摘要
现有的公用变压器状态信息聚类方法存在聚类有效率与平均分类准确率低的问题,故提出基于5G通信的公用变压器状态信息聚类方法。针对公用变压器状态信息的复杂化和多样化,通过计算采集状态信息数据之间的相关支持度、置信度和权重,判断公用变压器是否存在运行故障。通过聚类算法加快5G通信速度,应用聚类流程确定公用变压器的状态信息。实验结果表明,基于5G通信的公用变压器状态信息聚类方法能够提高聚类有效率,增加平均分类准确率。
The existing state information clustering methods of public transformers have the problems of low clustering efficiency and average classification accuracy.In view of this,a clustering method of state information of public transformers based on 5G communication is proposed.In view of the complexity and diversification of the state information of the public transformer,the relevant support degree,confidence degree and weight between the collected state information data are calculated to determine whether the public transformer has operation fault.The 5G communication speed is accelerated by clustering algorithm,and the state information of public transformer is determined by clustering process.The experimental results show that the clustering method based on 5G communication can improve the clustering efficiency and increase the average classification accuracy.
作者
廖绍成
LIAO Shaocheng(State Grid Zhejiang Xinchang Power Supply Company,Shaoxing 312500,China)
出处
《电子设计工程》
2021年第19期155-158,163,共5页
Electronic Design Engineering
基金
浙江省科技厅研究计划项目(LGG21E060003)。
关键词
5G通信
公用变压器
变压器状态
信息聚类
5G communication
public transformer
transformer state
information clustering
