基于多层架构的台区智慧物联体系架构设计  

The design of the metering production management platform with distributed multi-layer architecture

作　　者：于凯 崔伟 刘雅男 康伟 YU Kai;CUI Wei;LIU Ya-nan;KANG Wei(State Grid Shandong Electric Power Company Jinan Zhangqiu District Power Supply Company,Jinan 250200,China)

机构地区：[1]国网山东省电力公司济南市章丘区供电公司,济南250200

出　　处：《信息技术》2023年第8期182-188,共7页Information Technology

摘　　要：针对常规技术中的计量生产管理平台不完善、数据处理滞后等问题,提出了新型的计量生产管理平台。构建出包括数据层、访问层、计算层和应用层的分布式多层架构,还设计了包括计量客户服务层、计量业务处理层、计量工作质量监控层和计量调度层的计量生产调度数据管理平台,实现不同计量业务的管理,满足用户的多种需要。文中研究融入了大数据处理算法,应用KNN算法实现不同类型数据的分类,通过构建AdaBoost算法模型,实现电能计量故障数据的诊断。试验表明,文中研究的方法数据处理能力较强,工作效率较高,误差较低。To tackle the problems of imperfect measurement production management platform and lagging data processing in conventional technology,a new measurement production management platform is proposed.A distributed multi-layer architecture including data layer,access layer,computing layer and application layer is constructed.It also designs a metering production scheduling data management,including metering customer service layer,metering business processing layer,metering work quality monitoring layer and metering scheduling layer,which realizes the management of different metering services and meets the various needs of users.The research in this paper incorporates big data processing algorithms,applies KNN algorithm to realize the classification of different types of data,and builds the AdaBoost algorithm model to realize the diagnosis of electric energy metering fault data.The experiment shows that the method studied in the article has strong data processing ability,high work efficiency and low error.

关 键 词：计量生产管理平台 分布式多层架构 KNN算法 ADABOOST算法 计量业务 

分 类 号：TM72[电气工程—电力系统及自动化] TN931[电子电信—信号与信息处理]