确定数据聚类中心。根据熵值定义分段聚合条件,实现异构数据有序聚类。测试结果表明,所提方法在聚类中心空间中的异构数据分布较为均匀,分类聚类面积较大,聚类收敛较快,轮廓系数为0.4。所提方法的电力异构数据聚类效果更优,具有较好的适用性和有效性。The electric power datas in cloud space are diverse,with large data dimensions and complex structures,which can easily lead to electric power system operation security problems.Effective data clustering can eliminate the irrelevant and redundant data of electric power heterogeneity in cloud space.Adaptive segmented cloud model is introduced to propose a new clustering method for power heterogeneous data.The power heterogeneous data are preprocessed by Tent mapping to make the data spatial distribution more uniform and improve the data clustering accuracy.The adaptive normal cloud model is used to establish the cloud expectation curve equation,obtain the normal cloud drops,and determine the data clustering center.According to the entropy value,the segmented aggregation conditions are defined to realize the orderly clustering of heterogeneous data.The test results show that the heterogeneous data distribution of the proposed method in the space of clustering center is more uniform,the classification clustering area is larger,the clustering converges faster,and the contour coefficient is O.4.The clustering effect of the researched method for electric power heterogeneous data is better,and it has better applicability and effectiveness.

关 键 词：云模型 电力异构数据 自适应分段 TENT映射 聚类系数 云分段聚合 正态云滴 有序聚类 

分 类 号：TH70[机械工程—仪器科学与技术]